You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2022/10/26 23:15:48 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@04afd8311386b6293bda04843738e972df8e93f1)
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 381ebe81c0 deploying docs (apache/tvm@04afd8311386b6293bda04843738e972df8e93f1)
381ebe81c0 is described below
commit 381ebe81c0f5474c6ebdbde8339a448160d63c56
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Wed Oct 26 23:15:40 2022 +0000
deploying docs (apache/tvm@04afd8311386b6293bda04843738e972df8e93f1)
---
docs/_images/sphx_glr_micro_train_001.png | Bin 332672 -> 331892 bytes
docs/_images/sphx_glr_micro_train_thumb.png | Bin 24174 -> 23825 bytes
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_keras.rst.txt | 2 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.rst.txt | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 2 +-
.../compile_models/sg_execution_times.rst.txt | 22 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 20 +-
.../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 | 1779 ++++++++++----------
.../tune_network_cuda.rst.txt | 4 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 112 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 6 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 687 +++++++-
.../work_with_microtvm/micro_autotune.rst.txt | 16 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 18 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 10 +-
.../work_with_relay/sg_execution_times.rst.txt | 8 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../how_to/work_with_schedules/tensorize.rst.txt | 2 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 6 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 6 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 56 +-
.../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 | 47 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_darknet.html | 2 +-
docs/how_to/compile_models/from_keras.html | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 14 +-
docs/how_to/compile_models/from_pytorch.html | 12 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 26 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 43 +-
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 | 34 +-
docs/how_to/deploy_models/sg_execution_times.html | 20 +-
.../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 | 1779 ++++++++++----------
.../tune_with_autoscheduler/tune_network_cuda.html | 4 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 112 +-
.../tune_with_autotvm/sg_execution_times.html | 6 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 687 +++++++-
docs/how_to/work_with_microtvm/micro_autotune.html | 16 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 10 +-
.../how_to/work_with_relay/sg_execution_times.html | 8 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
docs/how_to/work_with_schedules/tensorize.html | 2 +-
docs/install/nnpack.html | 12 +-
docs/reference/api/python/auto_scheduler.html | 4 +-
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
.../api/typedoc/classes/graphexecutor.html | 16 +-
docs/reference/api/typedoc/classes/instance.html | 40 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 112 +-
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 6 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 6 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 6 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 262 ++-
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 26 +-
docs/tutorial/tensor_expr_get_started.html | 43 +-
126 files changed, 4120 insertions(+), 2647 deletions(-)
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 3f59a37dab..9ba43750df 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 6746d44e45..d025dae545 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 b8ff6f4150..278d0d9e4d 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -315,7 +315,7 @@ The process is no different from other examples.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 12.650 seconds)
+ **Total running time of the script:** ( 1 minutes 16.366 seconds)
.. _sphx_glr_download_how_to_compile_models_from_darknet.py:
diff --git a/docs/_sources/how_to/compile_models/from_keras.rst.txt b/docs/_sources/how_to/compile_models/from_keras.rst.txt
index 19a359e160..30629a6d70 100644
--- a/docs/_sources/how_to/compile_models/from_keras.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_keras.rst.txt
@@ -228,7 +228,7 @@ Look up prediction top 1 index in 1000 class synset.
.. code-block:: none
Relay top-1 id: 285, class name: Egyptian cat
-
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 993ms/step
+
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 1s/step
Keras top-1 id: 285, class name: Egyptian cat
diff --git a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
index ddb91a38a5..281d8d53f2 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -115,7 +115,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipd86b083d-67ea-4b7e-8991-31cdae3e67c8 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip80b28f56-b89e-45fb-b875-95558f226c3f 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 915679b563..945f507553 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -116,7 +116,7 @@ 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]
15%|#5 | 6.33M/41.5M [00:00<00:00, 62.2MB/s]
30%|##9 | 12.3M/41.5M [00:00<00:00, 55.5MB/s]
42%|####2 | 17.6M/41.5M [00:00<00:00, 33.3MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 33.3MB/s]
78%|#######8 | 32.5M/41.5M [00:00<00:00, 46.2MB/s]
92%|#########2| 38.3M/41.5M [00:00<00:00, 45.5MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 43.1MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
15%|#5 | 6.33M/41.5M [00:00<00:00, 58.8MB/s]
29%|##8 | 11.9M/41.5M [00:00<00:00, 48.5MB/s]
40%|#### | 16.6M/41.5M [00:00<00:00, 38.7MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 41.2MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 49.6MB/s]
96%|#########6| 40.0M/41.5M [00:00<00:00, 57.3MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 52.5MB/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 3d6ab70a8b..38f476201c 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -98,7 +98,7 @@ 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]
18%|#7 | 7.99M/44.7M [00:00<00:00, 59.7MB/s]
36%|###5 | 16.0M/44.7M [00:00<00:00, 69.4MB/s]
60%|###### | 27.0M/44.7M [00:00<00:00, 83.0MB/s]
85%|########4 | 37.7M/44.7M [00:00<00:00, 93.6MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 68.0MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
18%|#7 | 7.99M/44.7M [00:00<00:00, 60.7MB/s]
36%|###5 | 16.0M/44.7M [00:00<00:00, 65.8MB/s]
54%|#####3 | 24.0M/44.7M [00:00<00:00, 72.6MB/s]
69%|######9 | 31.0M/44.7M [00:00<00:00, 55.4MB/s]
82%|########2 | 36.8M/44.7M [00:00<00:00, 55.2MB/s]
95%|#########4| 42.3M/44.7M [00:00<00:00, 41.7MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 51.8MB/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 83f341fe18..1546eaec73 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -416,7 +416,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 10.056 seconds)
+ **Total running time of the script:** ( 1 minutes 17.756 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 9935a75da9..c6b7acb560 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
Computation times
=================
-**05:43.811** total execution time for **how_to_compile_models** files:
+**06:05.361** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:12.650 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:17.756 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:10.056 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:16.366 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:45.349 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:47.854 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:31.904 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:33.497 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:30.106 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:30.768 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:26.072 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:27.544 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:25.326 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:26.256 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:22.431 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:23.571 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:17.510 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:19.288 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.406 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.462 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index 303d3120b7..a0b94a64c3 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
@@ -434,7 +434,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.1361 16.0553 16.8488 15.7097 0.4122
+ 16.3309 16.3534 16.4332 16.1870 0.0770
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 fd29aa7cb6..1d0e753b50 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
@@ -127,7 +127,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MaskRCNN_ResNet50_FPN_Weights.COCO_V1`. You can also use `weights=MaskRCNN_ResNet50_FPN_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 7.99M/170M [00:00<00:02, 69.2MB/s]
13%|#2 | 21.9M/170M [00:00<00:01, 111MB/s]
19%|#9 | 32.8M/170M [00:00<00:01, 82.6MB/s]
25%|##5 | 42.6M/170M [00:00<00:01, 88.9MB/s]
33%|###2 | 56.0M/170M [00:00<00:01, 96.1MB/s]
42%|####2 | 72.0M/170M [00:00<00:01, 98.5MB/s]
48%|####8 | 81.6M/170M [00:00<00:00, 98.1MB/s]
54%|#####3 | 91.1M/170M [00:01<00:00, 94.5MB/s]
59%|#####8 | 100M/170M [00:01<00:00, 78.3MB/s]
64%|######3 | 108M/170M [00:01<00:00, 77.4MB/s]
68%|######8 | 116M/170M [00:01<00:00, 73.4MB/s]
72%|#######2 | 123M/170M [00:01<00:00, 68.5MB/s]
76%|#######6 | 129M/170M [00:01<00:00, 68.0MB/s]
80%|######## | 136M/170M [00:01<00:00, 63.5MB/s]
86%|########6 | 147M/170M [00:01<00:00, 74.4MB/s]
91%|######### | 154M/170M [00:02<00:00, 73.2MB/s]
95%|#########4| 161M/170M [00:02<00:00, 58.1MB/s]
99%|#########8| 168M/170M [00:02<00:00, 57.9MB/s]
100%|##########| 170M/170M [00:02<00:00, 75.8MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 7.99M/170M [00:00<00:02, 68.4MB/s]
9%|9 | 16.0M/170M [00:00<00:02, 63.3MB/s]
14%|#4 | 24.0M/170M [00:00<00:02, 56.1MB/s]
19%|#8 | 32.0M/170M [00:00<00:02, 54.3MB/s]
24%|##3 | 40.0M/170M [00:00<00:02, 51.9MB/s]
28%|##8 | 48.0M/170M [00:00<00:02, 54.6MB/s]
33%|###2 | 56.0M/170M [00:01<00:02, 56.9MB/s]
38%|###7 | 64.0M/170M [00:01<00:01, 56.1MB/s]
42%|####2 | 72.0M/170M [00:01<00:01, 59.3MB/s]
47%|####7 | 80.0M/170M [00:01<00:01, 60.6MB/s]
51%|##### | 86.4M/170M [00:01<00:01, 62.2MB/s]
54%|#####4 | 92.4M/170M [00:01<00:01, 60.0MB/s]
60%|###### | 102M/170M [00:01<00:01, 59.1MB/s]
64%|######3 | 108M/170M [00:01<00:01, 57.5MB/s]
67%|######6 | 113M/170M [00:02<00:01, 48.3MB/s]
71%|####### | 120M/170M [00:02<00:01, 49.7MB/s]
75%|#######5 | 128M/170M [00:02<00:00, 48.7MB/s]
80%|######## | 136M/170M [00:02<00:00, 46.9MB/s]
85%|########4 | 144M/170M [00:02<00:00, 48.7MB/s]
89%|########9 | 152M/170M [00:02<00:00, 48.6MB/s]
94%|#########4| 160M/170M [00:03<00:00, 45.5MB/s]
100%|##########| 170M/170M [00:03<00:00, 54.8MB/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').
@@ -296,7 +296,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 13.899 seconds)
+ **Total running time of the script:** ( 3 minutes 34.758 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 88a5618d2f..151ae5742d 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -236,7 +236,7 @@ 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]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 58.4MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 80.7MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
53%|#####2 | 7.16M/13.6M [00:00<00:00, 75.1MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 74.8MB/s]
@@ -418,7 +418,7 @@ 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)
- 90.5762 90.5239 91.5192 90.2164 0.2400
+ 90.6310 90.5840 91.9719 90.4586 0.1746
@@ -467,7 +467,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 20.607 seconds)
+ **Total running time of the script:** ( 1 minutes 25.775 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 d19311cedd..790b366dcc 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
@@ -432,7 +432,7 @@ 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)
- 120.2854 120.3054 122.0403 119.5772 0.3622
+ 122.8141 122.6929 130.2332 121.9411 0.8347
@@ -469,7 +469,7 @@ 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 25.230 seconds)
+ **Total running time of the script:** ( 2 minutes 25.010 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 a2c9f76db1..cca1eb9107 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -253,7 +253,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 32.362 seconds)
+ **Total running time of the script:** ( 1 minutes 37.902 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 c313d2b9c9..d7e2452b34 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
@@ -166,7 +166,7 @@ Convert and compile model for CPU.
data: None
input_sym_arg_type = in_param.infer_type()[0]
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
0%| | 0/132723 [00:00<?, ?KB/s]
5%|5 | 6841/132723 [00:00<00:01, 68400.12KB/s]
12%|#1 | 15300/132723 [00:00<00:01, 77918.57KB/s]
17%|#7 | 23092/132723 [00:00<00:01, 57563.24KB/s]
24%|##4 | 31879/132723 [00:00<00:01, 67416.62KB/s]
29%|##9 | 39124/132723 [00:00<00:01, 66338.25KB/s]
36%|###6 | 48010/132723 [00:00<00:01, 73167.37KB/s]
43%|####2 | 56842/132723 [00:00<00:00, 77743.30KB/s]
49%|####9 | 65664/132723 [00:00<00:00, 80901.50KB/s]
56%|#####6 | 74410/132723 [00:00<00:00, 82874.04KB/s]
63%|######2 | 83260/132723 [00:01<00:00, 84562.90KB/s]
69%|######9 | 92078/132723 [00:01<00:00, 85646.40KB/s]
76%|#######6 | 100906/132723 [00:01<00:00, 86432.47KB/s]
83%|########2 | 109752/132723 [00:01<00:00, 87038.70KB/s]
89%|########9 | 118490/132723 [00:01<00:00, 74534.69KB/s]
96%|#########5| 127164/132723 [00:01<00:00, 77804.19KB/s]
100%|#######
###| 132723/132723 [00:01<00:00, 77540.07KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
6%|5 | 7479/132723 [00:00<00:01, 74781.33KB/s]
12%|#2 | 16359/132723 [00:00<00:01, 83021.32KB/s]
19%|#9 | 25248/132723 [00:00<00:01, 85698.06KB/s]
26%|##5 | 34082/132723 [00:00<00:01, 86739.32KB/s]
32%|###2 | 42897/132723 [00:00<00:01, 87245.03KB/s]
39%|###8 | 51705/132723 [00:00<00:00, 87523.54KB/s]
46%|####5 | 60659/132723 [00:00<00:00, 88180.13KB/s]
52%|#####2 | 69612/132723 [00:00<00:00, 88608.31KB/s]
59%|#####9 | 78476/132723 [00:00<00:00, 88613.62KB/s]
66%|######5 | 87355/132723 [00:01<00:00, 88665.42KB/s]
73%|#######2 | 96361/132723 [00:01<00:00, 89090.47KB/s]
79%|#######9 | 105316/132723 [00:01<00:00, 89226.92KB/s]
86%|########6 | 114282/132723 [00:01<00:00, 89355.68KB/s]
93%|#########2| 123218/132723 [00:01<00:00, 89293.80KB/s]
100%|#########9| 132210/132723 [00:01<00:00, 89481.00KB/s]
100%|#######
###| 132723/132723 [00:01<00:00, 87982.81KB/s]
@@ -242,7 +242,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 56.028 seconds)
+ **Total running time of the script:** ( 3 minutes 8.283 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 231f885288..8906f6ebd3 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
@@ -5,24 +5,24 @@
Computation times
=================
-**12:52.353** total execution time for **how_to_deploy_models** files:
+**13:40.029** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:13.899 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:34.758 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:56.028 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 03:08.283 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:25.230 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:25.010 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:32.362 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:37.902 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:20.607 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:25.775 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:35.284 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:37.427 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:24.776 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:25.776 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:24.160 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:25.091 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.006 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.007 | 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 8842d8817b..b3e5507627 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
@@ -472,7 +472,7 @@ 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.zip247ccd88-0f24-48dd-b94c-c3937117ab31 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip02aff155-db9d-442b-b1a1-afafd93f7f13 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 c7aa72c33b..6e520dc952 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
=================
-**00:45.424** total execution time for **how_to_extend_tvm** files:
+**00:49.319** 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``) | 00:42.079 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:45.664 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.342 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.546 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:00.994 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.100 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.008 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.009 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index a1bcaa8ddf..da12f3b993 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -216,10 +216,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 6528us [6528us] (45.73%; 45.73%)
- FoldScaleAxis: 7748us [6us] (54.27%; 54.27%)
- FoldConstant: 7743us [1633us] (54.23%; 99.93%)
- InferType: 6110us [6110us] (42.80%; 78.91%)
+ InferType: 7339us [7339us] (47.39%; 47.39%)
+ FoldScaleAxis: 8147us [6us] (52.61%; 52.61%)
+ FoldConstant: 8141us [1635us] (52.57%; 99.92%)
+ InferType: 6506us [6506us] (42.01%; 79.91%)
@@ -258,10 +258,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 6192us [6192us] (44.37%; 44.37%)
- FoldScaleAxis: 7764us [5us] (55.63%; 55.63%)
- FoldConstant: 7760us [1569us] (55.60%; 99.94%)
- InferType: 6191us [6191us] (44.36%; 79.78%)
+ InferType: 6766us [6766us] (44.59%; 44.59%)
+ FoldScaleAxis: 8408us [7us] (55.41%; 55.41%)
+ FoldConstant: 8401us [1697us] (55.37%; 99.92%)
+ InferType: 6705us [6705us] (44.18%; 79.80%)
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 4e40593409..a1d6a71599 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -340,7 +340,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 48.037086 ms
+ Convolution: 36.324352 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 393981f793..baa2768115 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
@@ -659,7 +659,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 10.191680 ms
+ conv2d with tensor core: 13.360851 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 9cfdc8824e..63b346a822 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -143,8 +143,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.017844
- Baseline: 3.381535
+ Numpy running time: 0.019474
+ Baseline: 3.304775
@@ -239,7 +239,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.302799
+ Opt1: 0.330295
@@ -342,7 +342,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.334024
+ Opt2: 0.344460
@@ -438,7 +438,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.114958
+ Opt3: 0.140625
@@ -563,7 +563,7 @@ flattening.
.. code-block:: none
- Opt4: 0.110128
+ Opt4: 0.111854
@@ -685,7 +685,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.111559
+ Opt5: 0.114512
@@ -810,7 +810,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.146860
+ Opt6: 0.150728
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 06a6435c38..a18f0ec5b7 100644
--- a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
Computation times
=================
-**00:34.522** total execution time for **how_to_optimize_operators** files:
+**00:35.841** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.130 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:33.120 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.336 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.533 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.056 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.188 | 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 fbe95a64b3..e2ac46160e 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
=================
-**08:15.195** total execution time for **how_to_tune_with_autoscheduler** files:
+**08:44.052** 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``) | 04:54.702 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 05:08.510 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:29.296 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:34.419 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:01.447 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:04.102 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:27.538 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:33.297 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:11.501 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:12.230 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:10.710 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:11.494 | 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 116297ddb3..c75b0f1668 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -240,453 +240,461 @@ cooperative fetching, unrolling and operator fusion.
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 32;
- allocate(conv2d_nchw: Pointer(local float32), float32, [4]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [2016]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [1536]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [1], [], scope="local", align=4)[0] = 0f32
- conv2d_nchw_1[1] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [16]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [392]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [256]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [4], [], scope="local", align=8)[0] = 0f32
conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[6] = 0f32
+ conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[14] = 0f32
+ conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[3] = 0f32
- for (rc.outer.outer: int32, 0, 16) {
- for (rx.outer.outer: int32, 0, 3) {
- let cse_var_2: int32 = (rc.outer.outer*1568)
- let cse_var_1: int32 = (rc.outer.outer*288)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [2016], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[((((cse_var_2 + (floordiv(threadIdx.x_1, 63)*49)) + rx.outer.outer) + floormod(threadIdx.x_1, 63)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 588), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 4), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 4), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 784), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 4), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 980)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 980), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1176)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1176), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1372)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1372), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1568)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 8), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 8), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1568), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 8), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1764)] = @tir.if_then_else(((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[((((cse_var_2 + (floordiv(threadIdx.x_1, 63)*49)) + rx.outer.outer) + floormod(threadIdx.x_1, 63)) + 1364)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- if @tir.likely((threadIdx.x_1 < 56), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 1960)] = @tir.if_then_else((((threadIdx.x_1 < 49) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1960), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1: Buffer(kernel.shared, float32, [1536], [], scope="shared")[threadIdx.x_2] = kernel[(((((blockIdx.x*73728) + (floordiv(threadIdx.x_2, 96)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 96)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 196), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 4), 96), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 392), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 96), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 588)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 588), 96)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 4), 32)*9)) + (floormod(threadIdx.x_2, 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 784), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 96), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 980)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 980), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 96), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 1176), 96)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 32)*9)) + (floormod(threadIdx.x_2, 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- if @tir.likely((threadIdx.x_2 < 164), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 1372)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 1372), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 28), 96), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + rx.outer.outer)]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 384)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 768)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1152)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 385)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 769)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1153)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 386)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 770)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1154)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 387)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 771)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1155)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 388)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 772)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1156)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 389)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 773)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1157)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 390)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 774)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1158)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 391)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 775)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1159)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 392)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 776)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1160)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 393)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 777)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1161)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 394)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 778)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1162)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 395)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 779)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1163)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 396)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 780)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1164)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 397)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 781)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1165)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 398)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 782)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1166)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 399)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 783)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1167)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 400)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 784)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1168)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 401)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 785)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1169)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 402)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 786)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1170)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 403)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 787)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1171)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 404)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 788)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1172)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 405)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 789)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1173)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 406)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 790)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1174)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 407)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 791)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1175)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 408)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 792)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1176)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 409)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 793)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1177)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 410)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 794)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1178)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 411)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 795)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1179)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 412)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 796)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1180)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 413)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 797)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1181)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 414)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 798)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1182)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 415)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 799)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1183)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 416)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 800)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1184)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 417)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 801)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1185)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 418)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 802)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1186)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 419)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 803)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1187)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 420)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 804)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1188)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 421)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 805)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1189)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 422)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 806)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1190)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 423)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 807)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1191)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 424)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 808)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1192)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 425)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 809)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1193)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 426)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 810)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1194)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 427)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 811)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1195)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 428)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 812)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1196)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 429)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 813)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1197)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 430)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 814)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1198)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 431)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 815)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1199)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 432)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 816)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1200)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 433)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 817)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1201)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 434)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 818)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1202)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 435)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 819)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1203)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 436)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 820)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1204)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 437)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 821)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1205)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 438)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 822)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1206)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 439)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 823)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1207)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 440)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 824)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1208)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 441)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 825)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1209)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 442)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 826)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1210)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 443)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 827)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1211)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 444)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 828)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1212)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 445)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 829)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1213)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 446)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 830)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1214)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 447)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 831)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1215)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 448)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 832)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1216)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 449)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 833)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1217)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 450)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 834)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1218)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 451)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 835)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1219)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 452)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 836)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1220)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 453)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 837)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1221)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 454)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 838)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1222)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 455)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 839)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1223)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 456)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 840)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1224)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 457)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 841)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1225)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 458)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 842)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1226)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 459)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 843)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1227)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 460)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 844)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1228)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 461)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 845)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1229)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 462)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 846)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1230)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 463)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 847)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1231)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 464)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 848)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1232)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 465)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 849)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1233)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 466)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 850)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1234)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 467)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 851)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1235)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 468)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 852)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1236)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 469)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 853)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1237)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 470)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 854)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1238)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 471)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 855)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1239)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 472)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 856)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1240)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 473)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 857)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1241)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 474)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 858)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1242)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 475)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 859)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1243)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 476)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 860)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1244)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 477)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 861)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1245)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 478)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 862)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1246)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 479)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 863)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1247)]))
+ conv2d_nchw_1[5] = 0f32
+ conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[13] = 0f32
+ conv2d_nchw_1[15] = 0f32
+ for (rc.outer.outer: int32, 0, 64) {
+ for (ry.outer.outer: int32, 0, 3) {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [392], [], scope="shared")[(threadIdx.x_1*2)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1*2), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 1)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*2) + 1), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 7)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[((threadIdx.x_1*2) + 196)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1*2), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 188)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 197)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*2) + 1), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 189)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
+ kernel.shared_1: Buffer(kernel.shared, float32, [256], [], scope="shared")[ramp((threadIdx.x_2*4), 1, 4)] = kernel[ramp((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 2)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 2)*36)) + (ry.outer.outer*3)), 9, 4)]
+ }
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*16)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 32)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 64)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 96)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 128)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 160)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 192)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 224)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 1)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 33)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 65)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 97)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 129)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 161)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 193)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 225)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 8)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 40)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 72)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 104)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 136)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 200)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 232)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 9)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 41)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 73)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 105)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 137)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 201)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 233)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 2)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 34)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 66)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 98)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 130)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 162)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 194)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 226)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 3)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 35)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 67)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 99)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 131)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 163)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 195)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 227)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 10)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 42)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 74)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 106)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 138)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 202)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 234)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 11)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 43)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 75)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 107)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 139)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 203)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 235)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 4)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 36)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 68)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 100)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 132)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 164)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 196)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 228)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 5)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 37)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 69)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 101)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 133)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 165)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 197)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 229)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 12)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 44)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 76)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 108)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 140)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 204)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 236)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 13)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 45)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 77)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 109)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 141)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 205)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 237)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 6)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 38)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 70)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 102)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 134)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 166)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 198)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 230)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 7)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 39)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 71)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 103)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 135)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 167)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 199)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 231)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 14)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 46)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 78)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 110)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 142)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 206)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 238)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 15)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 47)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 79)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 111)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 143)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 207)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 239)]))
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[(threadIdx.x_1*2)] = @tir.if_then_else(((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 7)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 1)] = @tir.if_then_else(((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 6)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[((threadIdx.x_1*2) + 196)] = @tir.if_then_else(((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 189)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 197)] = @tir.if_then_else(((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 190)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
+ kernel.shared_1[ramp((threadIdx.x_2*4), 1, 4)] = kernel[ramp(((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 2)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 2)*36)) + (ry.outer.outer*3)) + 1), 9, 4)]
+ }
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*16)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 32)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 64)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 96)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 128)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 160)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 192)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 224)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 1)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 33)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 65)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 97)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 129)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 161)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 193)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 225)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 8)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 40)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 72)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 104)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 136)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 200)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 232)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 9)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 41)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 73)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 105)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 137)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 201)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 233)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 2)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 34)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 66)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 98)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 130)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 162)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 194)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 226)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 3)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 35)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 67)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 99)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 131)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 163)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 195)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 227)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 10)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 42)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 74)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 106)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 138)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 202)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 234)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 11)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 43)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 75)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 107)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 139)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 203)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 235)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 4)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 36)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 68)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 100)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 132)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 164)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 196)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 228)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 5)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 37)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 69)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 101)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 133)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 165)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 197)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 229)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 12)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 44)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 76)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 108)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 140)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 204)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 236)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 13)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 45)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 77)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 109)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 141)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 205)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 237)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 6)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 38)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 70)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 102)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 134)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 166)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 198)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 230)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 7)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 39)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 71)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 103)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 135)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 167)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 199)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 231)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 14)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 46)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 78)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 110)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 142)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 206)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 238)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 15)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 47)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 79)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 111)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 143)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 207)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 239)]))
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[(threadIdx.x_1*2)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (floormod((threadIdx.x_1*2), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 6)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 1)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (floormod(((threadIdx.x_1*2) + 1), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 5)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[((threadIdx.x_1*2) + 196)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (floormod((threadIdx.x_1*2), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 190)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 197)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (floormod(((threadIdx.x_1*2) + 1), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 191)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
+ kernel.shared_1[ramp((threadIdx.x_2*4), 1, 4)] = kernel[ramp(((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 2)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 2)*36)) + (ry.outer.outer*3)) + 2), 9, 4)]
}
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*16)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 32)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 64)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 96)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 128)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 160)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 192)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 224)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 1)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 33)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 65)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 97)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 129)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 161)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 193)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 225)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 8)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 40)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 72)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 104)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 136)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 200)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 232)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 9)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 41)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 73)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 105)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 137)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 201)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 233)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 2)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 34)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 66)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 98)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 130)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 162)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 194)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 226)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 3)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 35)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 67)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 99)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 131)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 163)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 195)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 227)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 10)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 42)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 74)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 106)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 138)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 202)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 234)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 11)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 43)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 75)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 107)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 139)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 203)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 235)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 4)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 36)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 68)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 100)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 132)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 164)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 196)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 228)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 5)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 37)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 69)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 101)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 133)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 165)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 197)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 229)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 12)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 44)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 76)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 108)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 140)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 204)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 236)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 13)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 45)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 77)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 109)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 141)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 205)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 237)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 6)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 38)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 70)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 102)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 134)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 166)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 198)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 230)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 7)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 39)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 71)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 103)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 135)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 167)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 199)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 231)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 14)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 46)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 78)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 110)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 142)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 206)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 238)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 15)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 47)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 79)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 111)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 143)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 207)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 239)]))
}
}
- compute[((blockIdx.x*784) + threadIdx.x)] = max((conv2d_nchw_1[0] + bias[((blockIdx.x*16) + floordiv(threadIdx.x, 49))]), 0f32)
- compute[(((blockIdx.x*784) + threadIdx.x) + 196)] = max((conv2d_nchw_1[1] + bias[(((blockIdx.x*16) + floordiv(threadIdx.x, 49)) + 4)]), 0f32)
- compute[(((blockIdx.x*784) + threadIdx.x) + 392)] = max((conv2d_nchw_1[2] + bias[(((blockIdx.x*16) + floordiv(threadIdx.x, 49)) + 8)]), 0f32)
- compute[(((blockIdx.x*784) + threadIdx.x) + 588)] = max((conv2d_nchw_1[3] + bias[(((blockIdx.x*16) + floordiv(threadIdx.x, 49)) + 12)]), 0f32)
+ for (i1.inner: int32, 0, 2) {
+ compute[((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 196)] = max((conv2d_nchw_1[(i1.inner + 2)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 4)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 392)] = max((conv2d_nchw_1[(i1.inner + 4)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 8)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 588)] = max((conv2d_nchw_1[(i1.inner + 6)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 12)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 784)] = max((conv2d_nchw_1[(i1.inner + 8)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 16)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 980)] = max((conv2d_nchw_1[(i1.inner + 10)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 20)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 1176)] = max((conv2d_nchw_1[(i1.inner + 12)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 24)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 1372)] = max((conv2d_nchw_1[(i1.inner + 14)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 28)]), 0f32)
+ }
}
}
@@ -740,7 +748,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.272 ms
+ Execution time of this operator: 0.333 ms
@@ -789,9 +797,9 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
- conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=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=4)
- 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=4)
+ conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=2)
+ conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=8)
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)
@@ -800,9 +808,9 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
- conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=32)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=1)
- conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+ conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_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=1)
@@ -810,9 +818,9 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
- compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=4)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=2)
+ compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=8)
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)
@@ -835,16 +843,16 @@ 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=1)
+ 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=4)
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=196)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
+ 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=2)
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=196)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
- s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+ s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -862,431 +870,454 @@ 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__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[4];
- __shared__ float pad_temp_shared[2016];
- __shared__ float kernel_shared[1536];
+ extern "C" __global__ void __launch_bounds__(98) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[16];
+ __shared__ float pad_temp_shared[392];
+ __shared__ float kernel_shared[256];
conv2d_nchw[0] = 0.000000e+00f;
- conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
+ conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
- for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
+ conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
+ for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
__syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = (((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 63) * 49)) + rx_outer_outer) + (((int)threadIdx.x) % 63)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 588)] = (((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 588) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((1 <= (((((int)threadIdx.x) / 7) + 4) % 9)) && ((((((int)threadIdx.x) / 7) + 4) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 784) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 4) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 980)] = (((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 980) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1176) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1372)] = (((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1372) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1568)] = (((((1 <= (((((int)threadIdx.x) / 7) + 8) % 9)) && ((((((int)threadIdx.x) / 7) + 8) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1568) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 8) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1764)] = (((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 63) * 49)) + rx_outer_outer) + (((int)threadIdx.x) % 63)) + 1364)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 56) {
- pad_temp_shared[(((int)threadIdx.x) + 1960)] = ((((((int)threadIdx.x) < 49) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1960) / 63) * 49)) + ((int)threadIdx.x)) + rx_outer_outer) - 1)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) * 2)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) * 2) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 1)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 2) + 1) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 196)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) * 2) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 188)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 197)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 2) + 1) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 189)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 64) {
+ int4 __1 = make_int4(((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)))+(9*0), ((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)))+(9*1), ((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) [...]
+ *(float4*)(kernel_shared + (((int)threadIdx.x) * 4)) = make_float4(kernel[__1.x],kernel[__1.y],kernel[__1.z],kernel[__1.w]);
}
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) % 96) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 196) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 4) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 392)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 392) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 8) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 2) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 588)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 588) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 4) & 31) * 9)) + ((((int)threadIdx.x) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 16) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 980)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 980) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 20) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 2) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1176) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 8) & 31) * 9)) + ((((int)threadIdx.x) % 3) * 3)) + rx_outer_outer)];
- if (((int)threadIdx.x) < 164) {
- kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1372) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 28) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + rx_outer_outer)];
+ __syncthreads();
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 16)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 32)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 64)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 96)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 128)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 192)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 224)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 33)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 65)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 97)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 129)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 193)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 225)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 8)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 40)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 72)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 104)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 136)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 200)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 232)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 9)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 41)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 73)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 105)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 137)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 201)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 233)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 2)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 34)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 66)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 98)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 130)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 194)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 226)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 3)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 35)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 67)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 99)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 131)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 195)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 227)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 10)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 42)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 74)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 106)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 138)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 202)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 234)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 11)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 43)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 75)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 107)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 139)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 203)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 235)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 4)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 36)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 68)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 100)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 132)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 196)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 228)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 5)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 37)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 69)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 101)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 133)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 197)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 229)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 12)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 44)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 76)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 108)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 140)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 204)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 236)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 13)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 45)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 77)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 109)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 141)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 205)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 237)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 6)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 38)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 70)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 102)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 134)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 198)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 230)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 7)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 39)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 71)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 103)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 135)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 199)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 231)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 14)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 46)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 78)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 110)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 142)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 206)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 238)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 15)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 47)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 79)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 111)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 143)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 207)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 239)]));
+ __syncthreads();
+ pad_temp_shared[(((int)threadIdx.x) * 2)] = (((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 1)] = (((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 196)] = (((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 189)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 197)] = (((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 190)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 64) {
+ int4 __2 = make_int4((((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 1))+(9*0), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 1))+(9*1), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x [...]
+ *(float4*)(kernel_shared + (((int)threadIdx.x) * 4)) = make_float4(kernel[__2.x],kernel[__2.y],kernel[__2.z],kernel[__2.w]);
}
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 384)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 768)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1152)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 385)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 769)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1153)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 386)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 770)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1154)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 387)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 771)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1155)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 388)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 772)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1156)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 389)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 773)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1157)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 390)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 774)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1158)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 391)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 775)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1159)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 392)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 776)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1160)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 393)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 777)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1161)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 394)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 778)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1162)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 395)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 779)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1163)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 396)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 780)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1164)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 397)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 781)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1165)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 398)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 782)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1166)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 399)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 783)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1167)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 400)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 784)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1168)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 401)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 785)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1169)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 402)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 786)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1170)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 403)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 787)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1171)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 404)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 788)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1172)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 405)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 789)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1173)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 406)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 790)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1174)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 407)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 791)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1175)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 408)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 792)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1176)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 409)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 793)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1177)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 410)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 794)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1178)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 411)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 795)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1179)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 412)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 796)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1180)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 413)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 797)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1181)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 414)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 798)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1182)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 415)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 799)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1183)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 416)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 800)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1184)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 417)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 801)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1185)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 418)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 802)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1186)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 419)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 803)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1187)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 420)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 804)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1188)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 421)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 805)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1189)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 422)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 806)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1190)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 423)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 807)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1191)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 424)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 808)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1192)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 425)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 809)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1193)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 426)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 810)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1194)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 427)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 811)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1195)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 428)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 812)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1196)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 429)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 813)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1197)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 430)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 814)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1198)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 431)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 815)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1199)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 432)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 816)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1200)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 433)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 817)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1201)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 434)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 818)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1202)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 435)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 819)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1203)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 436)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 820)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1204)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 437)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 821)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1205)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 438)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 822)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1206)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 439)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 823)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1207)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 440)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 824)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1208)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 441)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 825)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1209)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 442)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 826)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1210)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 443)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 827)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1211)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 444)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 828)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1212)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 445)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 829)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1213)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 446)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 830)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1214)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 447)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 831)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1215)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 448)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 832)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1216)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 449)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 833)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1217)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 450)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 834)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1218)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 451)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 835)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1219)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 452)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 836)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1220)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 453)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 837)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1221)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 454)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 838)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1222)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 455)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 839)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1223)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 456)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 840)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1224)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 457)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 841)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1225)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 458)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 842)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1226)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 459)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 843)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1227)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 460)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 844)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1228)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 461)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 845)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1229)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 462)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 846)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1230)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 463)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 847)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1231)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 464)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 848)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1232)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 465)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 849)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1233)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 466)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 850)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1234)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 467)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 851)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1235)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 468)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 852)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1236)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 469)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 853)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1237)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 470)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 854)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1238)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 471)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 855)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1239)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 472)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 856)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1240)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 473)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 857)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1241)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 474)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 858)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1242)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 475)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 859)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1243)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 476)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 860)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1244)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 477)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 861)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1245)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 478)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 862)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1246)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 479)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 863)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1247)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 16)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 32)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 64)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 96)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 128)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 192)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 224)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 33)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 65)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 97)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 129)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 193)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 225)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 8)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 40)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 72)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 104)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 136)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 200)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 232)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 9)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 41)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 73)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 105)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 137)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 201)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 233)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 2)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 34)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 66)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 98)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 130)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 194)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 226)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 3)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 35)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 67)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 99)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 131)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 195)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 227)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 10)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 42)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 74)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 106)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 138)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 202)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 234)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 11)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 43)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 75)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 107)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 139)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 203)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 235)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 4)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 36)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 68)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 100)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 132)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 196)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 228)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 5)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 37)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 69)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 101)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 133)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 197)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 229)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 12)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 44)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 76)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 108)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 140)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 204)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 236)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 13)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 45)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 77)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 109)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 141)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 205)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 237)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 6)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 38)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 70)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 102)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 134)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 198)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 230)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 7)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 39)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 71)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 103)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 135)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 199)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 231)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 14)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 46)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 78)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 110)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 142)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 206)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 238)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 15)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 47)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 79)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 111)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 143)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 207)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 239)]));
+ __syncthreads();
+ pad_temp_shared[(((int)threadIdx.x) * 2)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (((((int)threadIdx.x) * 2) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 1)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && ((((((int)threadIdx.x) * 2) + 1) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 196)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (((((int)threadIdx.x) * 2) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 190)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 197)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && ((((((int)threadIdx.x) * 2) + 1) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 191)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 64) {
+ int4 __3 = make_int4((((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 2))+(9*0), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 2))+(9*1), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x [...]
+ *(float4*)(kernel_shared + (((int)threadIdx.x) * 4)) = make_float4(kernel[__3.x],kernel[__3.y],kernel[__3.z],kernel[__3.w]);
+ }
+ __syncthreads();
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 16)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 32)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 64)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 96)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 128)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 192)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 224)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 33)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 65)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 97)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 129)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 193)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 225)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 8)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 40)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 72)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 104)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 136)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 200)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 232)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 9)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 41)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 73)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 105)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 137)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 201)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 233)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 2)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 34)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 66)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 98)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 130)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 194)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 226)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 3)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 35)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 67)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 99)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 131)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 195)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 227)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 10)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 42)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 74)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 106)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 138)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 202)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 234)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 11)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 43)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 75)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 107)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 139)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 203)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 235)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 4)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 36)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 68)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 100)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 132)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 196)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 228)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 5)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 37)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 69)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 101)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 133)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 197)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 229)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 12)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 44)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 76)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 108)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 140)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 204)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 236)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 13)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 45)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 77)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 109)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 141)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 205)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 237)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 6)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 38)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 70)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 102)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 134)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 198)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 230)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 7)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 39)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 71)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 103)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 135)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 199)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 231)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 14)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 46)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 78)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 110)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 142)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 206)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 238)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 15)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 47)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 79)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 111)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 143)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 207)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 239)]));
}
}
- compute[((((int)blockIdx.x) * 784) + ((int)threadIdx.x))] = max((conv2d_nchw[0] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49))]), 0.000000e+00f);
- compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[1] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 4)]), 0.000000e+00f);
- compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 392)] = max((conv2d_nchw[2] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 8)]), 0.000000e+00f);
- compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 588)] = max((conv2d_nchw[3] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 12)]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 196)] = max((conv2d_nchw[(i1_inner + 2)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 4)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 392)] = max((conv2d_nchw[(i1_inner + 4)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 8)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 588)] = max((conv2d_nchw[(i1_inner + 6)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 12)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 784)] = max((conv2d_nchw[(i1_inner + 8)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 980)] = max((conv2d_nchw[(i1_inner + 10)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 20)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 1176)] = max((conv2d_nchw[(i1_inner + 12)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 24)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 1372)] = max((conv2d_nchw[(i1_inner + 14)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 28)]), 0.000000e+00f);
+ }
}
@@ -1347,7 +1378,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 54.702 seconds)
+ **Total running time of the script:** ( 5 minutes 8.510 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 9ce0b2f653..9f73689dfe 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
@@ -643,7 +643,7 @@ 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)
- 8.1885 8.1875 8.1910 8.1871 0.0017
+ 8.1871 8.1868 8.1918 8.1828 0.0037
@@ -671,7 +671,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 1.447 seconds)
+ **Total running time of the script:** ( 1 minutes 4.102 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 6b72305a83..66475f4182 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
@@ -662,7 +662,7 @@ 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)
- 743.2654 743.3260 744.4337 742.0365 0.9796
+ 766.8664 766.8750 767.3471 766.3770 0.3961
@@ -690,7 +690,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 29.296 seconds)
+ **Total running time of the script:** ( 1 minutes 34.419 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 8f196b56da..2b25e823ed 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
@@ -386,28 +386,104 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 128) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [512]), storage_scope = global {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 16) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [512], [])[(((i.inner.init*32) + (nb_j.inner*16)) + j.init)] = 0f32
- }
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_9: placeholder_16: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 512) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [128]), storage_scope = global {
+ for (i.inner.init: int32, 0, 8) {
+ let cse_var_1: int32 = (i.inner.init*16)
+ {
+ compute_5: Buffer(compute_4, float32, [128], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (i.inner: int32, 0, 16) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- let cse_var_2: int32 = (((i.inner*32) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 16)*4096) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ for (elem_idx: int32, 0, let cse_var_2: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_2 + 1)] - placeholder_3[cse_var_2])) {
+ for (i.inner: int32, 0, 8) {
+ let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ {
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_4: int32 = (i.inner*16)
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[((placeholder_3[cse_var_3]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_5: int32 = ((i.inner*16) + 1)
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_6: int32 = ((i.inner*16) + 2)
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_7: int32 = ((i.inner*16) + 3)
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_8: int32 = ((i.inner*16) + 4)
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_9: int32 = ((i.inner*16) + 5)
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_10: int32 = ((i.inner*16) + 6)
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_11: int32 = ((i.inner*16) + 7)
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_12: int32 = ((i.inner*16) + 8)
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_13: int32 = ((i.inner*16) + 9)
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_14: int32 = ((i.inner*16) + 10)
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_15: int32 = ((i.inner*16) + 11)
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_16: int32 = ((i.inner*16) + 12)
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_17: int32 = ((i.inner*16) + 13)
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_18: int32 = ((i.inner*16) + 14)
+ compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_19: int32 = ((i.inner*16) + 15)
+ compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 16) {
- let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 8) {
+ let cse_var_20: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*4096) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_20, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_20, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -463,7 +539,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.508 ms
+ Execution time of this operator: 1.888 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 3968e22a4e..541037c847 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
Computation times
=================
-**00:34.812** total execution time for **how_to_tune_with_autotvm** files:
+**00:37.172** 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``) | 00:34.778 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:37.135 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.019 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.021 | 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 |
+--------------------------------------------------------------------------------------------------+-----------+--------+
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 bc5594c1dc..49af95510a 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
@@ -387,7 +387,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8046651
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7876715
No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -510,7 +510,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2799895
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9549576
No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -633,9 +633,9 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6746199
- No: 4 GFLOPS: 123.68/123.68 result: MeasureResult(costs=(0.001871792546875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.121532678604126, timestamp=1666819125.700304) [('tile_f', [-1, 1, 64, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7648213
- No: 5 GFLOPS: 0.00/123.68 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8447313
+ No: 4 GFLOPS: 1.29/1.29 result: MeasureResult(costs=(0.179328308,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.749636173248291, timestamp=1666822077.5919812) [('tile_f', [-1, 4, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8557307
+ No: 5 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -757,11 +757,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9742540
- No: 6 GFLOPS: 67.39/123.68 result: MeasureResult(costs=(0.0034352445000000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.685016870498657, timestamp=1666819130.5838377) [('tile_f', [-1, 8, 1, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4355178
- No: 7 GFLOPS: 167.02/167.02 result: MeasureResult(costs=(0.0013860574594594594,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6607511043548584, timestamp=1666819132.1200016) [('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7403292
- No: 8 GFLOPS: 31.14/167.02 result: MeasureResult(costs=(0.007434749571428571,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5834624767303467, timestamp=1666819132.8072135) [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1938378
- No: 9 GFLOPS: 0.00/167.02 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 8, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8315757
+ No: 6 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -883,11 +880,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 8, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6020236
- No: 10 GFLOPS: 61.38/167.02 result: MeasureResult(costs=(0.003771696363636364,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.913536787033081, timestamp=1666819137.3885193) [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8728026
- No: 11 GFLOPS: 705.96/705.96 result: MeasureResult(costs=(0.0003279260647249191,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2615830898284912, timestamp=1666819138.0211608) [('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7753312
- No: 12 GFLOPS: 69.82/705.96 result: MeasureResult(costs=(0.003315454791666667,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.487569570541382, timestamp=1666819138.9851844) [('tile_f', [-1, 1, 2, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9423808
- No: 13 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6919113
+ No: 7 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1009,8 +1003,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 16, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9829069
- No: 14 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 8]), ('tile_y', [-1, 1, 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', 0), ('unroll_explicit', 1)],None,6181257
+ No: 8 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1132,8 +1126,26 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1036307
- No: 15 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1913751
+ No: 9 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
+ res = future.result()
+ File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
+ return self.__get_result()
+ File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
+ raise self._exception
+ File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
+ result = self.fn(*self.args, **self.kwargs)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
+ worker = lambda *args: self._worker_run(*args)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
+ return proc.recv()
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
+ raise TimeoutError()
+ TimeoutError
+
+ [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9466171
+ No: 10 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1255,8 +1267,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 1, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9511704
- No: 16 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7060360
+ No: 11 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1378,10 +1390,9 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4813683
- No: 17 GFLOPS: 90.03/705.96 result: MeasureResult(costs=(0.002571501897435897,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.409144639968872, timestamp=1666819142.0864537) [('tile_f', [-1, 4, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2526554
- No: 18 GFLOPS: 21.16/705.96 result: MeasureResult(costs=(0.0109404319,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.6973342895507812, timestamp=1666819142.8425627) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8586278
- No: 19 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5223606
+ No: 12 GFLOPS: 145.84/145.84 result: MeasureResult(costs=(0.0015873300317460317,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7088942527770996, timestamp=1666822089.8054924) [('tile_f', [-1, 8, 8, 1]), ('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, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8725010
+ No: 13 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1503,8 +1514,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 128, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6775831
- No: 20 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6542100
+ No: 14 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1626,7 +1637,623 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 2, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3893234
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5646420
+ No: 15 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3322291
+ No: 16 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10438315
+ No: 17 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8412683
+ No: 18 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2106803
+ No: 19 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,825458
+ No: 20 GFLOPS: 74.85/145.84 result: MeasureResult(costs=(0.003092781515151515,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.605224609375, timestamp=1666822093.1458662) [('tile_f', [-1, 2, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1233575
@@ -1681,9 +2308,9 @@ and measure running time.
Finish loading 20 records
Best config:
- [('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7753312
+ [('tile_f', [-1, 8, 8, 1]), ('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, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8725010
Finish loading 20 records
- Time cost of this operator: 0.000739
+ Time cost of this operator: 0.000958
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 bc68849968..aaa184c1f6 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
@@ -327,10 +327,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 311.1 98.723 (1, 2, 10, 10, 3) 2 1 [311.1]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.042 0.965 (1, 6, 10, 10) 1 1 [3.042]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.983 0.312 (1, 1, 10, 10, 3) 1 1 [0.983]
- Total_time - 315.125 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 311.6 98.733 (1, 2, 10, 10, 3) 2 1 [311.6]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.027 0.959 (1, 6, 10, 10) 1 1 [3.027]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.973 0.308 (1, 1, 10, 10, 3) 1 1 [0.973]
+ Total_time - 315.6 - - - - -
@@ -394,10 +394,10 @@ 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 127.6 97.879 (1, 6, 10, 10, 1) 2 1 [127.6]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.811 1.389 (1, 6, 10, 10) 1 1 [1.811]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.954 0.732 (1, 1, 10, 10, 3) 1 1 [0.954]
- Total_time - 130.365 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 105.8 97.563 (1, 6, 10, 10, 1) 2 1 [105.8]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.796 1.656 (1, 6, 10, 10) 1 1 [1.796]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.847 0.781 (1, 3, 10, 10, 1) 1 1 [0.847]
+ Total_time - 108.443 - - - - -
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 8d36b93430..68281e0d5d 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -225,7 +225,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmp1suj57_8/images/random'
+ '/tmp/tmpbu2mt11i/images/random'
@@ -316,7 +316,7 @@ 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], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]
+ :alt: [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]
:srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
:class: sphx-glr-single-img
@@ -325,8 +325,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmp1suj57_8/images/target contains 8144 images
- /tmp/tmp1suj57_8/images/random contains 5000 images
+ /tmp/tmpbu2mt11i/images/target contains 8144 images
+ /tmp/tmpbu2mt11i/images/random contains 5000 images
@@ -501,13 +501,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 46s - loss: 0.2086 - accuracy: 0.9282 - val_loss: 0.2560 - val_accuracy: 0.9248 - 46s/epoch - 141ms/step
+ 328/328 - 47s - loss: 0.2181 - accuracy: 0.9245 - val_loss: 0.1206 - val_accuracy: 0.9566 - 47s/epoch - 144ms/step
Epoch 2/3
- 328/328 - 42s - loss: 0.0914 - accuracy: 0.9654 - val_loss: 0.1092 - val_accuracy: 0.9615 - 42s/epoch - 129ms/step
+ 328/328 - 44s - loss: 0.1039 - accuracy: 0.9615 - val_loss: 0.1075 - val_accuracy: 0.9600 - 44s/epoch - 133ms/step
Epoch 3/3
- 328/328 - 42s - loss: 0.0709 - accuracy: 0.9767 - val_loss: 0.1294 - val_accuracy: 0.9619 - 42s/epoch - 130ms/step
+ 328/328 - 44s - loss: 0.0673 - accuracy: 0.9749 - val_loss: 0.1221 - val_accuracy: 0.9619 - 44s/epoch - 133ms/step
- <keras.callbacks.History object at 0x7f57883edc50>
+ <keras.callbacks.History object at 0x7f1333acc950>
@@ -864,7 +864,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 37.024 seconds)
+ **Total running time of the script:** ( 4 minutes 42.046 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 113d80d1e9..5db0399899 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
@@ -5,16 +5,16 @@
Computation times
=================
-**05:37.830** total execution time for **how_to_work_with_microtvm** files:
+**05:47.208** total execution time for **how_to_work_with_microtvm** files:
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:37.024 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:42.046 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:48.043 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:51.418 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:09.126 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:09.811 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.635 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.930 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``) | 00:00.001 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index 88bde71baa..871444e57e 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
=================
-**00:44.332** total execution time for **how_to_work_with_relay** files:
+**00:46.966** 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``) | 00:32.689 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:34.811 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:10.113 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:10.510 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.523 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.637 | 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 4855d11fff..70c35566f6 100644
--- a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
@@ -261,7 +261,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7f5738aae830>
+ <function my_cuda_math_rule at 0x7f136402fdd0>
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 4a1147646e..75307e7bc7 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
=================
-**00:08.076** total execution time for **how_to_work_with_schedules** files:
+**00:08.715** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:05.760 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:06.239 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.020 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.122 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.553 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.580 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.537 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.552 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.115 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.121 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.046 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.052 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.029 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.031 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.016 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``) | 00:00.018 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
index 9ffbde4ab5..1082e3e9f4 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -347,7 +347,7 @@ The importing needs to happen before the tensorized GEMV being executed.
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmphhwb1r7r/input0.cc'\nsource_filename = \"/tmp/tmphhwb1r7r/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpnv73ocip/input0.cc'\nsource_filename = \"/tmp/tmpnv73ocip/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 001617262b..24984941b6 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
=================
-**00:25.051** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:27.264** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:25.045 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:27.258 | 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 d1ce59454d..f0b083a673 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -289,7 +289,7 @@ 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 27.64s!
+ resnet18_v1 inference graph built in 30.71s!
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 6eb6708761..db5ade165e 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -333,7 +333,7 @@ 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 18.77s!
+ yolov3-tiny inference graph built in 20.54s!
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 d4ffabec14..14aa63bd61 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
=================
-**01:37.692** total execution time for **topic_vta_tutorials_frontend** files:
+**01:43.621** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:50.506 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:52.706 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:47.187 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:50.915 | 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 c578e7b0e2..087bfb39c9 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
=================
-**00:03.074** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.103** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.648 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.655 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.426 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.449 | 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 ea6b9ef1c7..0ba5278cf0 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
=================
-**00:00.752** total execution time for **topic_vta_tutorials** files:
+**00:00.809** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.406 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.440 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.347 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.369 | 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 8ffd87b195..bf91d657c1 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -207,8 +207,8 @@ trials, we can load the best schedule from the log file and apply it.
.. code-block:: none
- *E
+ *E*E
@@ -333,7 +333,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 96.875 ms
+ Execution time of this operator: 99.469 ms
@@ -451,7 +451,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 32.489 seconds)
+ **Total running time of the script:** ( 1 minutes 44.723 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 c78a9d380f..7d878a7ad7 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -450,16 +450,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 10.63/10.63 result: MeasureResult(costs=(0.0252536908,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.620704174041748, timestamp=1666817750.891999) [('tile_y', [-1, 512]), ('tile_x', [-1, 128])],None,79
- No: 2 GFLOPS: 1.70/10.63 result: MeasureResult(costs=(0.1578523062,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.7097103595733643, timestamp=1666817753.6281903) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
- No: 3 GFLOPS: 8.82/10.63 result: MeasureResult(costs=(0.030430965200000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8486270904541016, timestamp=1666817755.0237305) [('tile_y', [-1, 2]), ('tile_x', [-1, 128])],None,71
- No: 4 GFLOPS: 13.53/13.53 result: MeasureResult(costs=(0.019834976,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5051016807556152, timestamp=1666817756.2478125) [('tile_y', [-1, 16]), ('tile_x', [-1, 512])],None,94
- No: 5 GFLOPS: 13.00/13.53 result: MeasureResult(costs=(0.020646058000000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4988253116607666, timestamp=1666817756.9051504) [('tile_y', [-1, 64]), ('tile_x', [-1, 512])],None,96
- No: 6 GFLOPS: 1.71/13.53 result: MeasureResult(costs=(0.1571568272,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.6974995136260986, timestamp=1666817759.6328373) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
- No: 7 GFLOPS: 3.93/13.53 result: MeasureResult(costs=(0.06837389399999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2710866928100586, timestamp=1666817761.6293411) [('tile_y', [-1, 32]), ('tile_x', [-1, 16])],None,45
- No: 8 GFLOPS: 9.18/13.53 result: MeasureResult(costs=(0.0292458614,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7069270610809326, timestamp=1666817762.3018649) [('tile_y', [-1, 16]), ('tile_x', [-1, 32])],None,54
- No: 9 GFLOPS: 1.92/13.53 result: MeasureResult(costs=(0.1395043818,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3404064178466797, timestamp=1666817764.7610335) [('tile_y', [-1, 1]), ('tile_x', [-1, 8])],None,30
- No: 10 GFLOPS: 11.28/13.53 result: MeasureResult(costs=(0.0238029414,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.49455833435058594, timestamp=1666817765.3044648) [('tile_y', [-1, 256]), ('tile_x', [-1, 32])],None,58
+ No: 1 GFLOPS: 1.58/1.58 result: MeasureResult(costs=(0.16970374119999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.875869035720825, timestamp=1666820602.6748846) [('tile_y', [-1, 32]), ('tile_x', [-1, 4])],None,25
+ No: 2 GFLOPS: 2.59/2.59 result: MeasureResult(costs=(0.1034866262,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8659942150115967, timestamp=1666820604.5371964) [('tile_y', [-1, 8]), ('tile_x', [-1, 4])],None,23
+ No: 3 GFLOPS: 2.11/2.59 result: MeasureResult(costs=(0.1269757282,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.188803195953369, timestamp=1666820607.5032542) [('tile_y', [-1, 8]), ('tile_x', [-1, 2])],None,13
+ No: 4 GFLOPS: 13.02/13.02 result: MeasureResult(costs=(0.020618882999999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.48042750358581543, timestamp=1666820608.0063584) [('tile_y', [-1, 4]), ('tile_x', [-1, 512])],None,92
+ No: 5 GFLOPS: 12.74/13.02 result: MeasureResult(costs=(0.0210762236,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4952247142791748, timestamp=1666820608.6314328) [('tile_y', [-1, 8]), ('tile_x', [-1, 512])],None,93
+ No: 6 GFLOPS: 3.02/13.02 result: MeasureResult(costs=(0.0888996746,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5804462432861328, timestamp=1666820610.989842) [('tile_y', [-1, 256]), ('tile_x', [-1, 8])],None,38
+ No: 7 GFLOPS: 1.76/13.02 result: MeasureResult(costs=(0.1524931758,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5825090408325195, timestamp=1666820614.3720727) [('tile_y', [-1, 4]), ('tile_x', [-1, 1])],None,2
+ No: 8 GFLOPS: 0.50/13.02 result: MeasureResult(costs=(0.5335169732,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.736244201660156, timestamp=1666820623.1172493) [('tile_y', [-1, 64]), ('tile_x', [-1, 1])],None,6
+ No: 9 GFLOPS: 2.84/13.02 result: MeasureResult(costs=(0.0944008996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.640514850616455, timestamp=1666820624.961784) [('tile_y', [-1, 4]), ('tile_x', [-1, 8])],None,32
+ No: 10 GFLOPS: 12.62/13.02 result: MeasureResult(costs=(0.021276204400000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.560152530670166, timestamp=1666820625.4736116) [('tile_y', [-1, 64]), ('tile_x', [-1, 128])],None,76
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 473131e11a..ab2722d0da 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -320,7 +320,7 @@ standard deviation.
.. code-block:: none
- {'mean': 513.548849240002, 'median': 513.8221270500026, 'std': 2.8991952502211213}
+ {'mean': 516.2833439300039, 'median': 516.0584373500114, 'std': 2.27093118676701}
@@ -554,31 +554,29 @@ 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: 11.27/ 15.90 GFLOPS | Progress: (4/20) | 7.41 s
[Task 1/25] Current/Best: 19.40/ 19.40 GFLOPS | Progress: (8/20) | 9.56 s
[Task 1/25] Current/Best: 9.01/ 19.40 GFLOPS | Progress: (12/20) | 12.78 s
[Task 1/25] Current/Best: 11.33/ 19.40 GFLOPS | Progress: (16/20) | 15.36 s
[Task 1/25] Current/Best: 14.27/ 19.40 GFLOPS | Progress: (20/20) | 19.11 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.79/ 16.16 GFLOPS | Progress: (4/20) | 2.89 s
[Task 2/25] Current/Best: 5.19/ 21.09 GFLOPS | Progress: (8/20) | 4.10 s
[Task 2/25] Current/Best: 8.78/ 21.09 GFLOPS | Progress: (12/20) | 5.25 s
[Task 2/25] Current/Best: 16.94/ 21.09 GFLOPS | Progress: (16/20) | 7.80 s
[Task 2/25] Current/Best: 18.39/ 21.09 GFLOPS | Progress: (20/20) | 9.20 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 17.08/ 17.08 GFLOPS | Progress: (4/20) | 4.21 s
[Task 3/25] Current/Best: 12.35/ 17.08 GFLOPS | Progress: (8/20) | 6.07 s
[Task 3/25] Current/Best: 9.43/ 18.50 GFLOPS | Progress: (12/20) | 8.03 s
[Task 3/25] Current/Best: 17.85/ 21.46 GFLOPS | Progress: (16/20) | 9.60 s
[Task 3/25] Current/Best: 11.14/ 21.46 GFLOPS | Progress: (20/20) | 11.52 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 17.43/ 17.43 GFLOPS | Progress: (4/20) | 3.33 s
[Task 4/25] Current/Best: 16.21/ 17.43 GFLOPS | Progress: (8/20) | 5.10 s
[Task 4/25] Current/Best: 9.03/ 17.43 GFLOPS | Progress: (12/20) | 12.47 s
[Task 4/25] Current/Best: 18.19/ 18.19 GFLOPS | Progress: (16/20) | 13.86 s
[Task 4/25] Current/Best: 10.87/ 18.19 GFLOPS | Progress: (20/20) | 16.57 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 10.91/ 18.05 GFLOPS | Progress: (4/20) | 3.63 s
[Task 5/25] Current/Best: 20.46/ 21.48 GFLOPS | Progress: (8/20) | 5.24 s
[Task 5/25] Current/Best: 16.06/ 21.48 GFLOPS | Progress: (12/20) | 6.86 s
[Task 5/25] Current/Best: 4.97/ 21.48 GFLOPS | Progress: (16/20) | 8.90 s
[Task 5/25] Current/Best: 4.80/ 21.48 GFLOPS | Progress: (20/20) | 11.41 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 14.37/ 14.37 GFLOPS | Progress: (4/20) | 4.00 s
[Task 6/25] Current/Best: 5.80/ 14.37 GFLOPS | Progress: (8/20) | 7.34 s
[Task 6/25] Current/Best: 9.40/ 14.37 GFLOPS | Progress: (12/20) | 14.11 s
[Task 6/25] Current/Best: 5.60/ 16.06 GFLOPS | Progress: (16/20) | 17.16 s
[Task 6/25] Current/Best: 13.95/ 16.06 GFLOPS | Progress: (20/20) | 20.13 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 13.35/ 16.24 GFLOPS | Progress: (4/20) | 4.13 s
[Task 7/25] Current/Best: 18.28/ 18.28 GFLOPS | Progress: (8/20) | 6.13 s
[Task 7/25] Current/Best: 14.10/ 18.28 GFLOPS | Progress: (12/20) | 8.23 s
[Task 7/25] Current/Best: 18.77/ 22.93 GFLOPS | Progress: (16/20) | 10.57 s
[Task 7/25] Current/Best: 14.83/ 22.93 GFLOPS | Progress: (20/20) | 13.58 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 2.67/ 18.86 GFLOPS | Progress: (4/20) | 3.93 s
[Task 8/25] Current/Best: 12.48/ 18.86 GFLOPS | Progress: (8/20) | 13.28 s
[Task 8/25] Current/Best: 13.77/ 18.86 GFLOPS | Progress: (12/20) | 17.70 s
[Task 8/25] Current/Best: 2.96/ 18.86 GFLOPS | Progress: (16/20) | 20.23 s
[Task 8/25] Current/Best: 3.66/ 18.86 GFLOPS | Progress: (20/20) | 27.85 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 5.98/ 22.07 GFLOPS | Progress: (4/20) | 3.51 s
[Task 9/25] Current/Best: 22.73/ 22.73 GFLOPS | Progress: (8/20) | 8.10 s
[Task 9/25] Current/Best: 9.12/ 22.73 GFLOPS | Progress: (12/20) | 18.85 s
[Task 9/25] Current/Best: 6.78/ 22.73 GFLOPS | Progress: (16/20) | 22.93 s
[Task 9/25] Current/Best: 7.26/ 23.67 GFLOPS | Progress: (20/20) | 24.38 s Done.
-
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 14.28/ 17.84 GFLOPS | Progress: (4/20) | 3.22 s
[Task 10/25] Current/Best: 12.28/ 17.84 GFLOPS | Progress: (8/20) | 4.98 s
[Task 10/25] Current/Best: 13.94/ 17.84 GFLOPS | Progress: (12/20) | 6.66 s
[Task 10/25] Current/Best: 6.25/ 17.84 GFLOPS | Progress: (16/20) | 8.31 s
[Task 10/25] Current/Best: 10.27/ 17.84 GFLOPS | Progress: (20/20) | 10.66 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 10.71/ 17.95 GFLOPS | Progress: (4/20) | 3.68 s
[Task 11/25] Current/Best: 7.02/ 20.77 GFLOPS | Progress: (8/20) | 5.51 s
[Task 11/25] Current/Best: 8.51/ 20.77 GFLOPS | Progress: (12/20) | 7.44 s
[Task 11/25] Current/Best: 16.28/ 20.77 GFLOPS | Progress: (16/20) | 9.51 s
[Task 11/25] Current/Best: 8.82/ 20.77 GFLOPS | Progress: (20/20) | 12.32 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 4.33/ 14.16 GFLOPS | Progress: (4/20) | 5.65 s
[Task 12/25] Current/Best: 10.98/ 22.26 GFLOPS | Progress: (8/20) | 8.43 s
[Task 12/25] Current/Best: 12.37/ 22.26 GFLOPS | Progress: (12/20) | 11.23 s
[Task 12/25] Current/Best: 11.87/ 22.26 GFLOPS | Progress: (16/20) | 16.78 s
[Task 12/25] Current/Best: 10.83/ 22.26 GFLOPS | Progress: (20/20) | 19.16 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 6.70/ 18.31 GFLOPS | Progress: (4/20) | 3.47 s
[Task 13/25] Current/Best: 9.36/ 18.31 GFLOPS | Progress: (8/20) | 5.66 s
[Task 13/25] Current/Best: 15.74/ 20.22 GFLOPS | Progress: (12/20) | 7.70 s
[Task 13/25] Current/Best: 17.22/ 20.22 GFLOPS | Progress: (16/20) | 10.64 s
[Task 13/25] Current/Best: 15.64/ 20.22 GFLOPS | Progress: (20/20) | 14.42 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 13.08/ 13.08 GFLOPS | Progress: (4/20) | 3.67 s
[Task 14/25] Current/Best: 13.25/ 14.83 GFLOPS | Progress: (8/20) | 6.05 s
[Task 14/25] Current/Best: 3.01/ 14.83 GFLOPS | Progress: (12/20) | 10.42 s
[Task 14/25] Current/Best: 6.30/ 20.55 GFLOPS | Progress: (16/20) | 12.15 s
[Task 14/25] Current/Best: 15.79/ 20.55 GFLOPS | Progress: (20/20) | 13.98 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 16.03/ 16.03 GFLOPS | Progress: (4/20) | 3.74 s
[Task 15/25] Current/Best: 16.74/ 18.67 GFLOPS | Progress: (8/20) | 6.01 s Done.
-
[Task 15/25] Current/Best: 19.10/ 19.83 GFLOPS | Progress: (12/20) | 7.92 s
[Task 15/25] Current/Best: 3.15/ 19.83 GFLOPS | Progress: (16/20) | 16.28 s
[Task 15/25] Current/Best: 8.46/ 19.83 GFLOPS | Progress: (20/20) | 22.79 s Done.
-
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 5.81/ 16.86 GFLOPS | Progress: (4/20) | 3.54 s
[Task 16/25] Current/Best: 13.49/ 21.05 GFLOPS | Progress: (8/20) | 5.05 s
[Task 16/25] Current/Best: 6.15/ 21.05 GFLOPS | Progress: (12/20) | 6.75 s
[Task 16/25] Current/Best: 7.44/ 21.05 GFLOPS | Progress: (16/20) | 8.44 s
[Task 16/25] Current/Best: 16.30/ 21.05 GFLOPS | Progress: (20/20) | 9.89 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 17.88/ 22.23 GFLOPS | Progress: (4/20) | 3.80 s
[Task 17/25] Current/Best: 12.28/ 22.57 GFLOPS | Progress: (8/20) | 5.55 s
[Task 17/25] Current/Best: 12.22/ 22.57 GFLOPS | Progress: (12/20) | 8.92 s
[Task 17/25] Current/Best: 19.63/ 22.57 GFLOPS | Progress: (16/20) | 11.65 s
[Task 17/25] Current/Best: 17.35/ 22.57 GFLOPS | Progress: (20/20) | 13.87 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 19.35/ 19.35 GFLOPS | Progress: (4/20) | 3.43 s
[Task 18/25] Current/Best: 13.84/ 19.35 GFLOPS | Progress: (8/20) | 6.38 s
[Task 18/25] Current/Best: 14.62/ 19.35 GFLOPS | Progress: (12/20) | 12.49 s
[Task 18/25] Current/Best: 11.99/ 20.53 GFLOPS | Progress: (16/20) | 15.95 s
[Task 18/25] Current/Best: 14.67/ 20.53 GFLOPS | Progress: (20/20) | 20.70 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 8.27/ 18.32 GFLOPS | Progress: (4/20) | 6.28 s
[Task 19/25] Current/Best: 7.58/ 18.78 GFLOPS | Progress: (8/20) | 9.24 s
[Task 19/25] Current/Best: 18.61/ 19.88 GFLOPS | Progress: (12/20) | 13.24 s
[Task 19/25] Current/Best: 11.43/ 19.88 GFLOPS | Progress: (16/20) | 16.77 s
[Task 19/25] Current/Best: 10.20/ 19.88 GFLOPS | Progress: (20/20) | 19.33 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 3.84/ 16.45 GFLOPS | Progress: (4/20) | 3.61 s
[Task 20/25] Current/Best: 11.29/ 16.45 GFLOPS | Progress: (8/20) | 5.83 s
[Task 20/25] Current/Best: 11.45/ 16.45 GFLOPS | Progress: (12/20) | 8.60 s
[Task 20/25] Current/Best: 19.35/ 19.35 GFLOPS | Progress: (16/20) | 11.90 s
[Task 20/25] Current/Best: 16.16/ 19.35 GFLOPS | Progress: (20/20) | 13.81 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 12.07/ 12.07 GFLOPS | Progress: (4/20) | 4.59 s
[Task 21/25] Current/Best: 2.74/ 16.26 GFLOPS | Progress: (8/20) | 8.54 s
[Task 21/25] Current/Best: 15.69/ 17.24 GFLOPS | Progress: (12/20) | 9.78 s Done.
-
[Task 21/25] Current/Best: 12.85/ 17.24 GFLOPS | Progress: (16/20) | 12.35 s
[Task 21/25] Current/Best: 2.70/ 19.01 GFLOPS | Progress: (20/20) | 13.90 s Done.
-
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 7.73/ 14.42 GFLOPS | Progress: (4/20) | 3.92 s
[Task 22/25] Current/Best: 19.86/ 19.86 GFLOPS | Progress: (8/20) | 5.29 s
[Task 22/25] Current/Best: 11.36/ 19.86 GFLOPS | Progress: (12/20) | 7.32 s
[Task 22/25] Current/Best: 5.21/ 19.86 GFLOPS | Progress: (16/20) | 10.16 s
[Task 22/25] Current/Best: 18.89/ 19.86 GFLOPS | Progress: (20/20) | 12.53 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 12.78/ 22.30 GFLOPS | Progress: (4/20) | 3.53 s
[Task 23/25] Current/Best: 10.33/ 22.30 GFLOPS | Progress: (8/20) | 9.34 s
[Task 23/25] Current/Best: 9.66/ 22.30 GFLOPS | Progress: (12/20) | 11.55 s
[Task 23/25] Current/Best: 14.95/ 22.30 GFLOPS | Progress: (16/20) | 14.00 s
[Task 23/25] Current/Best: 9.55/ 22.30 GFLOPS | Progress: (20/20) | 16.59 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.46/ 8.46 GFLOPS | Progress: (4/20) | 12.22 s
[Task 24/25] Current/Best: 3.58/ 9.24 GFLOPS | Progress: (8/20) | 23.76 s
[Task 24/25] Current/Best: 3.74/ 9.24 GFLOPS | Progress: (12/20) | 34.28 s
[Task 24/25] Current/Best: 3.05/ 9.24 GFLOPS | Progress: (16/20) | 45.91 s
[Task 24/25] Current/Best: 3.43/ 9.24 GFLOPS | Progress: (20/20) | 56.63 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
[Task 25/25] Current/Best: 3.56/ 8.43 GFLOPS | Progress: (4/20) | 13.01 s
[Task 25/25] Current/Best: 3.03/ 8.43 GFLOPS | Progress: (8/20) | 23.74 s
[Task 25/25] Current/Best: 3.01/ 8.43 GFLOPS | Progress: (12/20) | 25.49 s
[Task 25/25] Current/Best: 1.54/ 8.43 GFLOPS | Progress: (16/20) | 35.97 s
[Task 25/25] Current/Best: 8.69/ 8.69 GFLOPS | Progress: (20/20) | 47.20 s
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 15.32/ 22.73 GFLOPS | Progress: (4/20) | 8.39 s
[Task 1/25] Current/Best: 17.32/ 22.73 GFLOPS | Progress: (8/20) | 11.70 s
[Task 1/25] Current/Best: 3.21/ 22.73 GFLOPS | Progress: (12/20) | 14.44 s
[Task 1/25] Current/Best: 9.78/ 22.73 GFLOPS | Progress: (16/20) | 17.70 s
[Task 1/25] Current/Best: 18.45/ 22.73 GFLOPS | Progress: (20/20) | 20.70 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 14.59/ 17.78 GFLOPS | Progress: (4/20) | 2.90 s
[Task 2/25] Current/Best: 15.78/ 18.25 GFLOPS | Progress: (8/20) | 4.16 s
[Task 2/25] Current/Best: 10.42/ 18.25 GFLOPS | Progress: (12/20) | 5.50 s
[Task 2/25] Current/Best: 8.78/ 18.25 GFLOPS | Progress: (16/20) | 6.92 s
[Task 2/25] Current/Best: 2.03/ 19.26 GFLOPS | Progress: (20/20) | 8.03 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 13.82/ 14.39 GFLOPS | Progress: (4/20) | 3.58 s
[Task 3/25] Current/Best: 8.88/ 18.27 GFLOPS | Progress: (8/20) | 5.57 s
[Task 3/25] Current/Best: 12.35/ 18.27 GFLOPS | Progress: (12/20) | 7.86 s
[Task 3/25] Current/Best: 12.69/ 19.41 GFLOPS | Progress: (16/20) | 10.42 s
[Task 3/25] Current/Best: 13.21/ 19.41 GFLOPS | Progress: (20/20) | 13.56 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 13.61/ 18.27 GFLOPS | Progress: (4/20) | 3.03 s
[Task 4/25] Current/Best: 15.57/ 18.27 GFLOPS | Progress: (8/20) | 8.45 s
[Task 4/25] Current/Best: 11.19/ 18.27 GFLOPS | Progress: (12/20) | 11.15 s
[Task 4/25] Current/Best: 9.80/ 18.27 GFLOPS | Progress: (16/20) | 13.73 s
[Task 4/25] Current/Best: 6.67/ 18.27 GFLOPS | Progress: (20/20) | 15.22 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 10.96/ 15.96 GFLOPS | Progress: (4/20) | 3.60 s
[Task 5/25] Current/Best: 13.74/ 19.28 GFLOPS | Progress: (8/20) | 5.68 s
[Task 5/25] Current/Best: 8.96/ 19.28 GFLOPS | Progress: (12/20) | 7.42 s
[Task 5/25] Current/Best: 12.74/ 19.28 GFLOPS | Progress: (16/20) | 9.02 s
[Task 5/25] Current/Best: 19.40/ 19.40 GFLOPS | Progress: (20/20) | 10.97 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 4.64/ 17.37 GFLOPS | Progress: (4/20) | 4.21 s
[Task 6/25] Current/Best: 14.08/ 17.37 GFLOPS | Progress: (8/20) | 6.67 s
[Task 6/25] Current/Best: 7.48/ 17.37 GFLOPS | Progress: (12/20) | 8.82 s
[Task 6/25] Current/Best: 15.88/ 17.54 GFLOPS | Progress: (16/20) | 15.29 s
[Task 6/25] Current/Best: 12.08/ 18.37 GFLOPS | Progress: (20/20) | 17.35 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 12.27/ 21.26 GFLOPS | Progress: (4/20) | 3.97 s
[Task 7/25] Current/Best: 12.98/ 21.26 GFLOPS | Progress: (8/20) | 5.77 s
[Task 7/25] Current/Best: 8.87/ 21.26 GFLOPS | Progress: (12/20) | 8.08 s
[Task 7/25] Current/Best: 16.73/ 22.67 GFLOPS | Progress: (16/20) | 9.59 s
[Task 7/25] Current/Best: 6.56/ 22.67 GFLOPS | Progress: (20/20) | 12.23 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 1.69/ 15.87 GFLOPS | Progress: (4/20) | 13.92 s
[Task 8/25] Current/Best: 9.34/ 17.26 GFLOPS | Progress: (8/20) | 26.07 s
[Task 8/25] Current/Best: 7.78/ 17.84 GFLOPS | Progress: (12/20) | 30.79 s
[Task 8/25] Current/Best: 10.00/ 19.98 GFLOPS | Progress: (16/20) | 35.20 s
[Task 8/25] Current/Best: 4.72/ 19.98 GFLOPS | Progress: (20/20) | 46.83 s
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 13.93/ 16.18 GFLOPS | Progress: (4/20) | 3.66 s
[Task 9/25] Current/Best: 13.36/ 16.18 GFLOPS | Progress: (8/20) | 6.32 s
[Task 9/25] Current/Best: 6.19/ 16.18 GFLOPS | Progress: (12/20) | 9.56 s
[Task 9/25] Current/Best: 16.45/ 16.45 GFLOPS | Progress: (16/20) | 11.17 s
[Task 9/25] Current/Best: 3.19/ 17.95 GFLOPS | Progress: (20/2
0) | 13.46 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 20.21/ 20.21 GFLOPS | Progress: (4/20) | 3.23 s
[Task 10/25] Current/Best: 17.20/ 20.21 GFLOPS | Progress: (8/20) | 4.63 s
[Task 10/25] Current/Best: 18.27/ 20.21 GFLOPS | Progress: (12/20) | 7.14 s
[Task 10/25] Current/Best: 8.18/ 20.21 GFLOPS | Progress: (16/20) | 10.11 s
[Task 10/25] Current/Best: 13.42/ 20.21 GFLOPS | Progress: (20/20) | 13.42 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 12.23/ 12.23 GFLOPS | Progress: (4/20) | 4.26 s
[Task 11/25] Current/Best: 12.61/ 18.51 GFLOPS | Progress: (8/20) | 6.40 s
[Task 11/25] Current/Best: 7.75/ 18.51 GFLOPS | Progress: (12/20) | 8.77 s
[Task 11/25] Current/Best: 11.94/ 18.51 GFLOPS | Progress: (16/20) | 11.23 s
[Task 11/25] Current/Best: 12.21/ 18.51 GFLOPS | Progress: (20/20) | 15.42 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 10.36/ 19.06 GFLOPS | Progress: (4/20) | 7.88 s
[Task 12/25] Current/Best: 18.53/ 19.06 GFLOPS | Progress: (8/20) | 10.00 s
[Task 12/25] Current/Best: 19.67/ 19.67 GFLOPS | Progress: (12/20) | 13.55 s
[Task 12/25] Current/Best: 16.77/ 19.67 GFLOPS | Progress: (16/20) | 15.35 s
[Task 12/25] Current/Best: 13.00/ 19.67 GFLOPS | Progress: (20/20) | 20.89 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 3.07/ 12.00 GFLOPS | Progress: (4/20) | 4.52 s
[Task 13/25] Current/Best: 12.05/ 19.33 GFLOPS | Progress: (8/20) | 6.52 s
[Task 13/25] Current/Best: 9.06/ 19.97 GFLOPS | Progress: (12/20) | 8.42 s
[Task 13/25] Current/Best: 9.42/ 22.29 GFLOPS | Progress: (16/20) | 10.98 s
[Task 13/25] Current/Best: 11.91/ 22.29 GFLOPS | Progress: (20/20) | 14.27 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 9.53/ 15.58 GFLOPS | Progress: (4/20) | 4.30 s
[Task 14/25] Current/Best: 8.75/ 21.20 GFLOPS | Progress: (8/20) | 7.89 s
[Task 14/25] Current/Best: 9.37/ 21.20 GFLOPS | Progress: (12/20) | 10.31 s Done.
+
[Task 14/25] Current/Best: 17.51/ 21.20 GFLOPS | Progress: (16/20) | 12.91 s
[Task 14/25] Current/Best: 17.49/ 21.20 GFLOPS | Progress: (20/20) | 16.35 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 17.97/ 17.97 GFLOPS | Progress: (4/20) | 5.06 s
[Task 15/25] Current/Best: 18.01/ 18.01 GFLOPS | Progress: (8/20) | 6.58 s
[Task 15/25] Current/Best: 15.81/ 20.90 GFLOPS | Progress: (12/20) | 8.05 s
[Task 15/25] Current/Best: 13.99/ 21.68 GFLOPS | Progress: (16/20) | 14.55 s
[Task 15/25] Current/Best: 12.59/ 21.68 GFLOPS | Progress: (20/20) | 16.26 s Done.
+
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 9.49/ 19.70 GFLOPS | Progress: (4/20) | 3.69 s
[Task 16/25] Current/Best: 11.00/ 19.70 GFLOPS | Progress: (8/20) | 6.63 s
[Task 16/25] Current/Best: 14.59/ 19.70 GFLOPS | Progress: (12/20) | 8.28 s
[Task 16/25] Current/Best: 17.41/ 19.70 GFLOPS | Progress: (16/20) | 9.67 s
[Task 16/25] Current/Best: 13.60/ 19.70 GFLOPS | Progress: (20/20) | 11.65 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 13.19/ 13.19 GFLOPS | Progress: (4/20) | 4.87 s
[Task 17/25] Current/Best: 13.07/ 13.19 GFLOPS | Progress: (8/20) | 8.25 s
[Task 17/25] Current/Best: 3.09/ 17.44 GFLOPS | Progress: (12/20) | 10.98 s
[Task 17/25] Current/Best: 21.83/ 21.83 GFLOPS | Progress: (16/20) | 13.84 s
[Task 17/25] Current/Best: 17.90/ 21.83 GFLOPS | Progress: (20/20) | 16.06 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 11.29/ 16.18 GFLOPS | Progress: (4/20) | 4.98 s
[Task 18/25] Current/Best: 21.33/ 21.33 GFLOPS | Progress: (8/20) | 6.83 s
[Task 18/25] Current/Best: 9.21/ 21.33 GFLOPS | Progress: (12/20) | 9.28 s
[Task 18/25] Current/Best: 9.57/ 21.33 GFLOPS | Progress: (16/20) | 11.37 s
[Task 18/25] Current/Best: 11.73/ 21.33 GFLOPS | Progress: (20/20) | 15.06 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 14.50/ 18.93 GFLOPS | Progress: (4/20) | 5.12 s
[Task 19/25] Current/Best: 14.43/ 18.93 GFLOPS | Progress: (8/20) | 7.92 s
[Task 19/25] Current/Best: 20.49/ 20.49 GFLOPS | Progress: (12/20) | 11.11 s
[Task 19/25] Current/Best: 19.61/ 20.49 GFLOPS | Progress: (16/20) | 17.65 s
[Task 19/25] Current/Best: 3.09/ 20.49 GFLOPS | Progress: (20/20) | 20.45 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 15.80/ 18.39 GFLOPS | Progress: (4/20) | 3.79 s
[Task 20/25] Current/Best: 0.00/ 18.39 GFLOPS | Progress: (8/20) | 4.60 s
[Task 20/25] Current/Best: 16.11/ 18.39 GFLOPS | Progress: (12/20) | 7.32 s
[Task 20/25] Current/Best: 17.63/ 18.39 GFLOPS | Progress: (16/20) | 10.09 s
[Task 20/25] Current/Best: 2.70/ 18.39 GFLOPS | Progress: (20/20) | 14.02 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 4.56/ 11.90 GFLOPS | Progress: (4/20) | 3.46 s Done.
+
[Task 21/25] Current/Best: 21.21/ 21.21 GFLOPS | Progress: (8/20) | 5.02 s
[Task 21/25] Current/Best: 12.90/ 21.21 GFLOPS | Progress: (12/20) | 8.87 s
[Task 21/25] Current/Best: 17.02/ 21.21 GFLOPS | Progress: (16/20) | 10.29 s
[Task 21/25] Current/Best: 17.26/ 21.21 GFLOPS | Progress: (20/20) | 12.27 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 6.42/ 10.49 GFLOPS | Progress: (4/20) | 4.15 s
[Task 22/25] Current/Best: 6.59/ 16.57 GFLOPS | Progress: (8/20) | 9.15 s
[Task 22/25] Current/Best: 17.28/ 17.28 GFLOPS | Progress: (12/20) | 12.40 s
[Task 22/25] Current/Best: 10.96/ 18.16 GFLOPS | Progress: (16/20) | 13.89 s
[Task 22/25] Current/Best: 6.80/ 18.16 GFLOPS | Progress: (20/20) | 15.88 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 9.06/ 20.72 GFLOPS | Progress: (4/20) | 3.74 s
[Task 23/25] Current/Best: 10.55/ 20.72 GFLOPS | Progress: (8/20) | 6.76 s
[Task 23/25] Current/Best: 12.75/ 20.72 GFLOPS | Progress: (12/20) | 9.47 s
[Task 23/25] Current/Best: 13.58/ 21.78 GFLOPS | Progress: (16/20) | 12.02 s
[Task 23/25] Current/Best: 10.59/ 21.78 GFLOPS | Progress: (20/20) | 14.83 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 2.48/ 2.95 GFLOPS | Progress: (4/20) | 12.08 s
[Task 24/25] Current/Best: 5.65/ 8.11 GFLOPS | Progress: (8/20) | 21.03 s
[Task 24/25] Current/Best: 2.46/ 8.11 GFLOPS | Progress: (12/20) | 23.37 s
[Task 24/25] Current/Best: 3.16/ 8.11 GFLOPS | Progress: (16/20) | 33.90 s
[Task 24/25] Current/Best: 5.64/ 8.11 GFLOPS | Progress: (20/20) | 45.44 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
[Task 25/25] Current/Best: 7.00/ 8.76 GFLOPS | Progress: (4/20) | 3.92 s
[Task 25/25] Current/Best: 8.34/ 8.76 GFLOPS | Progress: (8/20) | 14.62 s
[Task 25/25] Current/Best: 9.04/ 9.04 GFLOPS | Progress: (12/20) | 16.82 s
[Task 25/25] Current/Best: 5.16/ 9.04 GFLOPS | Progress: (16/20) | 27.55 s
[Task 25/25] Current/Best: 2.95/ 9.04 GFLOPS | Progress: (20/20) | 29.54 s
@@ -732,8 +730,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 421.7461151999987, 'median': 421.13209345000087, 'std': 2.337065746045658}
- unoptimized: {'mean': 513.548849240002, 'median': 513.8221270500026, 'std': 2.8991952502211213}
+ optimized: {'mean': 430.95549447999474, 'median': 431.23146414998246, 'std': 1.1746246269723992}
+ unoptimized: {'mean': 516.2833439300039, 'median': 516.0584373500114, 'std': 2.27093118676701}
@@ -756,7 +754,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 11 minutes 18.895 seconds)
+ **Total running time of the script:** ( 11 minutes 4.313 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 00194a2578..0bb2f489fe 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -270,7 +270,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.284e-07 secs/op
+ 1.267e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 1e640cfe86..c8f76d4c05 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -263,7 +263,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x73279d0)), stage(b, placeholder(b, 0x236c0cf0)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min [...]
+ [stage(a, placeholder(a, 0x4231460)), stage(b, placeholder(b, 0x486cd40)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min= [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index 8bb9139e40..93e6231958 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,28 +5,28 @@
Computation times
=================
-**14:48.921** total execution time for **tutorial** files:
+**14:55.889** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 11:18.895 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 11:04.313 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:32.489 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:44.723 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:01.127 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 00:59.461 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:35.284 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:36.342 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:18.975 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:29.603 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.248 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.746 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.725 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.530 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.167 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.162 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.007 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.005 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.001 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.002 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index cd66727fd4..acb02ddf94 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -294,7 +294,7 @@ 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
@@ -394,7 +394,7 @@ compile and run this new schedule with the parallel operation applied:
.. code-block:: none
- parallel: 0.000007
+ parallel: 0.000008
@@ -501,10 +501,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 6.916600000295148e-06 1.0
- naive 6.6956e-06 0.9680478847575807
- parallel 6.9471e-06 1.0044096810143062
- vector 2.45552e-05 3.550183616076132
+ numpy 7.806870003150835e-06 1.0
+ naive 6.781800000000001e-06 0.8686964170356214
+ parallel 8.1979e-06 1.0500879349459318
+ vector 2.45417e-05 3.1436030048015433
@@ -925,7 +925,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.018247
+ Numpy running time: 0.018846
@@ -983,7 +983,7 @@ optimizations.
.. code-block:: none
- none: 3.434476
+ none: 3.234691
@@ -1086,7 +1086,7 @@ schedule.
.. code-block:: none
- blocking: 0.296841
+ blocking: 0.330272
@@ -1182,7 +1182,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.333410
+ vectorization: 0.353551
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1256,7 +1256,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.116325
+ loop permutation: 0.129061
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1355,7 +1355,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.108325
+ array packing: 0.110104
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1448,7 +1448,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.110139
+ block caching: 0.111901
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1534,7 +1534,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.146267
+ parallelization: 0.147987
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1615,13 +1615,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.4344755698 1.0
- blocking 0.2968414398 0.08642991739704993
- vectorization 0.3334102469 0.09707748391974018
- loop permutation 0.11632487529999999 0.03386976350126548
- array packing 0.108324767 0.03154040982341536
- block caching 0.11013888280000002 0.03206861733665316
- parallelization 0.1462670396 0.042587881796613734
+ none 3.2346905427 1.0
+ blocking 0.3302720767 0.10210314474914854
+ vectorization 0.35355094140000004 0.10929977280141631
+ loop permutation 0.1290610407 0.03989903794391181
+ array packing 0.11010375449999998 0.03403841976428949
+ block caching 0.1119010235 0.03459404293017659
+ parallelization 0.1479870525 0.0457499876870679
@@ -1661,11 +1661,6 @@ operations with tunable parameters that allows you to automatically optimize
the computation for specific platforms.
-.. rst-class:: sphx-glr-timing
-
- **Total running time of the script:** ( 1 minutes 1.127 seconds)
-
-
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
.. only:: html
diff --git a/docs/commit_hash b/docs/commit_hash
index acd19fde52..99504ef64b 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-697533e136b091b2833e4f6c16298496257e987f
+04afd8311386b6293bda04843738e972df8e93f1
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index fab7052faf..184c6d1b86 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -585,7 +585,7 @@ 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 12.650 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 16.366 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-darknet-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7716f96385bd5abb6e822041e285be54/from_darknet.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_darknet.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_keras.html b/docs/how_to/compile_models/from_keras.html
index 6d50ad265c..71f55e8c3e 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -506,7 +506,7 @@ pip install -U tensorflow --user
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Relay top-1 id: 285, class name: Egyptian cat
1/1 [==============================] - ETA: 0s
-1/1 [==============================] - 1s 993ms/step
+1/1 [==============================] - 1s 1s/step
Keras top-1 id: 285, class name: Egyptian cat
</pre></div>
</div>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 13d109e4f5..bbd1ed29b1 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -440,7 +440,7 @@ to download the full example code</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipd86b083d-67ea-4b7e-8991-31cdae3e67c8 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.zip80b28f56-b89e-45fb-b875-95558f226c3f 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 0872403ca0..9948fdd6b0 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -448,13 +448,13 @@ 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]
- 15%|#5 | 6.33M/41.5M [00:00<00:00, 62.2MB/s]
- 30%|##9 | 12.3M/41.5M [00:00<00:00, 55.5MB/s]
- 42%|####2 | 17.6M/41.5M [00:00<00:00, 33.3MB/s]
- 58%|#####7 | 24.0M/41.5M [00:00<00:00, 33.3MB/s]
- 78%|#######8 | 32.5M/41.5M [00:00<00:00, 46.2MB/s]
- 92%|#########2| 38.3M/41.5M [00:00<00:00, 45.5MB/s]
-100%|##########| 41.5M/41.5M [00:01<00:00, 43.1MB/s]
+ 15%|#5 | 6.33M/41.5M [00:00<00:00, 58.8MB/s]
+ 29%|##8 | 11.9M/41.5M [00:00<00:00, 48.5MB/s]
+ 40%|#### | 16.6M/41.5M [00:00<00:00, 38.7MB/s]
+ 58%|#####7 | 24.0M/41.5M [00:00<00:00, 41.2MB/s]
+ 77%|#######7 | 32.0M/41.5M [00:00<00:00, 49.6MB/s]
+ 96%|#########6| 40.0M/41.5M [00:00<00:00, 57.3MB/s]
+100%|##########| 41.5M/41.5M [00:00<00:00, 52.5MB/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 c4af23e41f..be82957c72 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -431,11 +431,13 @@ be unstable.</p>
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
- 18%|#7 | 7.99M/44.7M [00:00<00:00, 59.7MB/s]
- 36%|###5 | 16.0M/44.7M [00:00<00:00, 69.4MB/s]
- 60%|###### | 27.0M/44.7M [00:00<00:00, 83.0MB/s]
- 85%|########4 | 37.7M/44.7M [00:00<00:00, 93.6MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 68.0MB/s]
+ 18%|#7 | 7.99M/44.7M [00:00<00:00, 60.7MB/s]
+ 36%|###5 | 16.0M/44.7M [00:00<00:00, 65.8MB/s]
+ 54%|#####3 | 24.0M/44.7M [00:00<00:00, 72.6MB/s]
+ 69%|######9 | 31.0M/44.7M [00:00<00:00, 55.4MB/s]
+ 82%|########2 | 36.8M/44.7M [00:00<00:00, 55.2MB/s]
+ 95%|#########4| 42.3M/44.7M [00:00<00:00, 41.7MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 51.8MB/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 63d2ab1168..75e9d29dbf 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -645,7 +645,7 @@ 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 10.056 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 17.756 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 05c0740d6d..db0b6086ca 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -340,7 +340,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:43.811</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>06:05.361</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -348,44 +348,44 @@
<col style="width: 8%" />
</colgroup>
<tbody>
-<tr class="row-odd"><td><p><a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></td>
-<td><p>01:12.650</p></td>
+<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>01:17.756</p></td>
<td><p>0.0 MB</p></td>
</tr>
-<tr class="row-even"><td><p><a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></td>
-<td><p>01:10.056</p></td>
+<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>01:16.366</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>00:45.349</p></td>
+<td><p>00:47.854</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></td>
-<td><p>00:31.904</p></td>
+<td><p>00:33.497</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><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>00:30.106</p></td>
+<td><p>00:30.768</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><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>00:26.072</p></td>
+<td><p>00:27.544</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></td>
-<td><p>00:25.326</p></td>
+<td><p>00:26.256</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>00:22.431</p></td>
+<td><p>00:23.571</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>00:17.510</p></td>
+<td><p>00:19.288</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>00:02.406</p></td>
+<td><p>00:02.462</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 941ac19fe0..84cdabbc1d 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -662,7 +662,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.1361 16.0553 16.8488 15.7097 0.4122
+ 16.3309 16.3534 16.4332 16.1870 0.0770
</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 2f3a98bfed..aac19bf9fb 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -453,25 +453,28 @@ be unstable.</p>
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 5%|4 | 7.99M/170M [00:00<00:02, 69.2MB/s]
- 13%|#2 | 21.9M/170M [00:00<00:01, 111MB/s]
- 19%|#9 | 32.8M/170M [00:00<00:01, 82.6MB/s]
- 25%|##5 | 42.6M/170M [00:00<00:01, 88.9MB/s]
- 33%|###2 | 56.0M/170M [00:00<00:01, 96.1MB/s]
- 42%|####2 | 72.0M/170M [00:00<00:01, 98.5MB/s]
- 48%|####8 | 81.6M/170M [00:00<00:00, 98.1MB/s]
- 54%|#####3 | 91.1M/170M [00:01<00:00, 94.5MB/s]
- 59%|#####8 | 100M/170M [00:01<00:00, 78.3MB/s]
- 64%|######3 | 108M/170M [00:01<00:00, 77.4MB/s]
- 68%|######8 | 116M/170M [00:01<00:00, 73.4MB/s]
- 72%|#######2 | 123M/170M [00:01<00:00, 68.5MB/s]
- 76%|#######6 | 129M/170M [00:01<00:00, 68.0MB/s]
- 80%|######## | 136M/170M [00:01<00:00, 63.5MB/s]
- 86%|########6 | 147M/170M [00:01<00:00, 74.4MB/s]
- 91%|######### | 154M/170M [00:02<00:00, 73.2MB/s]
- 95%|#########4| 161M/170M [00:02<00:00, 58.1MB/s]
- 99%|#########8| 168M/170M [00:02<00:00, 57.9MB/s]
-100%|##########| 170M/170M [00:02<00:00, 75.8MB/s]
+ 5%|4 | 7.99M/170M [00:00<00:02, 68.4MB/s]
+ 9%|9 | 16.0M/170M [00:00<00:02, 63.3MB/s]
+ 14%|#4 | 24.0M/170M [00:00<00:02, 56.1MB/s]
+ 19%|#8 | 32.0M/170M [00:00<00:02, 54.3MB/s]
+ 24%|##3 | 40.0M/170M [00:00<00:02, 51.9MB/s]
+ 28%|##8 | 48.0M/170M [00:00<00:02, 54.6MB/s]
+ 33%|###2 | 56.0M/170M [00:01<00:02, 56.9MB/s]
+ 38%|###7 | 64.0M/170M [00:01<00:01, 56.1MB/s]
+ 42%|####2 | 72.0M/170M [00:01<00:01, 59.3MB/s]
+ 47%|####7 | 80.0M/170M [00:01<00:01, 60.6MB/s]
+ 51%|##### | 86.4M/170M [00:01<00:01, 62.2MB/s]
+ 54%|#####4 | 92.4M/170M [00:01<00:01, 60.0MB/s]
+ 60%|###### | 102M/170M [00:01<00:01, 59.1MB/s]
+ 64%|######3 | 108M/170M [00:01<00:01, 57.5MB/s]
+ 67%|######6 | 113M/170M [00:02<00:01, 48.3MB/s]
+ 71%|####### | 120M/170M [00:02<00:01, 49.7MB/s]
+ 75%|#######5 | 128M/170M [00:02<00:00, 48.7MB/s]
+ 80%|######## | 136M/170M [00:02<00:00, 46.9MB/s]
+ 85%|########4 | 144M/170M [00:02<00:00, 48.7MB/s]
+ 89%|########9 | 152M/170M [00:02<00:00, 48.6MB/s]
+ 94%|#########4| 160M/170M [00:03<00:00, 45.5MB/s]
+100%|##########| 170M/170M [00:03<00:00, 54.8MB/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=& [...]
@@ -569,7 +572,7 @@ 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 13.899 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 34.758 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 09903033c4..7009401698 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -497,8 +497,8 @@ 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]
- 59%|#####8 | 7.99M/13.6M [00:00<00:00, 58.4MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 80.7MB/s]
+ 53%|#####2 | 7.16M/13.6M [00:00<00:00, 75.1MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 74.8MB/s]
</pre></div>
</div>
</div>
@@ -589,7 +589,7 @@ 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)
- 90.5762 90.5239 91.5192 90.2164 0.2400
+ 90.6310 90.5840 91.9719 90.4586 0.1746
</pre></div>
</div>
<div class="admonition note">
@@ -628,7 +628,7 @@ 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 20.607 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 25.775 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 6cb4447f88..7412fe53f1 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -582,7 +582,7 @@ 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)
- 120.2854 120.3054 122.0403 119.5772 0.3622
+ 122.8141 122.6929 130.2332 121.9411 0.8347
</pre></div>
</div>
<div class="admonition note">
@@ -610,7 +610,7 @@ 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 25.230 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 25.010 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 42f1a487da..6244c58ac6 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -520,7 +520,7 @@ 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 32.362 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 37.902 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 b4fabafcb1..c99b088a83 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -462,22 +462,22 @@ to your device.</p>
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
0%| | 0/132723 [00:00<?, ?KB/s]
- 5%|5 | 6841/132723 [00:00<00:01, 68400.12KB/s]
- 12%|#1 | 15300/132723 [00:00<00:01, 77918.57KB/s]
- 17%|#7 | 23092/132723 [00:00<00:01, 57563.24KB/s]
- 24%|##4 | 31879/132723 [00:00<00:01, 67416.62KB/s]
- 29%|##9 | 39124/132723 [00:00<00:01, 66338.25KB/s]
- 36%|###6 | 48010/132723 [00:00<00:01, 73167.37KB/s]
- 43%|####2 | 56842/132723 [00:00<00:00, 77743.30KB/s]
- 49%|####9 | 65664/132723 [00:00<00:00, 80901.50KB/s]
- 56%|#####6 | 74410/132723 [00:00<00:00, 82874.04KB/s]
- 63%|######2 | 83260/132723 [00:01<00:00, 84562.90KB/s]
- 69%|######9 | 92078/132723 [00:01<00:00, 85646.40KB/s]
- 76%|#######6 | 100906/132723 [00:01<00:00, 86432.47KB/s]
- 83%|########2 | 109752/132723 [00:01<00:00, 87038.70KB/s]
- 89%|########9 | 118490/132723 [00:01<00:00, 74534.69KB/s]
- 96%|#########5| 127164/132723 [00:01<00:00, 77804.19KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 77540.07KB/s]
+ 6%|5 | 7479/132723 [00:00<00:01, 74781.33KB/s]
+ 12%|#2 | 16359/132723 [00:00<00:01, 83021.32KB/s]
+ 19%|#9 | 25248/132723 [00:00<00:01, 85698.06KB/s]
+ 26%|##5 | 34082/132723 [00:00<00:01, 86739.32KB/s]
+ 32%|###2 | 42897/132723 [00:00<00:01, 87245.03KB/s]
+ 39%|###8 | 51705/132723 [00:00<00:00, 87523.54KB/s]
+ 46%|####5 | 60659/132723 [00:00<00:00, 88180.13KB/s]
+ 52%|#####2 | 69612/132723 [00:00<00:00, 88608.31KB/s]
+ 59%|#####9 | 78476/132723 [00:00<00:00, 88613.62KB/s]
+ 66%|######5 | 87355/132723 [00:01<00:00, 88665.42KB/s]
+ 73%|#######2 | 96361/132723 [00:01<00:00, 89090.47KB/s]
+ 79%|#######9 | 105316/132723 [00:01<00:00, 89226.92KB/s]
+ 86%|########6 | 114282/132723 [00:01<00:00, 89355.68KB/s]
+ 93%|#########2| 123218/132723 [00:01<00:00, 89293.80KB/s]
+100%|#########9| 132210/132723 [00:01<00:00, 89481.00KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 87982.81KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -516,7 +516,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 56.028 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 8.283 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 572785f453..156ff7b46c 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -340,7 +340,7 @@
<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>12:52.353</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>13:40.029</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -349,39 +349,39 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>03:13.899</p></td>
+<td><p>03:34.758</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>02:56.028</p></td>
+<td><p>03:08.283</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>02:25.230</p></td>
+<td><p>02:25.010</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>01:32.362</p></td>
+<td><p>01:37.902</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>01:20.607</p></td>
+<td><p>01:25.775</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>00:35.284</p></td>
+<td><p>00:37.427</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>00:24.776</p></td>
+<td><p>00:25.776</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>00:24.160</p></td>
+<td><p>00:25.091</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>00:00.006</p></td>
+<td><p>00:00.007</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index 002cfd235a..c6f98f5baa 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -621,7 +621,7 @@ 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.zip247ccd88-0f24-48dd-b94c-c3937117ab31 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.zip02aff155-db9d-442b-b1a1-afafd93f7f13 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 e623a72157..3c6da4732d 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -340,7 +340,7 @@
<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>00:45.424</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:49.319</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -349,19 +349,19 @@
</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>00:42.079</p></td>
+<td><p>00:45.664</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>00:02.342</p></td>
+<td><p>00:02.546</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></td>
-<td><p>00:00.994</p></td>
+<td><p>00:01.100</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
-<td><p>00:00.008</p></td>
+<td><p>00:00.009</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index 34b590617b..d4c2de4e01 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -525,10 +525,10 @@ 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: 6528us [6528us] (45.73%; 45.73%)
-FoldScaleAxis: 7748us [6us] (54.27%; 54.27%)
- FoldConstant: 7743us [1633us] (54.23%; 99.93%)
- InferType: 6110us [6110us] (42.80%; 78.91%)
+InferType: 7339us [7339us] (47.39%; 47.39%)
+FoldScaleAxis: 8147us [6us] (52.61%; 52.61%)
+ FoldConstant: 8141us [1635us] (52.57%; 99.92%)
+ InferType: 6506us [6506us] (42.01%; 79.91%)
</pre></div>
</div>
</div>
@@ -550,10 +550,10 @@ 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: 6192us [6192us] (44.37%; 44.37%)
-FoldScaleAxis: 7764us [5us] (55.63%; 55.63%)
- FoldConstant: 7760us [1569us] (55.60%; 99.94%)
- InferType: 6191us [6191us] (44.36%; 79.78%)
+InferType: 6766us [6766us] (44.59%; 44.59%)
+FoldScaleAxis: 8408us [7us] (55.41%; 55.41%)
+ FoldConstant: 8401us [1697us] (55.37%; 99.92%)
+ InferType: 6705us [6705us] (44.18%; 79.80%)
</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 0049e12ecc..a4fe612d83 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -577,7 +577,7 @@ 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: 48.037086 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 36.324352 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 c400c6af9d..6b25c48d7c 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -916,7 +916,7 @@ 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: 10.191680 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 13.360851 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 0fc8fdd44b..ab7f90932a 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -474,8 +474,8 @@ 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: 0.017844
-Baseline: 3.381535
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019474
+Baseline: 3.304775
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -535,7 +535,7 @@ 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: 0.302799
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.330295
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -602,7 +602,7 @@ 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: 0.334024
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.344460
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -663,7 +663,7 @@ 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: 0.114958
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.140625
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -746,7 +746,7 @@ 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: 0.110128
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.111854
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -832,7 +832,7 @@ 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: 0.111559
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.114512
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -922,7 +922,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.146860
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.150728
</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 af644631e0..8263871e0a 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -340,7 +340,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:34.522</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.841</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -349,15 +349,15 @@
</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>00:32.130</p></td>
+<td><p>00:33.120</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>00:01.336</p></td>
+<td><p>00:01.533</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>00:01.056</p></td>
+<td><p>00:01.188</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 098331bf9a..8c2a3cce21 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -340,7 +340,7 @@
<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>08:15.195</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>08:44.052</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -349,27 +349,27 @@
</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>04:54.702</p></td>
+<td><p>05:08.510</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>01:29.296</p></td>
+<td><p>01:34.419</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>01:01.447</p></td>
+<td><p>01:04.102</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>00:27.538</p></td>
+<td><p>00:33.297</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>00:11.501</p></td>
+<td><p>00:12.230</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>00:10.710</p></td>
+<td><p>00:11.494</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 20c4ef6c0b..2613b13697 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
@@ -504,453 +504,461 @@ cooperative fetching, unrolling and operator fusion.</p>
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 32;
- allocate(conv2d_nchw: Pointer(local float32), float32, [4]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [2016]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [1536]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [1], [], scope="local", align=4)[0] = 0f32
- conv2d_nchw_1[1] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [16]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [392]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [256]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [4], [], scope="local", align=8)[0] = 0f32
conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[6] = 0f32
+ conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[14] = 0f32
+ conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[3] = 0f32
- for (rc.outer.outer: int32, 0, 16) {
- for (rx.outer.outer: int32, 0, 3) {
- let cse_var_2: int32 = (rc.outer.outer*1568)
- let cse_var_1: int32 = (rc.outer.outer*288)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [2016], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[((((cse_var_2 + (floordiv(threadIdx.x_1, 63)*49)) + rx.outer.outer) + floormod(threadIdx.x_1, 63)) - 8)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 1), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1 [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1 [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 588), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1 [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 4), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 4), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 784), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 4), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1 [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 980)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 980), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1 [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1176)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1176), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1372)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1372), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1568)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 8), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 8), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1568), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 8), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x [...]
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- pad_temp.shared_1[(threadIdx.x_1 + 1764)] = @tir.if_then_else(((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[((((cse_var_2 + (floordiv(threadIdx.x_1, 63)*49)) + rx.outer.outer) + floormod(threadIdx.x_1, 63)) + 1364)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- if @tir.likely((threadIdx.x_1 < 56), dtype=bool) {
- pad_temp.shared_1[(threadIdx.x_1 + 1960)] = @tir.if_then_else((((threadIdx.x_1 < 49) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 1960), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
- }
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1: Buffer(kernel.shared, float32, [1536], [], scope="shared")[threadIdx.x_2] = kernel[(((((blockIdx.x*73728) + (floordiv(threadIdx.x_2, 96)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 96)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 196), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 4), 96), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 392), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 96), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 588)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 588), 96)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 4), 32)*9)) + (floormod(threadIdx.x_2, 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 784), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 96), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 980)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 980), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 20), 96), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 1176), 96)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 32)*9)) + (floormod(threadIdx.x_2, 3)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
- if @tir.likely((threadIdx.x_2 < 164), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 1372)] = kernel[((((((blockIdx.x*73728) + (floordiv((threadIdx.x_2 + 1372), 96)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 28), 96), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + rx.outer.outer)]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*96)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 384)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 768)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1152)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 385)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 769)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1153)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 386)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 770)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1154)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 387)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 771)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1155)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 388)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 772)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1156)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 389)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 773)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1157)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 390)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 774)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1158)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 391)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 775)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1159)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 392)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 776)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1160)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 393)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 777)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1161)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 394)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 778)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1162)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 395)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 779)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1163)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 396)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 780)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1164)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 397)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 781)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1165)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 398)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 782)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1166)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 399)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 783)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1167)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 400)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 784)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1168)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 401)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 785)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1169)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 402)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 786)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1170)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 403)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 787)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1171)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 404)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 788)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1172)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 405)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 789)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1173)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 406)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 790)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1174)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 407)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 791)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1175)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 24)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 408)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 792)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 504)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1176)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 25)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 409)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 793)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 511)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1177)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 26)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 410)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 794)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 518)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1178)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 27)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 411)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 795)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 567)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1179)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 28)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 412)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 796)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 574)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1180)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 29)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 413)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 797)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 581)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1181)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 30)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 414)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 798)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 630)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1182)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 31)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 415)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 799)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 637)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1183)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 32)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 416)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 800)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 644)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1184)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 33)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 417)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 801)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 693)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1185)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 34)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 418)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 802)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 700)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1186)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 35)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 419)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 803)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 707)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1187)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 420)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 804)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 756)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1188)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 37)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 421)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 805)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 763)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1189)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 38)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 422)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 806)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 770)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1190)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 39)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 423)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 807)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 819)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1191)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 40)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 424)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 808)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 826)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1192)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 41)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 425)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 809)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 833)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1193)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 42)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 426)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 810)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 882)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1194)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 43)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 427)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 811)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 889)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1195)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 44)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 428)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 812)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 896)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1196)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 45)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 429)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 813)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 945)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1197)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 46)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 430)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 814)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 952)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1198)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 47)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 431)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 815)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 959)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1199)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 432)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 816)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1008)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1200)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 49)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 433)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 817)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1015)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1201)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 50)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 434)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 818)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1022)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1202)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 51)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 435)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 819)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1071)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1203)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 52)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 436)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 820)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1078)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1204)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 53)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 437)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 821)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1085)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1205)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 54)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 438)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 822)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1134)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1206)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 55)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 439)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 823)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1141)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1207)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 56)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 440)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 824)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1148)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1208)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 57)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 441)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 825)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1197)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1209)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 58)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 442)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 826)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1204)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1210)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 59)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 443)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 827)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1211)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1211)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 60)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 444)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 828)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1260)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1212)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 61)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 445)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 829)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1267)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1213)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 62)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 446)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 830)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1274)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1214)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 63)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 447)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 831)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1323)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1215)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 64)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 448)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 832)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1330)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1216)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 65)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 449)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 833)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1337)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1217)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 66)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 450)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 834)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1386)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1218)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 67)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 451)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 835)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1393)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1219)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 68)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 452)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 836)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1400)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1220)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 69)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 453)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 837)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1449)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1221)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 70)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 454)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 838)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1456)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1222)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 71)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 455)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 839)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1463)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1223)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 72)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 456)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 840)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1512)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1224)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 73)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 457)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 841)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1519)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1225)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 74)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 458)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 842)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1526)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1226)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 75)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 459)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 843)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1575)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1227)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 76)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 460)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 844)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1582)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1228)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 77)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 461)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 845)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1589)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1229)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 78)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 462)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 846)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1638)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1230)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 79)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 463)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 847)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1645)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1231)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 80)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 464)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 848)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1652)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1232)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 81)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 465)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 849)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1701)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1233)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 82)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 466)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 850)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1708)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1234)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 83)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 467)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 851)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1715)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1235)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 84)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 468)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 852)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1764)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1236)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 85)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 469)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 853)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1771)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1237)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 86)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 470)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 854)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1778)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1238)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 87)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 471)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 855)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1827)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1239)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 88)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 472)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 856)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1834)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1240)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 89)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 473)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 857)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1841)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1241)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 90)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 474)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 858)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1890)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1242)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 91)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 475)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 859)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1897)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1243)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 92)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 476)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 860)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1904)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1244)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 93)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 477)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 861)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1953)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1245)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 94)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 478)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 862)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1960)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1246)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 95)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 479)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 863)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 1967)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*96) + 1247)]))
+ conv2d_nchw_1[5] = 0f32
+ conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[13] = 0f32
+ conv2d_nchw_1[15] = 0f32
+ for (rc.outer.outer: int32, 0, 64) {
+ for (ry.outer.outer: int32, 0, 3) {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [392], [], scope="shared")[(threadIdx.x_1*2)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1*2), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 8)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 1)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*2) + 1), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 7)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[((threadIdx.x_1*2) + 196)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1*2), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 188)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 197)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (1 <= floormod(((threadIdx.x_1*2) + 1), 7))), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 189)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
+ kernel.shared_1: Buffer(kernel.shared, float32, [256], [], scope="shared")[ramp((threadIdx.x_2*4), 1, 4)] = kernel[ramp((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 2)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 2)*36)) + (ry.outer.outer*3)), 9, 4)]
+ }
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*16)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 32)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 64)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 96)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 128)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 160)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 192)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 224)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 1)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 33)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 65)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 97)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 129)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 161)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 193)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 225)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 8)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 40)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 72)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 104)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 136)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 200)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 232)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 9)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 41)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 73)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 105)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 137)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 201)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 233)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 2)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 34)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 66)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 98)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 130)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 162)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 194)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 226)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 3)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 35)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 67)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 99)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 131)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 163)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 195)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 227)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 10)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 42)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 74)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 106)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 138)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 202)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 234)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 11)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 43)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 75)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 107)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 139)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 203)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 235)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 4)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 36)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 68)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 100)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 132)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 164)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 196)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 228)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 5)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 37)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 69)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 101)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 133)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 165)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 197)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 229)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 12)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 44)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 76)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 108)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 140)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 204)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 236)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 13)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 45)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 77)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 109)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 141)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 205)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 237)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 6)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 38)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 70)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 102)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 134)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 166)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 198)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 230)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 7)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 39)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 71)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 103)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 135)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 167)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 199)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 231)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 14)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 46)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 78)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 110)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 142)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 206)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 238)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 15)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 47)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 79)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 111)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 143)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 207)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 239)]))
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[(threadIdx.x_1*2)] = @tir.if_then_else(((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 7)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 1)] = @tir.if_then_else(((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 6)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[((threadIdx.x_1*2) + 196)] = @tir.if_then_else(((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 189)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 197)] = @tir.if_then_else(((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 190)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
+ kernel.shared_1[ramp((threadIdx.x_2*4), 1, 4)] = kernel[ramp(((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 2)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 2)*36)) + (ry.outer.outer*3)) + 1), 9, 4)]
+ }
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*16)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 32)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 64)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 96)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 128)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 160)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 192)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 224)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 1)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 33)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 65)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 97)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 129)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 161)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 193)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 225)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 8)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 40)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 72)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 104)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 136)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 200)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 232)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 9)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 41)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 73)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 105)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 137)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 201)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 233)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 2)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 34)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 66)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 98)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 130)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 162)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 194)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 226)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 3)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 35)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 67)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 99)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 131)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 163)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 195)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 227)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 10)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 42)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 74)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 106)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 138)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 202)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 234)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 11)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 43)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 75)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 107)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 139)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 203)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 235)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 4)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 36)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 68)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 100)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 132)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 164)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 196)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 228)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 5)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 37)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 69)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 101)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 133)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 165)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 197)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 229)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 12)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 44)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 76)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 108)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 140)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 204)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 236)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 13)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 45)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 77)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 109)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 141)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 205)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 237)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 6)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 38)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 70)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 102)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 134)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 166)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 198)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 230)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 7)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 39)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 71)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 103)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 135)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 167)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 199)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 231)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 14)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 46)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 78)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 110)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 142)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 206)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 238)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 15)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 47)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 79)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 111)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 143)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 207)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 239)]))
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[(threadIdx.x_1*2)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (floormod((threadIdx.x_1*2), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 6)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 1)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (floormod(((threadIdx.x_1*2) + 1), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) - 5)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98 {
+ pad_temp.shared_1[((threadIdx.x_1*2) + 196)] = @tir.if_then_else((((1 <= (floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1*2), 49), 7) + ry.outer.outer) < 8)) && (floormod((threadIdx.x_1*2), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 190)], 0f32, dtype=float32)
+ pad_temp.shared_1[((threadIdx.x_1*2) + 197)] = @tir.if_then_else((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer)) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 49), 7) + ry.outer.outer) < 8)) && (floormod(((threadIdx.x_1*2) + 1), 7) < 6)), data[((((rc.outer.outer*392) + (ry.outer.outer*7)) + (threadIdx.x_1*2)) + 191)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 98;
+ if @tir.likely((threadIdx.x_2 < 64), dtype=bool) {
+ kernel.shared_1[ramp((threadIdx.x_2*4), 1, 4)] = kernel[ramp(((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 2)*4608)) + (rc.outer.outer*72)) + (floormod(threadIdx.x_2, 2)*36)) + (ry.outer.outer*3)) + 2), 9, 4)]
}
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*16)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 32)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 64)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 96)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 128)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 160)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 192)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 224)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 1)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 33)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 65)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 97)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 129)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 161)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 193)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 225)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 8)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 40)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 72)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 104)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 136)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 168)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 200)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 232)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 9)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 41)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 73)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 105)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 137)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 169)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 201)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 233)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 2)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 34)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 66)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 98)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 130)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 162)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 194)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 226)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 3)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 35)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 67)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 99)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 131)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 163)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 195)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 227)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 10)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 42)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 74)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 106)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 138)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 170)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 202)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 234)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 11)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 43)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 75)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 107)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 139)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 171)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 203)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 235)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 4)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 36)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 68)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 100)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 132)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 164)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 196)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 228)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 5)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 37)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 69)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 101)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 133)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 165)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 197)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 229)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 12)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 44)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 76)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 108)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 140)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 172)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 204)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 236)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 13)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 45)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 77)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 109)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 141)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 173)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 205)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 237)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 6)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 38)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 70)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 102)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 134)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 166)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 198)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 230)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 7)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 39)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 71)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 103)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 135)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 167)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 199)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 231)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 14)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 46)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 78)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 110)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 142)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 174)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 206)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 238)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 15)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 47)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 79)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 111)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 143)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 175)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 207)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*16) + 239)]))
}
}
- compute[((blockIdx.x*784) + threadIdx.x)] = max((conv2d_nchw_1[0] + bias[((blockIdx.x*16) + floordiv(threadIdx.x, 49))]), 0f32)
- compute[(((blockIdx.x*784) + threadIdx.x) + 196)] = max((conv2d_nchw_1[1] + bias[(((blockIdx.x*16) + floordiv(threadIdx.x, 49)) + 4)]), 0f32)
- compute[(((blockIdx.x*784) + threadIdx.x) + 392)] = max((conv2d_nchw_1[2] + bias[(((blockIdx.x*16) + floordiv(threadIdx.x, 49)) + 8)]), 0f32)
- compute[(((blockIdx.x*784) + threadIdx.x) + 588)] = max((conv2d_nchw_1[3] + bias[(((blockIdx.x*16) + floordiv(threadIdx.x, 49)) + 12)]), 0f32)
+ for (i1.inner: int32, 0, 2) {
+ compute[((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 196)] = max((conv2d_nchw_1[(i1.inner + 2)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 4)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 392)] = max((conv2d_nchw_1[(i1.inner + 4)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 8)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 588)] = max((conv2d_nchw_1[(i1.inner + 6)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 12)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 784)] = max((conv2d_nchw_1[(i1.inner + 8)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 16)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 980)] = max((conv2d_nchw_1[(i1.inner + 10)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 20)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 1176)] = max((conv2d_nchw_1[(i1.inner + 12)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 24)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*98)) + (i1.inner*49)) + floormod(threadIdx.x, 49)) + 1372)] = max((conv2d_nchw_1[(i1.inner + 14)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*2)) + i1.inner) + 28)]), 0f32)
+ }
}
}
</pre></div>
@@ -986,7 +994,7 @@ cooperative fetching, unrolling and operator fusion.</p>
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.272 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.333 ms
</pre></div>
</div>
</div>
@@ -1016,9 +1024,9 @@ conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
-conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=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=4)
-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=4)
+conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=2)
+conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=8)
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)
@@ -1027,9 +1035,9 @@ conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, fact
conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=32)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=1)
-conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_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=1)
@@ -1037,9 +1045,9 @@ s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nc
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
-compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=4)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=2)
+compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=8)
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)
@@ -1062,16 +1070,16 @@ 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=1)
+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=4)
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=196)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
+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=2)
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=196)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=98)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
-s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -1089,431 +1097,454 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[4];
- __shared__ float pad_temp_shared[2016];
- __shared__ float kernel_shared[1536];
+extern "C" __global__ void __launch_bounds__(98) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[16];
+ __shared__ float pad_temp_shared[392];
+ __shared__ float kernel_shared[256];
conv2d_nchw[0] = 0.000000e+00f;
- conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
+ conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
- for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
+ conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[13] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
+ for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
__syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = (((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 63) * 49)) + rx_outer_outer) + (((int)threadIdx.x) % 63)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((1 <= (((((int)threadIdx.x) / 7) + 1) % 9)) && ((((((int)threadIdx.x) / 7) + 1) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 1) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 588)] = (((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 588) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((1 <= (((((int)threadIdx.x) / 7) + 4) % 9)) && ((((((int)threadIdx.x) / 7) + 4) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 784) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 4) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 980)] = (((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 980) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1176) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1372)] = (((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1372) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1568)] = (((((1 <= (((((int)threadIdx.x) / 7) + 8) % 9)) && ((((((int)threadIdx.x) / 7) + 8) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1568) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 8) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1764)] = (((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 63) * 49)) + rx_outer_outer) + (((int)threadIdx.x) % 63)) + 1364)] : 0.000000e+00f);
- if (((int)threadIdx.x) < 56) {
- pad_temp_shared[(((int)threadIdx.x) + 1960)] = ((((((int)threadIdx.x) < 49) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 1960) / 63) * 49)) + ((int)threadIdx.x)) + rx_outer_outer) - 1)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) * 2)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) * 2) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 1)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 2) + 1) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 196)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) * 2) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 188)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 197)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && (1 <= (((((int)threadIdx.x) * 2) + 1) % 7))) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 189)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 64) {
+ int4 __1 = make_int4(((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)))+(9*0), ((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)))+(9*1), ((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)th [...]
+ *(float4*)(kernel_shared + (((int)threadIdx.x) * 4)) = make_float4(kernel[__1.x],kernel[__1.y],kernel[__1.z],kernel[__1.w]);
}
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) % 96) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 196) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 4) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 392)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 392) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 8) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 2) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 588)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 588) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 4) & 31) * 9)) + ((((int)threadIdx.x) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 16) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 980)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 980) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 20) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 2) % 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1176) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 8) & 31) * 9)) + ((((int)threadIdx.x) % 3) * 3)) + rx_outer_outer)];
- if (((int)threadIdx.x) < 164) {
- kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1372) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 28) % 96) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + rx_outer_outer)];
+ __syncthreads();
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 16)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 32)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 64)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 96)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 128)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 192)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 224)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 33)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 65)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 97)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 129)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 193)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 225)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 8)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 40)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 72)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 104)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 136)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 200)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 232)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 9)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 41)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 73)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 105)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 137)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 201)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 233)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 2)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 34)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 66)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 98)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 130)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 194)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 226)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 3)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 35)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 67)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 99)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 131)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 195)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 227)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 10)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 42)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 74)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 106)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 138)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 202)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 234)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 11)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 43)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 75)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 107)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 139)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 203)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 235)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 4)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 36)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 68)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 100)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 132)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 196)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 228)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 5)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 37)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 69)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 101)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 133)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 197)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 229)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 12)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 44)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 76)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 108)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 140)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 204)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 236)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 13)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 45)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 77)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 109)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 141)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 205)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 237)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 6)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 38)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 70)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 102)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 134)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 198)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 230)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 7)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 39)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 71)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 103)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 135)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 199)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 231)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 14)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 46)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 78)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 110)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 142)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 206)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 238)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 15)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 47)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 79)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 111)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 143)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 207)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 239)]));
+ __syncthreads();
+ pad_temp_shared[(((int)threadIdx.x) * 2)] = (((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 7)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 1)] = (((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 196)] = (((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 189)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 197)] = (((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 190)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 64) {
+ int4 __2 = make_int4((((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 1))+(9*0), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 1))+(9*1), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) [...]
+ *(float4*)(kernel_shared + (((int)threadIdx.x) * 4)) = make_float4(kernel[__2.x],kernel[__2.y],kernel[__2.z],kernel[__2.w]);
}
__syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 96)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 384)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 768)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1152)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 385)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 769)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 7)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1153)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 386)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 770)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 14)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1154)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 387)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 771)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 63)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1155)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 388)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 772)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 70)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1156)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 389)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 773)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 77)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1157)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 390)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 774)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 126)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1158)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 391)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 775)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 133)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1159)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 392)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 776)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 140)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1160)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 393)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 777)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 189)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1161)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 394)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 778)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1162)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 395)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 779)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 203)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1163)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 396)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 780)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 252)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1164)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 397)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 781)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 259)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1165)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 398)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 782)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 266)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1166)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 399)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 783)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 315)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1167)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 400)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 784)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 322)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1168)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 401)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 785)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 329)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1169)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 402)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 786)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 378)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1170)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 403)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 787)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 385)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1171)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 404)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 788)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 392)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1172)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 405)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 789)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 441)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1173)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 406)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 790)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 448)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1174)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 407)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 791)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 455)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1175)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 408)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 792)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 504)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1176)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 409)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 793)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 511)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1177)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 410)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 794)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 518)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1178)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 411)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 795)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 567)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1179)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 412)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 796)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 574)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1180)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 413)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 797)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 581)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1181)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 414)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 798)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 630)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1182)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 415)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 799)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 637)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1183)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 416)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 800)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 644)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1184)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 417)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 801)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 693)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1185)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 418)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 802)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 700)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1186)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 419)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 803)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 707)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1187)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 420)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 804)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 756)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1188)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 421)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 805)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 763)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1189)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 422)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 806)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 770)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1190)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 423)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 807)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 819)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1191)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 424)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 808)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 826)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1192)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 425)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 809)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 833)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1193)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 426)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 810)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 882)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1194)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 427)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 811)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 889)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1195)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 428)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 812)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 896)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1196)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 429)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 813)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 945)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1197)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 430)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 814)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 952)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1198)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 431)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 815)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 959)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1199)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 432)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 816)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1008)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1200)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 433)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 817)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1015)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1201)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 434)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 818)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1022)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1202)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 51)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 435)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 819)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1071)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1203)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 52)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 436)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 820)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1078)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1204)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 53)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 437)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 821)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1085)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1205)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 54)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 438)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 822)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1134)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1206)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 55)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 439)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 823)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1141)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1207)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 56)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 440)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 824)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1148)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1208)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 57)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 441)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 825)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1197)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1209)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 58)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 442)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 826)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1204)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1210)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 59)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 443)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 827)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1211)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1211)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 60)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 444)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 828)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1260)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1212)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 61)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 445)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 829)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1267)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1213)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 62)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 446)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 830)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1274)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1214)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 63)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 447)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 831)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1323)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1215)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 64)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 448)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 832)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1330)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1216)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 65)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 449)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 833)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1337)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1217)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 66)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 450)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 834)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1386)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1218)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 67)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 451)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 835)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1393)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1219)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 68)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 452)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 836)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1400)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1220)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 69)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 453)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 837)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1449)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1221)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 70)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 454)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 838)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1456)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1222)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 71)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 455)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 839)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1463)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1223)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 72)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 456)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 840)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1512)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1224)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 73)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 457)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 841)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1519)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1225)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 74)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 458)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 842)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1526)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1226)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 75)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 459)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 843)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1575)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1227)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 76)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 460)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 844)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1582)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1228)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 77)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 461)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 845)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1589)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1229)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 78)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 462)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 846)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1638)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1230)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 79)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 463)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 847)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1645)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1231)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 80)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 464)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 848)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1652)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1232)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 81)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 465)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 849)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1701)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1233)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 82)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 466)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 850)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1708)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1234)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 83)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 467)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 851)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1715)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1235)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 84)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 468)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 852)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1764)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1236)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 85)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 469)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 853)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1771)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1237)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 86)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 470)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 854)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1778)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1238)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 87)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 471)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 855)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1827)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1239)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 88)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 472)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 856)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1834)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1240)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 89)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 473)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 857)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1841)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1241)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 90)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 474)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 858)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1890)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1242)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 91)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 475)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 859)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1897)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1243)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 92)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 476)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 860)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1904)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1244)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 93)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 477)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 861)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1953)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1245)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 94)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 478)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 862)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1960)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1246)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 95)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 479)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 863)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 1967)] * kernel_shared[(((((int)threadIdx.x) / 49) * 96) + 1247)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 16)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 32)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 64)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 96)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 128)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 192)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 224)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 33)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 65)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 97)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 129)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 193)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 225)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 8)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 40)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 72)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 104)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 136)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 200)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 232)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 9)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 41)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 73)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 105)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 137)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 201)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 233)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 2)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 34)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 66)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 98)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 130)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 194)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 226)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 3)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 35)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 67)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 99)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 131)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 195)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 227)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 10)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 42)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 74)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 106)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 138)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 202)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 234)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 11)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 43)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 75)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 107)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 139)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 203)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 235)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 4)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 36)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 68)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 100)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 132)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 196)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 228)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 5)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 37)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 69)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 101)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 133)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 197)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 229)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 12)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 44)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 76)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 108)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 140)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 204)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 236)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 13)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 45)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 77)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 109)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 141)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 205)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 237)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 6)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 38)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 70)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 102)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 134)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 198)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 230)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 7)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 39)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 71)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 103)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 135)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 199)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 231)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 14)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 46)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 78)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 110)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 142)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 206)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 238)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 15)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 47)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 79)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 111)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 143)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 207)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 239)]));
+ __syncthreads();
+ pad_temp_shared[(((int)threadIdx.x) * 2)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (((((int)threadIdx.x) * 2) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 6)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 1)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && ((((((int)threadIdx.x) * 2) + 1) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) - 5)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 196)] = ((((1 <= ((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer)) && (((((((int)threadIdx.x) * 2) % 49) / 7) + ry_outer_outer) < 8)) && (((((int)threadIdx.x) * 2) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 190)] : 0.000000e+00f);
+ pad_temp_shared[((((int)threadIdx.x) * 2) + 197)] = ((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer)) && ((((((((int)threadIdx.x) * 2) + 1) % 49) / 7) + ry_outer_outer) < 8)) && ((((((int)threadIdx.x) * 2) + 1) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (ry_outer_outer * 7)) + (((int)threadIdx.x) * 2)) + 191)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 64) {
+ int4 __3 = make_int4((((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 2))+(9*0), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) & 1) * 36)) + (ry_outer_outer * 3)) + 2))+(9*1), (((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) >> 1) * 4608)) + (rc_outer_outer * 72)) [...]
+ *(float4*)(kernel_shared + (((int)threadIdx.x) * 4)) = make_float4(kernel[__3.x],kernel[__3.y],kernel[__3.z],kernel[__3.w]);
+ }
+ __syncthreads();
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 16)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 32)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 64)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 96)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 128)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 160)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 192)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 224)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 1)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 33)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 65)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 97)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 129)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 161)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 193)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 225)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 8)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 40)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 72)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 104)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 136)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 168)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 200)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 232)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 9)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 41)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 73)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 105)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 137)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 169)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 201)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 233)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 2)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 34)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 66)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 98)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 130)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 162)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 194)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 226)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 3)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 35)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 67)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 99)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 131)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 163)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 195)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 227)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 10)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 42)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 74)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 106)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 138)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 170)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 202)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 234)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 11)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 43)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 75)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 107)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 139)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 171)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 203)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 235)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 4)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 36)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 68)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 100)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 132)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 164)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 196)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 228)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 5)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 37)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 69)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 101)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 133)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 165)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 197)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 229)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 12)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 44)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 76)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 108)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 140)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 172)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 204)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 196)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 236)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 13)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 45)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 77)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 109)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 141)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 173)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 205)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 245)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 237)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 6)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 38)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 70)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 102)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 134)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 166)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 198)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 230)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 7)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 39)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 71)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 103)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 135)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 167)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 199)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 231)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 14)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 46)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 78)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 110)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 142)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 174)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 206)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 294)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 238)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 15)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 47)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 79)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 111)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 143)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 175)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 207)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 343)] * kernel_shared[(((((int)threadIdx.x) / 49) * 16) + 239)]));
}
}
- compute[((((int)blockIdx.x) * 784) + ((int)threadIdx.x))] = max((conv2d_nchw[0] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49))]), 0.000000e+00f);
- compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[1] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 4)]), 0.000000e+00f);
- compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 392)] = max((conv2d_nchw[2] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 8)]), 0.000000e+00f);
- compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 588)] = max((conv2d_nchw[3] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 12)]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 196)] = max((conv2d_nchw[(i1_inner + 2)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 4)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 392)] = max((conv2d_nchw[(i1_inner + 4)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 8)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 588)] = max((conv2d_nchw[(i1_inner + 6)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 12)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 784)] = max((conv2d_nchw[(i1_inner + 8)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 980)] = max((conv2d_nchw[(i1_inner + 10)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 20)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 1176)] = max((conv2d_nchw[(i1_inner + 12)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 24)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 98)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49)) + 1372)] = max((conv2d_nchw[(i1_inner + 14)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 2)) + i1_inner) + 28)]), 0.000000e+00f);
+ }
}
</pre></div>
</div>
@@ -1549,7 +1580,7 @@ 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> ( 4 minutes 54.702 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes 8.510 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 dc2fde909f..04b9bf22de 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -915,7 +915,7 @@ 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)
- 8.1885 8.1875 8.1910 8.1871 0.0017
+ 8.1871 8.1868 8.1918 8.1828 0.0037
</pre></div>
</div>
</div>
@@ -937,7 +937,7 @@ 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 1.447 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 4.102 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 59d3304f36..61d1a93a9f 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -934,7 +934,7 @@ 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)
- 743.2654 743.3260 744.4337 742.0365 0.9796
+ 766.8664 766.8750 767.3471 766.3770 0.3961
</pre></div>
</div>
</div>
@@ -956,7 +956,7 @@ 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 29.296 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 34.419 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 7cf40e665d..8bf10dca30 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -632,28 +632,104 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 128) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [512]), storage_scope = global {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 16) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [512], [])[(((i.inner.init*32) + (nb_j.inner*16)) + j.init)] = 0f32
- }
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_9: placeholder_16: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 512) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [128]), storage_scope = global {
+ for (i.inner.init: int32, 0, 8) {
+ let cse_var_1: int32 = (i.inner.init*16)
+ {
+ compute_5: Buffer(compute_4, float32, [128], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (i.inner: int32, 0, 16) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- let cse_var_2: int32 = (((i.inner*32) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 16)*4096) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ for (elem_idx: int32, 0, let cse_var_2: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_2 + 1)] - placeholder_3[cse_var_2])) {
+ for (i.inner: int32, 0, 8) {
+ let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ {
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_4: int32 = (i.inner*16)
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[((placeholder_3[cse_var_3]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_5: int32 = ((i.inner*16) + 1)
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_6: int32 = ((i.inner*16) + 2)
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_7: int32 = ((i.inner*16) + 3)
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_8: int32 = ((i.inner*16) + 4)
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_9: int32 = ((i.inner*16) + 5)
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_10: int32 = ((i.inner*16) + 6)
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_11: int32 = ((i.inner*16) + 7)
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_12: int32 = ((i.inner*16) + 8)
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_13: int32 = ((i.inner*16) + 9)
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_14: int32 = ((i.inner*16) + 10)
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_15: int32 = ((i.inner*16) + 11)
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_16: int32 = ((i.inner*16) + 12)
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_17: int32 = ((i.inner*16) + 13)
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_18: int32 = ((i.inner*16) + 14)
+ compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_3 + 1)] - placeholder_3[cse_var_3])), dtype=bool) {
+ let cse_var_19: int32 = ((i.inner*16) + 15)
+ compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*2048) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 16) {
- let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 8) {
+ let cse_var_20: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*4096) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_20, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_20, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -691,7 +767,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.508 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.888 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 6c196ed020..ba7e994dca 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -340,7 +340,7 @@
<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>00:34.812</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:37.172</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -349,11 +349,11 @@
</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>00:34.778</p></td>
+<td><p>00:37.135</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></td>
-<td><p>00:00.019</p></td>
+<td><p>00:00.021</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></td>
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 c242b74ca7..bc58bf3f5b 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -689,7 +689,7 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8046651
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7876715
No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -812,7 +812,7 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2799895
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9549576
No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -935,9 +935,9 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6746199
-No: 4 GFLOPS: 123.68/123.68 result: MeasureResult(costs=(0.001871792546875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.121532678604126, timestamp=1666819125.700304) [('tile_f', [-1, 1, 64, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7648213
-No: 5 GFLOPS: 0.00/123.68 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8447313
+No: 4 GFLOPS: 1.29/1.29 result: MeasureResult(costs=(0.179328308,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.749636173248291, timestamp=1666822077.5919812) [('tile_f', [-1, 4, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8557307
+No: 5 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1059,11 +1059,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9742540
-No: 6 GFLOPS: 67.39/123.68 result: MeasureResult(costs=(0.0034352445000000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.685016870498657, timestamp=1666819130.5838377) [('tile_f', [-1, 8, 1, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4355178
-No: 7 GFLOPS: 167.02/167.02 result: MeasureResult(costs=(0.0013860574594594594,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6607511043548584, timestamp=1666819132.1200016) [('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7403292
-No: 8 GFLOPS: 31.14/167.02 result: MeasureResult(costs=(0.007434749571428571,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5834624767303467, timestamp=1666819132.8072135) [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1938378
-No: 9 GFLOPS: 0.00/167.02 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 8, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8315757
+No: 6 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1185,11 +1182,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 8, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6020236
-No: 10 GFLOPS: 61.38/167.02 result: MeasureResult(costs=(0.003771696363636364,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.913536787033081, timestamp=1666819137.3885193) [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8728026
-No: 11 GFLOPS: 705.96/705.96 result: MeasureResult(costs=(0.0003279260647249191,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2615830898284912, timestamp=1666819138.0211608) [('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7753312
-No: 12 GFLOPS: 69.82/705.96 result: MeasureResult(costs=(0.003315454791666667,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.487569570541382, timestamp=1666819138.9851844) [('tile_f', [-1, 1, 2, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9423808
-No: 13 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6919113
+No: 7 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1311,8 +1305,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 16, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9829069
-No: 14 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 8]), ('tile_y', [-1, 1, 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', 0), ('unroll_explicit', 1)],None,6181257
+No: 8 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1434,8 +1428,26 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1036307
-No: 15 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 2, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1913751
+No: 9 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
+ res = future.result()
+ File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
+ return self.__get_result()
+ File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
+ raise self._exception
+ File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
+ result = self.fn(*self.args, **self.kwargs)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
+ worker = lambda *args: self._worker_run(*args)
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
+ return proc.recv()
+ File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
+ raise TimeoutError()
+TimeoutError
+
+ [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9466171
+No: 10 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1557,8 +1569,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 1, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9511704
-No: 16 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7060360
+No: 11 GFLOPS: 0.00/1.29 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1680,10 +1692,9 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4813683
-No: 17 GFLOPS: 90.03/705.96 result: MeasureResult(costs=(0.002571501897435897,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.409144639968872, timestamp=1666819142.0864537) [('tile_f', [-1, 4, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2526554
-No: 18 GFLOPS: 21.16/705.96 result: MeasureResult(costs=(0.0109404319,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.6973342895507812, timestamp=1666819142.8425627) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8586278
-No: 19 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5223606
+No: 12 GFLOPS: 145.84/145.84 result: MeasureResult(costs=(0.0015873300317460317,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7088942527770996, timestamp=1666822089.8054924) [('tile_f', [-1, 8, 8, 1]), ('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, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8725010
+No: 13 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1805,8 +1816,8 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 128, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6775831
-No: 20 GFLOPS: 0.00/705.96 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6542100
+No: 14 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1928,7 +1939,623 @@ 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 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 2, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3893234
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5646420
+No: 15 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3322291
+No: 16 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10438315
+No: 17 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8412683
+No: 18 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2106803
+No: 19 GFLOPS: 0.00/145.84 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+ func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+ func = build(s, args, target_host=task.target_host, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, 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:1731
+ 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:1671
+ 19: run<>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1631
+ 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+ at ../include/tvm/runtime/packed_func.h:1646
+ 13: operator()
+ at ../src/driver/driver_api.cc:391
+ 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:377
+ 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+ at ../src/driver/driver_api.cc:272
+ 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:453
+ 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:1750
+ 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:1694
+ 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:1618
+ 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 871, in verify_pass
+ raise InstantiationError("Skipped because of invalid gpu kernel")
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,825458
+No: 20 GFLOPS: 74.85/145.84 result: MeasureResult(costs=(0.003092781515151515,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.605224609375, timestamp=1666822093.1458662) [('tile_f', [-1, 2, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1233575
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -1967,9 +2594,9 @@ and measure running time.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Finish loading 20 records
Best config:
-[('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7753312
+[('tile_f', [-1, 8, 8, 1]), ('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, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8725010
Finish loading 20 records
-Time cost of this operator: 0.000739
+Time cost of this operator: 0.000958
</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 3aea3fdcd8..2805b2fa9d 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -595,10 +595,10 @@ 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 311.1 98.723 (1, 2, 10, 10, 3) 2 1 [311.1]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.042 0.965 (1, 6, 10, 10) 1 1 [3.042]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.983 0.312 (1, 1, 10, 10, 3) 1 1 [0.983]
-Total_time - 315.125 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 311.6 98.733 (1, 2, 10, 10, 3) 2 1 [311.6]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.027 0.959 (1, 6, 10, 10) 1 1 [3.027]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.973 0.308 (1, 1, 10, 10, 3) 1 1 [0.973]
+Total_time - 315.6 - - - - -
</pre></div>
</div>
</div>
@@ -649,10 +649,10 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 127.6 97.879 (1, 6, 10, 10, 1) 2 1 [127.6]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.811 1.389 (1, 6, 10, 10) 1 1 [1.811]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.954 0.732 (1, 1, 10, 10, 3) 1 1 [0.954]
-Total_time - 130.365 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 105.8 97.563 (1, 6, 10, 10, 1) 2 1 [105.8]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.796 1.656 (1, 6, 10, 10) 1 1 [1.796]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.847 0.781 (1, 3, 10, 10, 1) 1 1 [0.847]
+Total_time - 108.443 - - - - -
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index 1f6b133b3f..4088435b3d 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -529,7 +529,7 @@ 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/tmp1suj57_8/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpbu2mt11i/images/random'
</pre></div>
</div>
</div>
@@ -589,8 +589,8 @@ 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], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [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/tmp1suj57_8/images/target contains 8144 images
-/tmp/tmp1suj57_8/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [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/tmpbu2mt11i/images/target contains 8144 images
+/tmp/tmpbu2mt11i/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -702,13 +702,13 @@ 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 - 46s - loss: 0.2086 - accuracy: 0.9282 - val_loss: 0.2560 - val_accuracy: 0.9248 - 46s/epoch - 141ms/step
+328/328 - 47s - loss: 0.2181 - accuracy: 0.9245 - val_loss: 0.1206 - val_accuracy: 0.9566 - 47s/epoch - 144ms/step
Epoch 2/3
-328/328 - 42s - loss: 0.0914 - accuracy: 0.9654 - val_loss: 0.1092 - val_accuracy: 0.9615 - 42s/epoch - 129ms/step
+328/328 - 44s - loss: 0.1039 - accuracy: 0.9615 - val_loss: 0.1075 - val_accuracy: 0.9600 - 44s/epoch - 133ms/step
Epoch 3/3
-328/328 - 42s - loss: 0.0709 - accuracy: 0.9767 - val_loss: 0.1294 - val_accuracy: 0.9619 - 42s/epoch - 130ms/step
+328/328 - 44s - loss: 0.0673 - accuracy: 0.9749 - val_loss: 0.1221 - val_accuracy: 0.9619 - 44s/epoch - 133ms/step
-<keras.callbacks.History object at 0x7f57883edc50>
+<keras.callbacks.History object at 0x7f1333acc950>
</pre></div>
</div>
</div>
@@ -970,7 +970,7 @@ 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 37.024 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 42.046 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 dae629212b..7732dac6f3 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -340,7 +340,7 @@
<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>05:37.830</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>05:47.208</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -349,19 +349,19 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>04:37.024</p></td>
+<td><p>04:42.046</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>00:48.043</p></td>
+<td><p>00:51.418</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><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">microTVM Host-Driven AoT</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:09.126</p></td>
+<td><p>00:09.811</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">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:03.635</p></td>
+<td><p>00:03.930</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">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 1ccb40a83a..0f4bde2c9f 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -340,7 +340,7 @@
<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>00:44.332</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:46.966</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -349,15 +349,15 @@
</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>00:32.689</p></td>
+<td><p>00:34.811</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>00:10.113</p></td>
+<td><p>00:10.510</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>00:01.523</p></td>
+<td><p>00:01.637</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 ba918c15ee..b16f0253c5 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -535,7 +535,7 @@ 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 0x7f5738aae830>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f136402fdd0>
</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 0d789a59b0..3b24516bfb 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -340,7 +340,7 @@
<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>00:08.076</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:08.715</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -349,35 +349,35 @@
</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>00:05.760</p></td>
+<td><p>00:06.239</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>00:01.020</p></td>
+<td><p>00:01.122</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>00:00.553</p></td>
+<td><p>00:00.580</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>00:00.537</p></td>
+<td><p>00:00.552</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>00:00.115</p></td>
+<td><p>00:00.121</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.046</p></td>
+<td><p>00:00.052</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></td>
-<td><p>00:00.029</p></td>
+<td><p>00:00.031</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></td>
-<td><p>00:00.016</p></td>
+<td><p>00:00.018</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index 4c6a49c4b2..216b4dd87c 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -590,7 +590,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmphhwb1r7r/input0.cc'\nsource_filename = \"/tmp/tmphhwb1r7r/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpnv73ocip/input0.cc'\nsource_filename = \"/tmp/tmpnv73ocip/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index c5d6b0f653..bf24d2a4b7 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -229,7 +229,17 @@
<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/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index d1b015dc54..4ea055bfbe 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1615,7 +1615,7 @@ 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>
@@ -1899,7 +1899,7 @@ Candidates:
<dl class="py function">
<dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
<dd><p>THIS API IS DEPRECATED.</p>
<p>Run auto scheduling search for a task.</p>
<dl class="field-list simple">
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 787b955801..2153823a48 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/697533e13/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
</section>
@@ -151,7 +151,7 @@
<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
</ul>
</aside>
</section>
@@ -168,7 +168,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index 5813f0b5c3..1b74b0818a 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/697533e13/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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 88a7c64925..f7066ae65e 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/697533e13/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L262">runtime.ts:262</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L260">runtime.ts:260</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L260">runtime.ts:260</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">code<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L258">runtime.ts:258</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L258">runtime.ts:258</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -177,7 +177,7 @@
<div class="tsd-signature tsd-kind-icon">lanes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L262">runtime.ts:262</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -199,7 +199,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L279">runtime.ts:279</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L279">runtime.ts:279</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -216,7 +216,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L270">runtime.ts:270</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L270">runtime.ts:270</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/dldevice.html b/docs/reference/api/typedoc/classes/dldevice.html
index e177acb231..b104d8a33e 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/697533e13/web/src/runtime.ts#L202">runtime.ts:202</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L202">runtime.ts:202</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L200">runtime.ts:200</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L200">runtime.ts:200</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -161,7 +161,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L198">runtime.ts:198</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -183,7 +183,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L223">runtime.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L223">runtime.ts:223</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -205,7 +205,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L230">runtime.ts:230</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L230">runtime.ts:230</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/environment.html b/docs/reference/api/typedoc/classes/environment.html
index 32faf6f733..879705d2d8 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/697533e13/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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 f0e1a7ce8d..cfa390f7a5 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/697533e13/web/src/runtime.ts#L49">runtime.ts:49</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L49">runtime.ts:49</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/697533e13/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L46">runtime.ts:46</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/697533e13/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L45">runtime.ts:45</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/697533e13/web/src/runtime.ts#L44">runtime.ts:44</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L44">runtime.ts:44</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/697533e13/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L47">runtime.ts:47</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/697533e13/web/src/runtime.ts#L76">runtime.ts:76</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L76">runtime.ts:76</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/697533e13/web/src/runtime.ts#L66">runtime.ts:66</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L66">runtime.ts:66</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/697533e13/web/src/runtime.ts#L84">runtime.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L84">runtime.ts:84</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/697533e13/web/src/runtime.ts#L95">runtime.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L95">runtime.ts:95</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/697533e13/web/src/runtime.ts#L72">runtime.ts:72</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L72">runtime.ts:72</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/graphexecutor.html b/docs/reference/api/typedoc/classes/graphexecutor.html
index ec761e3549..9e985f5982 100644
--- a/docs/reference/api/typedoc/classes/graphexecutor.html
+++ b/docs/reference/api/typedoc/classes/graphexecutor.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/697533e13/web/src/runtime.ts#L583">runtime.ts:583</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L583">runtime.ts:583</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">module<span class="tsd-signature-symbol">:</span> <a href="module.html" class="tsd-signature-type">Module</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L579">runtime.ts:579</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L579">runtime.ts:579</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/697533e13/web/src/runtime.ts#L654">runtime.ts:654</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L654">runtime.ts:654</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/697533e13/web/src/runtime.ts#L597">runtime.ts:597</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L597">runtime.ts:597</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -241,7 +241,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L631">runtime.ts:631</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L631">runtime.ts:631</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -279,7 +279,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L644">runtime.ts:644</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L644">runtime.ts:644</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -310,7 +310,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L621">runtime.ts:621</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L621">runtime.ts:621</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -332,7 +332,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L609">runtime.ts:609</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L609">runtime.ts:609</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/instance.html b/docs/reference/api/typedoc/classes/instance.html
index 9e60351f53..fc3f69e1fc 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -139,7 +139,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L692">runtime.ts:692</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L692">runtime.ts:692</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -202,7 +202,7 @@
<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/697533e13/web/src/runtime.ts#L684">runtime.ts:684</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L684">runtime.ts:684</a></li>
</ul>
</aside>
</section>
@@ -212,7 +212,7 @@
<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/697533e13/web/src/runtime.ts#L683">runtime.ts:683</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L683">runtime.ts:683</a></li>
</ul>
</aside>
</section>
@@ -229,7 +229,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L932">runtime.ts:932</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -260,7 +260,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L994">runtime.ts:994</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L994">runtime.ts:994</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -303,7 +303,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L924">runtime.ts:924</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L924">runtime.ts:924</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -341,7 +341,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L732">runtime.ts:732</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L732">runtime.ts:732</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -358,7 +358,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L952">runtime.ts:952</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L952">runtime.ts:952</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -402,7 +402,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L816">runtime.ts:816</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L816">runtime.ts:816</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -434,7 +434,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L1033">runtime.ts:1033</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L1033">runtime.ts:1033</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -465,7 +465,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L846">runtime.ts:846</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L846">runtime.ts:846</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -497,7 +497,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L750">runtime.ts:750</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L750">runtime.ts:750</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -520,7 +520,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L1013">runtime.ts:1013</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L1013">runtime.ts:1013</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -568,7 +568,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L789">runtime.ts:789</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L789">runtime.ts:789</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -608,7 +608,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L914">runtime.ts:914</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L914">runtime.ts:914</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -646,7 +646,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L1145">runtime.ts:1145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L1145">runtime.ts:1145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -698,7 +698,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L740">runtime.ts:740</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L740">runtime.ts:740</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -722,7 +722,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L868">runtime.ts:868</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L868">runtime.ts:868</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -754,7 +754,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L857">runtime.ts:857</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L857">runtime.ts:857</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -786,7 +786,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L940">runtime.ts:940</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L940">runtime.ts:940</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 6c12ba63f7..d7bff45f2d 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/697533e13/web/src/memory.ts#L40">memory.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L32">memory.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L33">memory.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L154">memory.ts:154</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L90">memory.ts:90</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L97">memory.ts:97</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L74">memory.ts:74</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L81">memory.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L104">memory.ts:104</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L132">memory.ts:132</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L145">memory.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L60">memory.ts:60</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L67">memory.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L53">memory.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L114">memory.ts:114</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L124">memory.ts:124</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/memory.ts#L175">memory.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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 911c5a9ee1..9865ca13e9 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.html
@@ -124,7 +124,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L504">runtime.ts:504</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L504">runtime.ts:504</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -170,7 +170,7 @@
<div class="tsd-signature tsd-kind-icon">handle<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/697533e13/web/src/runtime.ts#L502">runtime.ts:502</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L502">runtime.ts:502</a></li>
</ul>
</aside>
</section>
@@ -187,7 +187,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L516">runtime.ts:516</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L516">runtime.ts:516</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -204,7 +204,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L530">runtime.ts:530</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L530">runtime.ts:530</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -236,7 +236,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L561">runtime.ts:561</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L561">runtime.ts:561</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 9e6e3f829a..27dd468704 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/697533e13/web/src/runtime.ts#L304">runtime.ts:304</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L304">runtime.ts:304</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/697533e13/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L297">runtime.ts:297</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/697533e13/web/src/runtime.ts#L293">runtime.ts:293</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L293">runtime.ts:293</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
<div class="tsd-signature tsd-kind-icon">handle<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/697533e13/web/src/runtime.ts#L289">runtime.ts:289</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L289">runtime.ts:289</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
<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/697533e13/web/src/runtime.ts#L291">runtime.ts:291</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L291">runtime.ts:291</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
<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/697533e13/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L295">runtime.ts:295</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -240,7 +240,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L370">runtime.ts:370</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L370">runtime.ts:370</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -273,7 +273,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L414">runtime.ts:414</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L414">runtime.ts:414</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -305,7 +305,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L355">runtime.ts:355</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -322,7 +322,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L474">runtime.ts:474</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L474">runtime.ts:474</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -346,7 +346,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L443">runtime.ts:443</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L443">runtime.ts:443</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 ba51032333..97faf74f68 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -122,7 +122,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L158">runtime.ts:158</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L158">runtime.ts:158</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">handle<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/697533e13/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L157">runtime.ts:157</a></li>
</ul>
</aside>
</section>
@@ -164,7 +164,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/697533e13/web/src/runtime.ts#L165">runtime.ts:165</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L165">runtime.ts:165</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index 8f00e534a4..e33eea766c 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/697533e13/web/src/rpc_server.ts#L92">rpc_server.ts:92</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L92">rpc_server.ts:92</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/697533e13/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L82">rpc_server.ts:82</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/697533e13/web/src/rpc_server.ts#L78">rpc_server.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L78">rpc_server.ts:78</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/697533e13/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L81">rpc_server.ts:81</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/697533e13/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L79">rpc_server.ts:79</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/697533e13/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L80">rpc_server.ts:80</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/697533e13/web/src/rpc_server.ts#L77">rpc_server.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L77">rpc_server.ts:77</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index 7deb7cc460..35251b0d46 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/697533e13/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L145">runtime.ts:145</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/697533e13/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L145">runtime.ts:145</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/697533e13/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L143">runtime.ts:143</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 5982bedf42..e742a00b38 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/697533e13/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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/697533e13/web/src/webgpu.ts#L170">webgpu.ts:170</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/webgpu.ts#L170">webgpu.ts:170</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/697533e13/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/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 03088bce35..23d84f0a7b 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/697533e13/web/src/ctypes.ts#L220">ctypes.ts:220</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L220">ctypes.ts:220</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/697533e13/web/src/ctypes.ts#L216">ctypes.ts:216</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L216">ctypes.ts:216</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/697533e13/web/src/ctypes.ts#L214">ctypes.ts:214</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L214">ctypes.ts:214</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/697533e13/web/src/ctypes.ts#L218">ctypes.ts:218</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L218">ctypes.ts:218</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/697533e13/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L226">ctypes.ts:226</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/697533e13/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L221">ctypes.ts:221</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/697533e13/web/src/ctypes.ts#L219">ctypes.ts:219</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L219">ctypes.ts:219</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/697533e13/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L223">ctypes.ts:223</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/697533e13/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L227">ctypes.ts:227</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/697533e13/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L222">ctypes.ts:222</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/697533e13/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L228">ctypes.ts:228</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/697533e13/web/src/ctypes.ts#L217">ctypes.ts:217</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L217">ctypes.ts:217</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/697533e13/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L224">ctypes.ts:224</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/697533e13/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L225">ctypes.ts:225</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/697533e13/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/aynccallbackcode.html b/docs/reference/api/typedoc/enums/aynccallbackcode.html
index d38b70d11e..94616c7c93 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/697533e13/web/src/runtime.ts#L676">runtime.ts:676</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L676">runtime.ts:676</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/697533e13/web/src/runtime.ts#L675">runtime.ts:675</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L675">runtime.ts:675</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/dldatatypecode.html b/docs/reference/api/typedoc/enums/dldatatypecode.html
index f1f3e49bdd..48142f3600 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/697533e13/web/src/runtime.ts#L242">runtime.ts:242</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L242">runtime.ts:242</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/697533e13/web/src/runtime.ts#L240">runtime.ts:240</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L240">runtime.ts:240</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/697533e13/web/src/runtime.ts#L243">runtime.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L243">runtime.ts:243</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/697533e13/web/src/runtime.ts#L241">runtime.ts:241</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/runtime.ts#L241">runtime.ts:241</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/rpcserverstate.html b/docs/reference/api/typedoc/enums/rpcserverstate.html
index 1f4eae5a45..8b84367a9d 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/697533e13/web/src/rpc_server.ts#L27">rpc_server.ts:27</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L27">rpc_server.ts:27</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/697533e13/web/src/rpc_server.ts#L28">rpc_server.ts:28</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L28">rpc_server.ts:28</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/697533e13/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L29">rpc_server.ts:29</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/697533e13/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L32">rpc_server.ts:32</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/697533e13/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L31">rpc_server.ts:31</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/697533e13/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/sizeof.html b/docs/reference/api/typedoc/enums/sizeof.html
index ebab3cb0ce..1a94bfc030 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/697533e13/web/src/ctypes.ts#L206">ctypes.ts:206</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L206">ctypes.ts:206</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/697533e13/web/src/ctypes.ts#L207">ctypes.ts:207</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L207">ctypes.ts:207</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/697533e13/web/src/ctypes.ts#L203">ctypes.ts:203</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L203">ctypes.ts:203</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/697533e13/web/src/ctypes.ts#L204">ctypes.ts:204</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L204">ctypes.ts:204</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/697533e13/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L201">ctypes.ts:201</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/697533e13/web/src/ctypes.ts#L202">ctypes.ts:202</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L202">ctypes.ts:202</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/697533e13/web/src/ctypes.ts#L205">ctypes.ts:205</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L205">ctypes.ts:205</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/697533e13/web/src/ctypes.ts#L200">ctypes.ts:200</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L200">ctypes.ts:200</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/697533e13/web/src/ctypes.ts#L199">ctypes.ts:199</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L199">ctypes.ts:199</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/index.html b/docs/reference/api/typedoc/index.html
index 01773503a2..2ce914a09a 100644
--- a/docs/reference/api/typedoc/index.html
+++ b/docs/reference/api/typedoc/index.html
@@ -174,7 +174,7 @@
<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/697533e13/web/src/ctypes.ts#L112">ctypes.ts:112</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L112">ctypes.ts:112</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
<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/697533e13/web/src/ctypes.ts#L128">ctypes.ts:128</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L128">ctypes.ts:128</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -282,7 +282,7 @@
<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/697533e13/web/src/ctypes.ts#L144">ctypes.ts:144</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L144">ctypes.ts:144</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -326,7 +326,7 @@
<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/697533e13/web/src/ctypes.ts#L136">ctypes.ts:136</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L136">ctypes.ts:136</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -370,7 +370,7 @@
<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/697533e13/web/src/ctypes.ts#L121">ctypes.ts:121</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/04afd8311/web/src/ctypes.ts#L121">ctypes.ts:121</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -406,7 +406,7 @@
<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/697533e13/web/src/ctypes.ts#L160">ctypes.ts:160</a></li>
... 1297 lines suppressed ...