You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2023/03/14 00:54:13 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@ff12a2032352d8376cc902ca996e1121d405b3e1)
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 fe6a0f53c3 deploying docs (apache/tvm@ff12a2032352d8376cc902ca996e1121d405b3e1)
fe6a0f53c3 is described below
commit fe6a0f53c38afac46b718679e73fba3297be1df9
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Tue Mar 14 00:54:06 2023 +0000
deploying docs (apache/tvm@ff12a2032352d8376cc902ca996e1121d405b3e1)
---
docs/_images/sphx_glr_micro_train_001.png | Bin 346533 -> 333957 bytes
docs/_images/sphx_glr_micro_train_thumb.png | Bin 24178 -> 23827 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_adreno.rst.txt | 2 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 20 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 2 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 8 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 14 +-
.../tune_conv2d_layer_cuda.rst.txt | 1116 +++++---------------
.../tune_network_cuda.rst.txt | 4 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 106 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 6 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 335 +++---
.../work_with_microtvm/micro_autotune.rst.txt | 18 +-
.../work_with_microtvm/micro_pytorch.rst.txt | 4 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 18 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 12 +-
.../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 | 12 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 4 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 6 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 57 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 18 +-
.../tutorial/tensor_expr_get_started.rst.txt | 45 +-
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 | 12 +-
docs/how_to/compile_models/from_pytorch.html | 9 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 22 +-
.../deploy_models/deploy_model_on_adreno.html | 2 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 45 +-
docs/how_to/deploy_models/deploy_prequantized.html | 9 +-
.../deploy_models/deploy_prequantized_tflite.html | 4 +-
docs/how_to/deploy_models/deploy_quantized.html | 2 +-
docs/how_to/deploy_models/deploy_ssd_gluoncv.html | 36 +-
docs/how_to/deploy_models/sg_execution_times.html | 20 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
docs/how_to/extend_tvm/sg_execution_times.html | 8 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 1116 +++++---------------
.../tune_with_autoscheduler/tune_network_cuda.html | 4 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 106 +-
.../tune_with_autotvm/sg_execution_times.html | 6 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 335 +++---
docs/how_to/work_with_microtvm/micro_autotune.html | 18 +-
docs/how_to/work_with_microtvm/micro_pytorch.html | 4 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 12 +-
.../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 | 12 +-
docs/install/nnpack.html | 12 +-
.../api/doxygen/detail_2broadcast_8h_source.html | 2 +-
.../api/doxygen/group__norm_8h_source.html | 2 +-
.../api/doxygen/layer__norm_8h_source.html | 2 +-
.../api/doxygen/nn_2pooling_8h_source.html | 8 +-
.../reference/api/doxygen/reduction_8h_source.html | 22 +-
docs/reference/api/python/auto_scheduler.html | 4 +-
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
docs/reference/api/typedoc/classes/instance.html | 58 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
.../api/typedoc/classes/runtimecontext.html | 22 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
docs/reference/api/typedoc/classes/tvmarray.html | 16 +-
docs/reference/api/typedoc/classes/tvmobject.html | 12 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 124 +--
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 4 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 6 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 5 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 267 ++---
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 18 +-
docs/tutorial/tensor_expr_get_started.html | 41 +-
135 files changed, 1802 insertions(+), 3004 deletions(-)
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 1612aabc2c..b0c19c3436 100644
Binary files a/docs/_images/sphx_glr_micro_train_001.png and b/docs/_images/sphx_glr_micro_train_001.png differ
diff --git a/docs/_images/sphx_glr_micro_train_thumb.png b/docs/_images/sphx_glr_micro_train_thumb.png
index b25d3cb0c8..475ea5ff8a 100644
Binary files a/docs/_images/sphx_glr_micro_train_thumb.png and b/docs/_images/sphx_glr_micro_train_thumb.png differ
diff --git a/docs/_sources/how_to/compile_models/from_darknet.rst.txt b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
index 05a6fc13e2..07793db574 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -318,7 +318,7 @@ The process is no different from other examples.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 19.115 seconds)
+ **Total running time of the script:** ( 1 minutes 21.306 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 6dd9420e81..aa07564c94 100644
--- a/docs/_sources/how_to/compile_models/from_keras.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_keras.rst.txt
@@ -232,7 +232,7 @@ Look up prediction top 1 index in 1000 class synset.
.. code-block:: none
Relay top-1 id: 285, class name: Egyptian cat
-
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 941ms/step
+
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 990ms/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 e390ecf31f..30670b1057 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -116,7 +116,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip4fd6033c-3c45-4b1b-84d2-4c6ffd33b2f3 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip3a91dd03-5f9a-467a-88e2-d4a308d99f35 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 38c139c9b5..ae6604d91c 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -121,7 +121,7 @@ Load a pretrained OneFlow model and save model
.. code-block:: none
Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 76.7MB/s]
39%|###8 | 16.0M/41.5M [00:00<00:00, 53.5MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 57.7MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 65.6MB/s]
93%|#########3| 38.6M/41.5M [00:00<00:00, 66.6MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 64.6MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 66.9MB/s]
39%|###8 | 16.0M/41.5M [00:00<00:00, 64.7MB/s]
58%|#####7 | 24.0M/41.5M [00:00<00:00, 62.1MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 64.9MB/s]
96%|#########6| 40.0M/41.5M [00:00<00:00, 66.3MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 67.2MB/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 0b79ec645d..27ec0680c5 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -101,7 +101,7 @@ Load a pretrained PyTorch model
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
0%| | 0.00/44.7M [00:00<?, ?B/s]
28%|##7 | 12.4M/44.7M [00:00<00:00, 130MB/s]
55%|#####5 | 24.8M/44.7M [00:00<00:00, 98.4MB/s]
82%|########1 | 36.6M/44.7M [00:00<00:00, 108MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 101MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
18%|#7 | 7.99M/44.7M [00:00<00:00, 78.1MB/s]
49%|####9 | 22.0M/44.7M [00:00<00:00, 117MB/s]
75%|#######4 | 33.3M/44.7M [00:00<00:00, 92.8MB/s]
95%|#########5| 42.6M/44.7M [00:00<00:00, 85.4MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 87.5MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
index c3e49a324e..03ed0b6c31 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -424,7 +424,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 23.459 seconds)
+ **Total running time of the script:** ( 1 minutes 23.775 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 a5f81bed85..bee272d2a0 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
Computation times
=================
-**06:37.193** total execution time for **how_to_compile_models** files:
+**06:42.775** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:23.459 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:23.775 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:19.115 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:21.306 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:56.187 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:57.512 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:37.473 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:37.996 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:31.830 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:32.531 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:31.078 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:31.196 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:27.340 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:27.306 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:26.181 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:25.628 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:21.810 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:22.771 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.719 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.753 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
index ff72069817..0ee7307026 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
@@ -727,7 +727,7 @@ well as provides information about the model's performance
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 2752.8736 2751.4635 2763.1592 2750.2746 3.7295
+ 2540.5413 2541.5224 2542.5263 2537.8086 1.8890
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 4a37f8d3f2..3d6845e875 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -437,7 +437,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.8604 15.8693 16.0653 15.6673 0.1238
+ 16.2988 16.1252 16.8254 15.7942 0.3921
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 55a9e0af3f..cc01a74c69 100644
--- a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
@@ -130,7 +130,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MaskRCNN_ResNet50_FPN_Weights.COCO_V1`. You can also use `weights=MaskRCNN_ResNet50_FPN_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 8.12M/170M [00:00<00:02, 83.0MB/s]
9%|9 | 16.0M/170M [00:00<00:01, 80.9MB/s]
14%|#4 | 24.0M/170M [00:00<00:02, 74.1MB/s]
21%|##1 | 35.8M/170M [00:00<00:01, 92.4MB/s]
26%|##6 | 44.8M/170M [00:00<00:01, 74.1MB/s]
31%|### | 52.4M/170M [00:00<00:01, 74.1MB/s]
35%|###5 | 59.8M/170M [00:00<00:01, 75.1MB/s]
42%|####2 | 72.0M/170M [00:00<00:01, 76.8MB/s]
52%|#####1 | 88.0M/170M [00:01<00:00, 91.2MB/s]
61%|###### | 103M/170M [00:01<00:00, 108MB/s]
67%|######7 | 114M/170M [00:01<00:00, 103MB/s]
73%|#######2 | 124M/170M [00:01<00:00, 85.2MB/s]
78%|#######8 | 133M/170M [00:01<00:00, 84.5MB/s]
83%|########3 | 141M/170M [00:01<00:00, 84.6MB/s]
89%|########9 | 152M/170M [00:01<00:00, 90.9MB/s]
94%|#########4| 160M/170M [00:01<00:00, 82.5MB/s]
100%|##########| 170M/170M [00:02<00:00, 87.1MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 7.99M/170M [00:00<00:03, 50.8MB/s]
9%|8 | 14.8M/170M [00:00<00:02, 60.2MB/s]
12%|#2 | 20.8M/170M [00:00<00:02, 59.5MB/s]
16%|#5 | 26.6M/170M [00:00<00:03, 48.8MB/s]
19%|#8 | 31.5M/170M [00:00<00:03, 40.3MB/s]
21%|## | 35.6M/170M [00:00<00:03, 38.8MB/s]
24%|##3 | 40.0M/170M [00:00<00:03, 36.9MB/s]
28%|##8 | 48.0M/170M [00:01<00:02, 43.9MB/s]
33%|###2 | 56.0M/170M [00:01<00:02, 51.6MB/s]
38%|###7 | 64.0M/170M [00:01<00:02, 48.6MB/s]
42%|####2 | 72.0M/170M [00:01<00:01, 53.9MB/s]
47%|####7 | 80.0M/170M [00:01<00:01, 52.3MB/s]
52%|#####1 | 88.0M/170M [00:01<00:01, 44.7MB/s]
56%|#####5 | 94.3M/170M [00:02<00:01, 45.4MB/s]
60%|###### | 102M/170M [00:02<00:01, 41.8MB/s]
63%|######2 | 107M/170M [00:02<00:01, 40.4MB/s]
67%|######6 | 114M/170M [00:02<00:01, 47.3MB/s]
71%|####### | 120M/170M [00:02<00:01, 41.3MB/s]
75%|#######5 | 128M/170M [00:02<00:00, 43.9MB/s]
80%|######## | 136M/170M [00:03<00:00, 46.0MB/s]
85%|########4 | 144M/170M [00:03<00:00, 49.8MB/s]
88%|########8 | 150M/170M [00:03<00:00, 45.2MB/s]
91%|#########1| 155M/170M [00:03<00:00, 40.8MB/s]
94%|#########4| 160M/170M [00:03<00:00, 39.9MB/s]
99%|#########8| 168M/170M [00:03<00:00, 49.0MB/s]
100%|##########| 170M/170M [00:03<00:00, 46.3MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -299,7 +299,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 33.082 seconds)
+ **Total running time of the script:** ( 3 minutes 39.018 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 248b06cee3..50a4984628 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -227,7 +227,7 @@ training. Other models require a full post training calibration.
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MobileNet_V2_Weights.IMAGENET1K_V1`. You can also use `weights=MobileNet_V2_Weights.DEFAULT` to get the most up-to-date weights.
warnings.warn(msg)
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
0%| | 0.00/13.6M [00:00<?, ?B/s]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 80.1MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 102MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
47%|####6 | 6.30M/13.6M [00:00<00:00, 44.1MB/s]
78%|#######7 | 10.5M/13.6M [00:00<00:00, 43.6MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 45.7MB/s]
@@ -409,7 +409,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.3305 90.1795 95.6701 90.0084 0.5895
+ 90.3555 90.3418 92.0020 90.0462 0.2361
@@ -458,7 +458,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 16.788 seconds)
+ **Total running time of the script:** ( 1 minutes 17.863 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 32693fdef0..2ef834c870 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -423,7 +423,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 118.7670 118.6938 121.5630 117.8118 0.4996
+ 119.9148 119.8763 122.1049 119.2164 0.4122
@@ -460,7 +460,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 34.564 seconds)
+ **Total running time of the script:** ( 2 minutes 31.118 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 7a49bdc8be..ce5ed5d911 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -257,7 +257,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 36.031 seconds)
+ **Total running time of the script:** ( 1 minutes 36.012 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 d4208fd558..bc976c2ba2 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -170,7 +170,7 @@ Convert and compile model for CPU.
data: None
input_sym_arg_type = in_param.infer_type()[0]
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
0%| | 0/132723 [00:00<?, ?KB/s]
1%|1 | 1989/132723 [00:00<00:06, 19888.69KB/s]
5%|4 | 6158/132723 [00:00<00:03, 32711.42KB/s]
10%|# | 13445/132723 [00:00<00:02, 51047.11KB/s]
16%|#5 | 20863/132723 [00:00<00:01, 60176.73KB/s]
22%|##2 | 29504/132723 [00:00<00:01, 69633.26KB/s]
29%|##8 | 38185/132723 [00:00<00:01, 75471.14KB/s]
35%|###5 | 46783/132723 [00:00<00:01, 78904.17KB/s]
42%|####1 | 55443/132723 [00:00<00:00, 81352.12KB/s]
48%|####8 | 64109/132723 [00:00<00:00, 83006.98KB/s]
55%|#####4 | 72804/132723 [00:01<00:00, 84221.48KB/s]
61%|######1 | 81458/132723 [00:01<00:00, 84928.31KB/s]
68%|######7 | 90085/132723 [00:01<00:00, 85333.86KB/s]
74%|#######4 | 98726/132723 [00:01<00:00, 85658.75KB/s]
81%|######## | 107389/132723 [00:01<00:00, 85947.97KB/s]
87%|########7 | 116015/132723 [00:01<00:00, 86034.50KB/s]
94%|#########3
| 124648/132723 [00:01<00:00, 86122.47KB/s]
100%|##########| 132723/132723 [00:01<00:00, 78304.85KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
4%|4 | 5376/132723 [00:00<00:02, 53756.72KB/s]
9%|9 | 12535/132723 [00:00<00:01, 64244.31KB/s]
14%|#4 | 18960/132723 [00:00<00:01, 63847.55KB/s]
21%|## | 27659/132723 [00:00<00:01, 72946.38KB/s]
26%|##6 | 34958/132723 [00:00<00:01, 70871.06KB/s]
33%|###2 | 43677/132723 [00:00<00:01, 76301.14KB/s]
39%|###8 | 51327/132723 [00:00<00:01, 56104.09KB/s]
45%|####5 | 60042/132723 [00:00<00:01, 63950.61KB/s]
52%|#####1 | 68705/132723 [00:01<00:00, 69964.92KB/s]
58%|#####8 | 77495/132723 [00:01<00:00, 74891.88KB/s]
65%|######5 | 86287/132723 [00:01<00:00, 78562.92KB/s]
72%|#######1 | 95105/132723 [00:01<00:00, 81320.90KB/s]
78%|#######8 | 103826/132723 [00:01<00:00, 83026.80KB/s]
85%|########4 | 112531/132723 [00:01<00:00, 84204.05KB/s]
91%|#########1| 121325/132723 [00:01<00:00, 85303.44KB/s]
98%|########
#8| 130124/132723 [00:01<00:00, 86097.96KB/s]
100%|##########| 132723/132723 [00:01<00:00, 75710.28KB/s]
@@ -246,7 +246,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 45.518 seconds)
+ **Total running time of the script:** ( 3 minutes 48.579 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 385183bb59..e0067c3e54 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
@@ -5,26 +5,26 @@
Computation times
=================
-**15:22.218** total execution time for **how_to_deploy_models** files:
+**15:27.211** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 03:45.518 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 03:48.579 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:33.082 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:39.018 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:34.564 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:31.118 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:36.031 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:36.012 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:16.788 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:17.863 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:57.082 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``) | 00:54.802 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:42.485 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:43.500 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:28.484 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:28.466 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:28.179 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:27.847 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.006 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
index b73d0c69ca..b5a81be5d4 100644
--- a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
@@ -463,7 +463,7 @@ First let us define two helper functions to get the mobilenet model and a cat im
.. code-block:: none
- Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip41bcb284-c50e-4de0-8317-aec415239e34 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipe688c588-85b8-486a-b730-9ee5303afc08 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 2a3dff9387..fc9aca8674 100644
--- a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
Computation times
=================
-**00:54.048** total execution time for **how_to_extend_tvm** files:
+**00:55.422** total execution time for **how_to_extend_tvm** files:
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:50.204 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:51.480 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.745 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.816 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.089 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.117 | 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 18b9cf769b..d16139a60b 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -220,10 +220,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 22478us [22478us] (48.55%; 48.55%)
- FoldScaleAxis: 23818us [24us] (51.45%; 51.45%)
- FoldConstant: 23794us [1761us] (51.40%; 99.90%)
- InferType: 22033us [22033us] (47.59%; 92.60%)
+ InferType: 22340us [22340us] (48.11%; 48.11%)
+ FoldScaleAxis: 24100us [8us] (51.89%; 51.89%)
+ FoldConstant: 24092us [1742us] (51.88%; 99.97%)
+ InferType: 22350us [22350us] (48.13%; 92.77%)
@@ -262,10 +262,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 22158us [22158us] (48.39%; 48.39%)
- FoldScaleAxis: 23635us [6us] (51.61%; 51.61%)
- FoldConstant: 23629us [1778us] (51.60%; 99.97%)
- InferType: 21851us [21851us] (47.72%; 92.48%)
+ InferType: 22486us [22486us] (48.14%; 48.14%)
+ FoldScaleAxis: 24222us [9us] (51.86%; 51.86%)
+ FoldConstant: 24213us [1817us] (51.84%; 99.96%)
+ InferType: 22396us [22396us] (47.95%; 92.49%)
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 dd3521c1af..ac77de9f1a 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -331,7 +331,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 36.923774 ms
+ Convolution: 39.969024 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 0e8ce02400..ed7eca9c2b 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
@@ -598,7 +598,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 13.355830 ms
+ conv2d with tensor core: 13.256323 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 205baa3f41..083ddf2776 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -134,8 +134,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.018244
- Baseline: 3.427344
+ Numpy running time: 0.018360
+ Baseline: 3.272373
@@ -227,7 +227,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.299723
+ Opt1: 0.310553
@@ -318,7 +318,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.331611
+ Opt2: 0.344205
@@ -406,7 +406,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.116872
+ Opt3: 0.113700
@@ -523,7 +523,7 @@ flattening.
.. code-block:: none
- Opt4: 0.109720
+ Opt4: 0.107906
@@ -635,7 +635,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.111627
+ Opt5: 0.110479
@@ -748,7 +748,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.146855
+ Opt6: 0.147061
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 bdebb3bde5..aa219f3227 100644
--- a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
Computation times
=================
-**00:35.023** total execution time for **how_to_optimize_operators** files:
+**00:34.632** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.293 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.048 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.612 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.539 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.118 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.045 | 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 ef89416177..6db587dc78 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
=================
-**09:53.843** total execution time for **how_to_tune_with_autoscheduler** files:
+**09:52.900** total execution time for **how_to_tune_with_autoscheduler** files:
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:04.967 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:02.638 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:42.003 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:42.460 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:07.337 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:07.681 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:31.824 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:32.030 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:14.197 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:14.221 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:13.515 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:13.869 | 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 37f58039fd..78c461d430 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
@@ -243,479 +243,146 @@ cooperative fetching, unrolling and operator fusion.
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- blockIdx_x = T.launch_thread("blockIdx.x", 56)
- conv2d_nchw = T.allocate([7], "float32", "local")
- pad_temp_shared = T.allocate([336], "float32", "shared")
- kernel_shared = T.allocate([3072], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 64)
- conv2d_nchw_1 = T.Buffer((7,), data=conv2d_nchw, scope="local", align=16)
+ blockIdx_x = T.launch_thread("blockIdx.x", 32)
+ conv2d_nchw = T.allocate([14], "float32", "local")
+ pad_temp_shared = T.allocate([6272], "float32", "shared")
+ kernel_shared = T.allocate([2048], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 56)
+ conv2d_nchw_1 = T.Buffer((14,), data=conv2d_nchw, scope="local", align=32)
conv2d_nchw_1[0] = T.float32(0)
+ conv2d_nchw_1[7] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
+ conv2d_nchw_1[8] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
+ conv2d_nchw_1[9] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
+ conv2d_nchw_1[10] = T.float32(0)
conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[11] = T.float32(0)
conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[12] = T.float32(0)
conv2d_nchw_1[6] = T.float32(0)
- for rc_outer_outer, ry_outer_outer in T.grid(32, 3):
- cse_var_4: T.int32 = rc_outer_outer * 784
- cse_var_3: T.int32 = ry_outer_outer * 7
- cse_var_2: T.int32 = rc_outer_outer * 144
+ conv2d_nchw_1[13] = T.float32(0)
+ for rc_outer_outer, ry_outer_outer, rx_outer_outer in T.grid(4, 3, 3):
+ cse_var_2: T.int32 = rc_outer_outer * 1152
cse_var_1: T.int32 = ry_outer_outer * 3
+ pad_temp_shared_1 = T.Buffer((6272,), data=pad_temp_shared, scope="shared")
+ for ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer in range(112):
+ cse_var_3: T.int32 = ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer * 56
+ threadIdx_x_1 = T.launch_thread("threadIdx.x", 56)
+ data_1 = T.Buffer((25088,), data=data.data)
+ pad_temp_shared_1[cse_var_3 + threadIdx_x_1] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 6272 + cse_var_3 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer - 8], T.float32(0))
threadIdx_x_1 = T.env_thread("threadIdx.x")
- pad_temp_shared_1 = T.Buffer((336,), data=pad_temp_shared, scope="shared")
- data_1 = T.Buffer((25088,), data=data.data)
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 <= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 64] = T.if_then_else(1 <= (threadIdx_x_1 + 1) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 1) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 64) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 128] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 2) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 128) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 192] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 3 + 1) % 7 and ry_outer_outer + (threadIdx_x_1 // 3 + 1) % 7 < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 440], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 256] = T.if_then_else(1 <= (threadIdx_x_1 + 4) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 4) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 256) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- if T.likely(threadIdx_x_1 < 16):
- pad_temp_shared_1[threadIdx_x_1 + 320] = T.if_then_else((threadIdx_x_1 + 5) // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 320) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
- threadIdx_x_2 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope="shared")
+ kernel_shared_1 = T.Buffer((2048,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 64) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 64) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 128) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 128) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 192] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 18432]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 256) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 256) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 320) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 320) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 384] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 36864]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 448) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 448) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 512) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 512) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 576] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 55296]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 640) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 640) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 704) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 704) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 768] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 73728]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 832) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 832) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 896) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 896) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 960] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 92160]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1024) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1024) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1088) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1088) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1152] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 110592]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1216) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1216) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1280) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1280) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 129024]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1408) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1408) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1472) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1472) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1536] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 147456]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1600) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1600) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1664) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1664) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1728] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 165888]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1792) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1792) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1856) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1856) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1920] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 184320]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1984) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1984) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2048) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2048) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2112] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 202752]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2176) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2176) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2240) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2240) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2304] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 221184]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2368) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2368) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2432) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2432) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2496] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 239616]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2560) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2560) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2624) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2624) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 258048]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2752) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2752) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2816) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2816) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2880] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 276480]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2944) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2944) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 3008) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 3008) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[72] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[75] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[78] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[81] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[73] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[76] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[79] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[82] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[74] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[77] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[80] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[83] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[84] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[105] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[126] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[147] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[87] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[108] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[129] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[150] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[90] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[111] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[132] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[153] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[93] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[114] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[135] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[156] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[96] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[117] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[138] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[159] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[99] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[120] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[141] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[162] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[102] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[123] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[144] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[165] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[85] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[106] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[127] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[148] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[88] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[109] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[130] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[151] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[91] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[112] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[133] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[154] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[94] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[115] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[136] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[157] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[97] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[118] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[139] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[160] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[100] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[121] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[142] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[163] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[103] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[124] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[145] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[166] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[86] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[107] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[128] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[149] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[89] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[110] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[131] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[152] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[92] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[113] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[134] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[155] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[95] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[116] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[137] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[158] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[98] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[119] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[140] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[161] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[101] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[122] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[143] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[164] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[104] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[125] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[146] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[167] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[168] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[189] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[210] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[231] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[171] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[192] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[213] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[234] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[174] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[195] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[216] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[237] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[177] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[198] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[219] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[240] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[180] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[201] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[222] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[243] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[183] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[204] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[225] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[246] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[186] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[207] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[228] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[249] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[169] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[190] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[211] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[232] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[172] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[193] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[214] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[235] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[175] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[196] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[217] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[238] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[178] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[199] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[220] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[241] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[181] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[202] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[223] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[244] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[184] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[205] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[226] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[247] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[187] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[208] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[229] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[250] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[170] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[191] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[212] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[233] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[173] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[194] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[215] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[236] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[176] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[197] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[218] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[239] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[179] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[200] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[221] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[242] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[182] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[203] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[224] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[245] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[185] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[206] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[227] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[248] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[188] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[209] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[230] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[251] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[252] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[273] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[294] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[315] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[255] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[276] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[297] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[318] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[258] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[279] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[300] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[321] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[261] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[282] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[303] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[324] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[264] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[285] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[306] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[327] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[267] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[288] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[309] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[330] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[270] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[291] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[312] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[333] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[253] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[274] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[295] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[316] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[256] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[277] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[298] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[319] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[259] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[280] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[301] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[322] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[262] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[283] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[304] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[325] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[265] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[286] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[307] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[328] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[268] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[289] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[310] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[331] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[271] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[292] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[313] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[334] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[254] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[275] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[296] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[317] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[257] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[278] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[299] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[320] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[260] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[281] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[302] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[323] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[263] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[284] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[305] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[326] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[266] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[287] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[308] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[329] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[269] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[290] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[311] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[332] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[272] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[293] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[314] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[335] * kernel_shared_1[threadIdx_x * 48 + 47]
- for i2_inner in range(7):
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 56] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 504]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 112] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 112) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 112) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 168] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 168) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 360]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 224] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 224) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 96) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 280] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 280) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 216]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 336] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 336) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 80) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 392] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 392) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 72]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 448] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 448) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 576]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 504] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 504) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 120) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 560] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 560) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 432]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 616] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 616) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 104) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 672] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 672) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 288]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 728] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 728) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 88) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 784] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 784) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 144]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 840] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 840) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 648]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 896] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 32256]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 952] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 952) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 504]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1008] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1008) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 112) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1064] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1064) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 360]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1120] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1120) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 96) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1176] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1176) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 216]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1232] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1232) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 80) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1288] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1288) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 72]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1344] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1344) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 576]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1400] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1400) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 120) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1456] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1456) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 432]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1512] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1512) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 104) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1568] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1568) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 288]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1624] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1624) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 88) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1680] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1680) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 144]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1736] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1736) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 648]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1792] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 64512]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1848] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1848) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 504]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1904] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1904) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 112) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1960] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1960) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 360]
+ with T.launch_thread(threadIdx_x_1, 56):
+ if T.likely(threadIdx_x_1 < 32):
+ kernel_shared_1[threadIdx_x_1 + 2016] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 2016) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 864]
+ for rc_outer_inner in range(64):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 49] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 49] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 50] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 50] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 51] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 51] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 3] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 3] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 52] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 52] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 4] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 4] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 53] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 53] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 5] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 5] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 6] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 6] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ for i1_inner, i3_inner in T.grid(2, 7):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x // 7 * 3136 + threadIdx_x * 49 + i2_inner * 7 + blockIdx_x % 7] = T.max(conv2d_nchw_1[i2_inner] + bias_1[blockIdx_x // 7 * 64 + threadIdx_x], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x // 7 * 98 + i1_inner * 49 + threadIdx_x % 7 * 7 + i3_inner] = T.max(conv2d_nchw_1[i1_inner * 7 + i3_inner] + bias_1[blockIdx_x * 16 + threadIdx_x // 7 * 2 + i1_inner], T.float32(0))
@@ -765,7 +432,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.360 ms
+ Execution time of this operator: 0.355 ms
@@ -813,35 +480,35 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
- conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+ conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
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=64)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
- conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=7)
- conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
+ conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
+ conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
- conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+ conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
conv2d_nchw_xx_o_o_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=4)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+ conv2d_nchw_rc_o_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=64)
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=3)
+ conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
- compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
- compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+ compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+ compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
- compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+ compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
@@ -862,14 +529,14 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
- s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+ s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 64)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -887,417 +554,108 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[7];
- __shared__ float pad_temp_shared[336];
- __shared__ float kernel_shared[3072];
+ extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[14];
+ __shared__ float pad_temp_shared[6272];
+ __shared__ float kernel_shared[2048];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
+ conv2d_nchw[13] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 4; ++rc_outer_outer) {
for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) && ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) - 8)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 64)] = (((((1 <= ((((((int)threadIdx.x) + 1) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 64) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 1) % 3)) - [...]
- pad_temp_shared[(((int)threadIdx.x) + 128)] = (((((1 <= ((((((int)threadIdx.x) + 2) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 128) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 2) % 3)) [...]
- pad_temp_shared[(((int)threadIdx.x) + 192)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 3) + 1) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) + 440)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 256)] = (((((1 <= ((((((int)threadIdx.x) + 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 256) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 1) % 3)) [...]
- if (((int)threadIdx.x) < 16) {
- pad_temp_shared[(((int)threadIdx.x) + 320)] = (((((((((int)threadIdx.x) + 5) / 3) + ry_outer_outer) < 8) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 320) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 2) % 3)) - 8)] : 0.000000e+00f);
+ for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
+ __syncthreads();
+ for (int ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer = 0; ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer < 112; ++ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) {
+ pad_temp_shared[((ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer * 56) + ((int)threadIdx.x))] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 6272) + (ax0_ax1_fused_ [...]
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 504)];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 112) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 112) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 168) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 360)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 224) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 96) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 280) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 336) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 80) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 392) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 448) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 576)];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 504) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 120) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 560) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 432)];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 616) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 104) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 672) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 288)];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 728) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 88) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 840) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 648)];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 952) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 504)];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1008) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 112) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1064) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 360)];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1120) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 96) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1176) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1232) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 80) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1288) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1344) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 576)];
+ kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1400) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 120) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1456) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 432)];
+ kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1512) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 104) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1568) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 288)];
+ kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1624) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 88) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1680) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1736) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 648)];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1848) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 504)];
+ kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1904) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 112) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1960) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 360)];
+ if (((int)threadIdx.x) < 32) {
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2016) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 864)];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 64; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7))] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 49)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 49)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 50)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 50)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 51)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 51)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 3)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 52)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 52)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 4)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 53)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 53)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 5)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 6)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 55)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 55)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ }
}
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 64) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 64) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 128) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 128) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 18432)];
- kernel_shared[(((((((int)threadIdx.x) + 256) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 256) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 320) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 320) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((((((int)threadIdx.x) + 448) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 448) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 512) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 512) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 55296)];
- kernel_shared[(((((((int)threadIdx.x) + 640) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 640) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 704) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 704) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((((((int)threadIdx.x) + 832) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 832) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 896) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 896) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 92160)];
- kernel_shared[(((((((int)threadIdx.x) + 1024) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1024) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1088) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1088) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((((((int)threadIdx.x) + 1216) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1216) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1280) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1280) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
- kernel_shared[(((((((int)threadIdx.x) + 1408) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1408) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1472) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1472) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((((((int)threadIdx.x) + 1600) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1600) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1664) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1664) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 165888)];
- kernel_shared[(((((((int)threadIdx.x) + 1792) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1792) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1856) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1856) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((((((int)threadIdx.x) + 1984) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1984) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2048) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2048) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 202752)];
- kernel_shared[(((((((int)threadIdx.x) + 2176) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2176) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2240) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2240) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((((((int)threadIdx.x) + 2368) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2368) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2432) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2432) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 239616)];
- kernel_shared[(((((((int)threadIdx.x) + 2560) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2560) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2624) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2624) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((((((int)threadIdx.x) + 2752) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2752) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2816) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2816) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 276480)];
- kernel_shared[(((((((int)threadIdx.x) + 2944) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2944) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 3008) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3008) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[9] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[12] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[18] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[231] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[234] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[216] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[237] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[219] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[240] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[222] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[243] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[225] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[246] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[228] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[249] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[232] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[235] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[217] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[238] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[220] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[241] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[223] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[244] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[226] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[247] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[229] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[250] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[233] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[236] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[218] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[239] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[221] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[242] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[224] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[245] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[227] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[248] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[230] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[251] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[252] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[273] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[294] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[315] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[255] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[276] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[297] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[318] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[258] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[279] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[300] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[321] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[261] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[282] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[303] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[324] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[264] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[285] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[306] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[327] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[267] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[288] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[309] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[330] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[270] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[291] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[312] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[333] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[253] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[274] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[295] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[316] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[256] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[277] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[298] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[319] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[259] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[280] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[301] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[322] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[262] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[283] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[304] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[325] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[265] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[286] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[307] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[328] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[268] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[289] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[310] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[331] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[271] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[292] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[313] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[334] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[254] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[275] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[296] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[317] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[257] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[278] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[299] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[320] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[260] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[281] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[302] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[323] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[263] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[284] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[305] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[326] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[266] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[287] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[308] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[329] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[269] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[290] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[311] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[332] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[272] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[293] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[314] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[335] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
}
}
- for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[(((((((int)blockIdx.x) / 7) * 3136) + (((int)threadIdx.x) * 49)) + (i2_inner * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[(((((int)blockIdx.x) / 7) * 64) + ((int)threadIdx.x))]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
+ compute[(((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[(((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ }
}
}
@@ -1357,7 +715,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 6 minutes 4.967 seconds)
+ **Total running time of the script:** ( 6 minutes 2.638 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 fecde64bcc..f892d27856 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -647,7 +647,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 7.9149 7.9138 7.9185 7.9123 0.0026
+ 7.9235 7.9218 7.9282 7.9203 0.0034
@@ -675,7 +675,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 7.337 seconds)
+ **Total running time of the script:** ( 1 minutes 7.681 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 b75d0ba8f1..3d4f800f8d 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -666,7 +666,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 750.5247 748.0445 756.7973 746.7324 4.4676
+ 759.8578 760.0620 760.0696 759.4420 0.2941
@@ -694,7 +694,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 42.003 seconds)
+ **Total running time of the script:** ( 1 minutes 42.460 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 38d181b4d6..d415ef7f2d 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
@@ -389,12 +389,12 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- for i0_outer_i1_outer_fused in T.parallel(256):
- compute_1 = T.allocate([256], "float32", "global")
- compute_2 = T.Buffer((256,), data=compute_1)
- for i_outer_inner in range(4):
- for i_inner_init in range(4):
- cse_var_1: T.int32 = i_outer_inner * 64 + i_inner_init * 16
+ for i0_outer_i1_outer_fused in T.parallel(16):
+ compute_1 = T.allocate([4096], "float32", "global")
+ compute_2 = T.Buffer((4096,), data=compute_1)
+ for i_outer_inner, nb_j_inner in T.grid(2, 2):
+ for i_inner_init in range(64):
+ cse_var_1: T.int32 = i_outer_inner * 2048 + i_inner_init * 32 + nb_j_inner * 16
compute_2[cse_var_1] = T.float32(0)
compute_2[cse_var_1 + 1] = T.float32(0)
compute_2[cse_var_1 + 2] = T.float32(0)
@@ -411,66 +411,52 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
compute_2[cse_var_1 + 13] = T.float32(0)
compute_2[cse_var_1 + 14] = T.float32(0)
compute_2[cse_var_1 + 15] = T.float32(0)
- for elem_idx, i_inner in T.grid(T.Let(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2], where={cse_var_2: i0_outer_i1_outer_fused % 32}), 4):
+ for elem_idx, i_inner in T.grid(T.Let(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2], where={cse_var_2: i0_outer_i1_outer_fused * 2 + nb_j_inner}), 64):
cse_var_2 = T.int32()
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_3: T.int32 = i0_outer_i1_outer_fused % 32
+ cse_var_21: T.int32 = elem_idx * 16
+ cse_var_20: T.int32 = i0_outer_i1_outer_fused * 2 + nb_j_inner
+ cse_var_19: T.int32 = i_outer_inner * 16384 + i_inner * 256
+ cse_var_18: T.int32 = i_outer_inner * 2048 + i_inner * 32 + nb_j_inner * 16
+ cse_var_17: T.int32 = cse_var_18 + 9
+ cse_var_16: T.int32 = cse_var_18 + 8
+ cse_var_15: T.int32 = cse_var_18 + 7
+ cse_var_14: T.int32 = cse_var_18 + 6
+ cse_var_13: T.int32 = cse_var_18 + 5
+ cse_var_12: T.int32 = cse_var_18 + 4
+ cse_var_11: T.int32 = cse_var_18 + 3
+ cse_var_10: T.int32 = cse_var_18 + 2
+ cse_var_9: T.int32 = cse_var_18 + 15
+ cse_var_8: T.int32 = cse_var_18 + 14
+ cse_var_7: T.int32 = cse_var_18 + 13
+ cse_var_6: T.int32 = cse_var_18 + 12
+ cse_var_5: T.int32 = cse_var_18 + 11
+ cse_var_4: T.int32 = cse_var_18 + 10
+ cse_var_3: T.int32 = cse_var_18 + 1
placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
placeholder_7 = T.Buffer((32768,), data=placeholder.data)
placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_4: T.int32 = i_outer_inner * 64 + i_inner * 16
- compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_5: T.int32 = i_outer_inner * 64 + i_inner * 16 + 1
- compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 1] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_6: T.int32 = i_outer_inner * 64 + i_inner * 16 + 2
- compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 2] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_7: T.int32 = i_outer_inner * 64 + i_inner * 16 + 3
- compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 3] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_8: T.int32 = i_outer_inner * 64 + i_inner * 16 + 4
- compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 4] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_9: T.int32 = i_outer_inner * 64 + i_inner * 16 + 5
- compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 5] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_10: T.int32 = i_outer_inner * 64 + i_inner * 16 + 6
- compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 6] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_11: T.int32 = i_outer_inner * 64 + i_inner * 16 + 7
- compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 7] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_12: T.int32 = i_outer_inner * 64 + i_inner * 16 + 8
- compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 8] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_13: T.int32 = i_outer_inner * 64 + i_inner * 16 + 9
- compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 9] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_14: T.int32 = i_outer_inner * 64 + i_inner * 16 + 10
- compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 10] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_15: T.int32 = i_outer_inner * 64 + i_inner * 16 + 11
- compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 11] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_16: T.int32 = i_outer_inner * 64 + i_inner * 16 + 12
- compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 12] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_17: T.int32 = i_outer_inner * 64 + i_inner * 16 + 13
- compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 13] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_18: T.int32 = i_outer_inner * 64 + i_inner * 16 + 14
- compute_2[cse_var_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 14] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_19: T.int32 = i_outer_inner * 64 + i_inner * 16 + 15
- compute_2[cse_var_19] = compute_2[cse_var_19] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 15] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- for i0_inner in range(16):
- cse_var_20: T.int32 = i0_outer_i1_outer_fused // 32 * 8192 + i0_inner * 512 + i0_outer_i1_outer_fused % 32 * 16
+ compute_2[cse_var_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 1] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 2] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 3] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 4] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 5] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 6] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 7] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 8] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 9] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 10] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 11] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 12] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 13] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 14] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 15] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ for i0_inner in range(128):
+ cse_var_22: T.int32 = i0_inner * 512 + i0_outer_i1_outer_fused * 32
compute_3 = T.Buffer((65536,), data=compute.data)
placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_20:cse_var_20 + 16] = T.max(compute_2[i0_inner * 16:i0_inner * 16 + 16] + placeholder_5[cse_var_20:cse_var_20 + 16], T.Broadcast(T.float32(0), 16))
+ compute_3[cse_var_22:cse_var_22 + 32] = T.max(compute_2[i0_inner * 32:i0_inner * 32 + 32] + placeholder_5[cse_var_22:cse_var_22 + 32], T.Broadcast(T.float32(0), 32))
@@ -520,7 +506,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 2.139 ms
+ Execution time of this operator: 1.871 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 8ce3ecaf52..05cb4d135b 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
=================
-**00:33.006** total execution time for **how_to_tune_with_autotvm** files:
+**00:47.394** 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``) | 00:32.972 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:47.358 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.020 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.022 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``) | 00:00.005 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
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 0416e0039c..85ad69d70d 100644
--- a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
@@ -390,7 +390,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9435978
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4477522
No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -513,7 +513,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 1, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6333769
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5170306
No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -636,9 +636,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8582861
- No: 4 GFLOPS: 103.89/103.89 result: MeasureResult(costs=(0.002228368358490566,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.504410028457642, timestamp=1678686052.69968) [('tile_f', [-1, 16, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,6987874
- No: 5 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2301592
+ No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -760,8 +759,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9866495
- No: 6 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9435764
+ No: 5 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -883,8 +882,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5987806
- No: 7 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 32, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2557412
+ No: 6 GFLOPS: 17.60/17.60 result: MeasureResult(costs=(0.013150684777777777,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.790144443511963, timestamp=1678752180.9416506) [('tile_f', [-1, 2, 2, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2560249
+ No: 7 GFLOPS: 0.00/17.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1006,8 +1006,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3677119
- No: 8 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8628384
+ No: 8 GFLOPS: 0.00/17.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1129,8 +1129,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6613715
- No: 9 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1638929
+ No: 9 GFLOPS: 0.00/17.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1252,8 +1252,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5742039
- No: 10 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4206012
+ No: 10 GFLOPS: 149.26/149.26 result: MeasureResult(costs=(0.0015510194563106796,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5596656799316406, timestamp=1678752182.7189288) [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4883131
+ No: 11 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1375,9 +1376,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 2, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5190827
- No: 11 GFLOPS: 42.24/103.89 result: MeasureResult(costs=(0.00548004262962963,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.892000436782837, timestamp=1678686056.4503279) [('tile_f', [-1, 1, 32, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,656390
- No: 12 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3648223
+ No: 12 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1499,8 +1499,26 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7580699
- No: 13 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 32, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7272364
+ No: 13 GFLOPS: 0.00/149.26 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, 4, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9154477
+ No: 14 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1622,8 +1640,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 256]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8257258
- No: 14 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7382068
+ No: 15 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1745,8 +1763,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2499208
- No: 15 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 32, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9552881
+ No: 16 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1868,9 +1886,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8491318
- No: 16 GFLOPS: 232.07/232.07 result: MeasureResult(costs=(0.0009975514785276072,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9048430919647217, timestamp=1678686058.5952504) [('tile_f', [-1, 1, 8, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7089621
- No: 17 GFLOPS: 0.00/232.07 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6150584
+ No: 17 GFLOPS: 1.31/149.26 result: MeasureResult(costs=(0.177029724,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.406627178192139, timestamp=1678752198.4004164) [('tile_f', [-1, 64, 1, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5616962
+ No: 18 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1992,9 +2010,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 8, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4579057
- No: 18 GFLOPS: 91.97/232.07 result: MeasureResult(costs=(0.0025171067560975607,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.0822193622589111, timestamp=1678686059.8699977) [('tile_f', [-1, 1, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1246310
- No: 19 GFLOPS: 0.00/232.07 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 4, 128]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9865015
+ No: 19 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2116,130 +2133,160 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,636036
- No: 20 GFLOPS: 0.00/232.07 result: Traceback (most recent call last):
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
- func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8895135
+ No: 20 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 742, in __call__
+ yield remote, remote.load_module(os.path.split(build_result.filename)[1])
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 706, in run_through_rpc
+ costs = time_f(*args).results
+ File "/workspace/python/tvm/runtime/module.py", line 357, in evaluator
+ blob = feval(*args)
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/./packed_func.pxi", line 262, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 251, in tvm._ffi._cy3.core.FuncCall3
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
+ 4: TVMFuncCall
at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:402
- 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
- at ../src/driver/driver_api.cc:388
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:283
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../src/runtime/rpc/rpc_module.cc:129
+ 1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1012
+ 0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
+ at ../src/runtime/rpc/rpc_endpoint.cc:804
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 804
+ TVMError:
+ ---------------------------------------------------------------
+ An error occurred during the execution of TVM.
+ For more information, please see: https://tvm.apache.org/docs/errors.html
+ ---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
+
+ During handling of the above exception, another exception occurred:
Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:402
- 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
- at ../src/driver/driver_api.cc:388
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:283
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 706, in run_through_rpc
+ costs = time_f(*args).results
+ File "/usr/lib/python3.7/contextlib.py", line 130, in __exit__
+ self.gen.throw(type, value, traceback)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 746, in __call__
+ remote.remove(build_result.filename)
+ File "/workspace/python/tvm/rpc/client.py", line 144, in remove
+ self._remote_funcs["remove"] = self.get_function("tvm.rpc.server.remove")
+ File "/workspace/python/tvm/rpc/client.py", line 72, in get_function
+ return self._sess.get_function(name)
+ File "/workspace/python/tvm/runtime/module.py", line 171, in get_function
+ self.handle, c_str(name), ctypes.c_int(query_imports), ctypes.byref(ret_handle)
+ File "/workspace/python/tvm/_ffi/base.py", line 348, in check_call
+ raise get_last_ffi_error()
+ tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 52: 0xffffffffffffffff
+ 51: _start
+ 50: __libc_start_main
+ 49: _Py_UnixMain
+ 48: 0x0000000000650da0
+ 47: 0x0000000000650afa
+ 46: _PyFunction_FastCallDict
+ 45: _PyEval_EvalCodeWithName
+ 44: _PyEval_EvalFrameDefault
+ 43: _PyFunction_FastCallKeywords
+ 42: _PyEval_EvalCodeWithName
+ 41: _PyEval_EvalFrameDefault
+ 40: _PyMethodDef_RawFastCallKeywords
+ 39: 0x0000000000546369
+ 38: _PyEval_EvalCodeWithName
+ 37: _PyEval_EvalFrameDefault
+ 36: _PyFunction_FastCallKeywords
+ 35: _PyEval_EvalCodeWithName
+ 34: _PyEval_EvalFrameDefault
+ 33: _PyFunction_FastCallDict
+ 32: _PyEval_EvalCodeWithName
+ 31: _PyEval_EvalFrameDefault
+ 30: _PyObject_FastCallDict
+ 29: 0x00000000004c06e1
+ 28: _PyFunction_FastCallDict
+ 27: _PyEval_EvalFrameDefault
+ 26: _PyMethodDescr_FastCallKeywords
+ 25: 0x00000000005dcb58
+ 24: 0x00000000005dc83f
+ 23: 0x00000000004ba127
+ 22: _PyEval_EvalFrameDefault
+ 21: _PyFunction_FastCallKeywords
+ 20: _PyEval_EvalFrameDefault
+ 19: _PyFunction_FastCallKeywords
+ 18: _PyEval_EvalFrameDefault
+ 17: _PyFunction_FastCallKeywords
+ 16: _PyEval_EvalCodeWithName
+ 15: _PyEval_EvalFrameDefault
+ 14: 0x0000000000537c30
+ 13: _PyObject_FastCallKeywords
+ 12: 0x00007f3eb5baafa2
+ 11: _ctypes_callproc
+ 10: ffi_call
+ 9: ffi_call_unix64
+ 8: TVMModGetFunction
+ at ../src/runtime/c_runtime_api.cc:408
+ 7: tvm::runtime::ModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)
+ at ../src/runtime/module.cc:66
+ 6: tvm::runtime::RPCModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, tvm::runtime::ObjectPtr<tvm::runtime::Object> const&)
+ at ../src/runtime/rpc/rpc_module.cc:185
+ 5: tvm::runtime::RPCClientSession::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1007
+ 4: tvm::runtime::TVMRetValue tvm::runtime::RPCEndpoint::SysCallRemote<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(tvm::runtime::RPCCode, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
+ at ../src/runtime/rpc/rpc_endpoint.h:223
+ 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(int&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const
at ../include/tvm/runtime/packed_func.h:1621
2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
at ../include/tvm/runtime/packed_func.h:1217
1: Call
at ../include/tvm/runtime/packed_func.h:1213
0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 128]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2213632
+ at ../src/runtime/rpc/rpc_endpoint.cc:684
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 684
+ TVMError:
+ ---------------------------------------------------------------
+ An error occurred during the execution of TVM.
+ For more information, please see: https://tvm.apache.org/docs/errors.html
+ ---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=1
+
+ Traceback (most recent call last):
+ 52: 0xffffffffffffffff
+ 51: _start
+ 50: __libc_start_main
+ 49: _Py_UnixMain
+ 48: 0x0000000000650da0
+ 47: 0x0000000000650afa
+ 46: _PyFunction_FastCallDict
+ 45: _PyEval_EvalCodeWithName
+ 44: _PyEval_EvalFrameDefault
+ 43: _PyFunction_FastCallKeywords
+ 42: _PyEval_EvalCodeWithName
+ 41: _PyEval_EvalFrameDefault
+ 40: _PyMethodDef_RawFastCallKeywords
+ 39: 0x0000000000546369
+ 38: _PyEval_EvalCodeWithName
+ 37: _PyEval_EvalFrameDefault
+ 36: _PyFunction_FastCallKeywords
+ 35: _PyEval_EvalCodeWithName
+ 34: _PyEval_EvalFrameDefault
+ 33: _PyFunction_FastCallDict
+ 32: _PyEval_EvalCodeWithName
+ 31: _PyEval_EvalFrameDefault
+ 30: _PyObject_FastCallDict
+ 29: 0x00000000004c06e1
+ 28: _PyFunction_FastCallDict
+ 27: _PyEval_EvalFrameDefault
+ 26: _PyMethodDescr_FastCallKeywords
+ 25: 0x00000000005dcb58
+ 24: 0x00000000005dc83f
+ 23: 0x00000000004ba127
+ 22: _PyEval_EvalFrameDefault
+ 21: _PyFunction_FastCallKeywords
+ 20: _PyEval_EvalFrameDefault
+ 19: _PyFunction_FastCall [('tile_f', [-1, 8, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3105708
@@ -2294,9 +2341,9 @@ and measure running time.
Finish loading 20 records
Best config:
- [('tile_f', [-1, 1, 8, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7089621
+ [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4883131
Finish loading 20 records
- Time cost of this operator: 0.001352
+ Time cost of this operator: 0.001962
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 8f4b2760f7..18f981a60d 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -360,10 +360,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 374.6 98.916 (1, 2, 10, 10, 3) 2 1 [374.6]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.114 0.822 (1, 6, 10, 10) 1 1 [3.114]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.992 0.262 (1, 1, 10, 10, 3) 1 1 [0.992]
- Total_time - 378.706 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 313.8 98.742 (1, 2, 10, 10, 3) 2 1 [313.8]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.041 0.957 (1, 6, 10, 10) 1 1 [3.041]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.958 0.301 (1, 1, 10, 10, 3) 1 1 [0.958]
+ Total_time - 317.799 - - - - -
@@ -428,10 +428,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.4 97.358 (1, 6, 10, 10, 1) 2 1 [100.4]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.762 1.709 (1, 6, 10, 10) 1 1 [1.762]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.962 0.933 (1, 1, 10, 10, 3) 1 1 [0.962]
- Total_time - 103.124 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 102.6 97.423 (1, 6, 10, 10, 1) 2 1 [102.6]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.745 1.657 (1, 6, 10, 10) 1 1 [1.745]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 0.92 (1, 1, 10, 10, 3) 1 1 [0.969]
+ Total_time - 105.314 - - - - -
@@ -439,7 +439,7 @@ Timing the tuned program
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 21.943 seconds)
+ **Total running time of the script:** ( 1 minutes 20.989 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_autotune.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
index 09eb1e7d8e..e0434f1239 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
@@ -118,7 +118,7 @@ download a cat image and preprocess it to use as the model input.
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/ao/quantization/utils.py:281: UserWarning: must run observer before calling calculate_qparams. Returning default values.
"must run observer before calling calculate_qparams. " +
Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
-
0%| | 0.00/3.42M [00:00<?, ?B/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 43.3MB/s]
+
0%| | 0.00/3.42M [00:00<?, ?B/s]
100%|##########| 3.42M/3.42M [00:00<00:00, 40.4MB/s]
/workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
return LooseVersion(torch_ver) > ver
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -324,7 +324,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 18.283 seconds)
+ **Total running time of the script:** ( 1 minutes 18.623 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_pytorch.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
index 520550dfac..f6a2fcab0b 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
@@ -218,7 +218,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmpm32a5n_m/images/random'
+ '/tmp/tmp03odc2we/images/random'
@@ -309,7 +309,7 @@ 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], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0]
+ :alt: [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]
:srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
:class: sphx-glr-single-img
@@ -318,8 +318,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmpm32a5n_m/images/target contains 8144 images
- /tmp/tmpm32a5n_m/images/random contains 5000 images
+ /tmp/tmp03odc2we/images/target contains 8144 images
+ /tmp/tmp03odc2we/images/random contains 5000 images
@@ -494,13 +494,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 46s - loss: 0.2104 - accuracy: 0.9252 - val_loss: 0.1579 - val_accuracy: 0.9494 - 46s/epoch - 141ms/step
+ 328/328 - 47s - loss: 0.2161 - accuracy: 0.9268 - val_loss: 0.1325 - val_accuracy: 0.9539 - 47s/epoch - 144ms/step
Epoch 2/3
- 328/328 - 43s - loss: 0.0956 - accuracy: 0.9635 - val_loss: 0.1417 - val_accuracy: 0.9566 - 43s/epoch - 131ms/step
+ 328/328 - 43s - loss: 0.1036 - accuracy: 0.9620 - val_loss: 0.1488 - val_accuracy: 0.9513 - 43s/epoch - 132ms/step
Epoch 3/3
- 328/328 - 43s - loss: 0.0641 - accuracy: 0.9759 - val_loss: 0.1162 - val_accuracy: 0.9622 - 43s/epoch - 131ms/step
+ 328/328 - 43s - loss: 0.0646 - accuracy: 0.9777 - val_loss: 0.1283 - val_accuracy: 0.9588 - 43s/epoch - 132ms/step
- <keras.callbacks.History object at 0x7f16b577bc90>
+ <keras.callbacks.History object at 0x7f5a9577ed10>
@@ -861,7 +861,7 @@ 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 18.505 seconds)
+ **Total running time of the script:** ( 4 minutes 46.174 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 5e49ac1285..b7f2ce174e 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
@@ -5,18 +5,18 @@
Computation times
=================
-**07:16.046** total execution time for **how_to_work_with_microtvm** files:
+**07:43.444** total execution time for **how_to_work_with_microtvm** files:
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:18.505 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:46.174 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:21.943 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 01:20.989 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:18.283 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``) | 01:18.623 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.076 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:10.297 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.239 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:07.361 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``) | 00:00.000 | 0.0 MB |
+-----------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index aa034c5ba4..86f73684d5 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
=================
-**00:45.562** total execution time for **how_to_work_with_relay** files:
+**00:45.955** 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``) | 00:33.531 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:33.652 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:10.369 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:10.463 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.656 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.834 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.006 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
index 292afcdde1..ce67da769a 100644
--- a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
@@ -264,7 +264,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7f1563938d40>
+ <function my_cuda_math_rule at 0x7f5934085d40>
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 8dde14a1a1..ef2509b20a 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
@@ -5,18 +5,18 @@
Computation times
=================
-**00:07.402** total execution time for **how_to_work_with_schedules** files:
+**00:09.103** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:04.893 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:06.511 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.151 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.228 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.573 | 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``) | 00:00.555 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.556 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.119 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.117 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.052 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 88fc3deb1e..3b2a8a1c1e 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
=================
-**00:31.022** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:31.463** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:31.016 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:31.456 | 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 9833c77e53..98cd51c252 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -293,7 +293,7 @@ The compilation steps are:
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
- resnet18_v1 inference graph built in 32.96s!
+ resnet18_v1 inference graph built in 34.04s!
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 c23ba9e1c2..2eef398f3b 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -337,7 +337,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 22.67s!
+ yolov3-tiny inference graph built in 22.89s!
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 29f93e90c1..7b9b066912 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**01:39.194** total execution time for **topic_vta_tutorials_frontend** files:
+**01:40.731** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:49.715 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:50.884 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:49.479 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:49.848 | 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 8c6c9bc627..ee94826582 100644
--- a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:03.153** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.183** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.694 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.725 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.459 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.458 | 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 0a0812246a..96ae754138 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
=================
-**00:00.756** total execution time for **topic_vta_tutorials** files:
+**00:00.766** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.392 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.397 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.364 | 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 15bb621a18..e843b489b5 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -318,7 +318,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 98.147 ms
+ Execution time of this operator: 95.234 ms
@@ -416,7 +416,7 @@ resume the status and do more 5 trials.
.. code-block:: none
Resume search:
- .T
+
@@ -434,7 +434,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 30.840 seconds)
+ **Total running time of the script:** ( 1 minutes 22.267 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 318de38e20..da8e9cebc1 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -454,16 +454,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 3.09/3.09 result: MeasureResult(costs=(0.0867567328,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6469478607177734, timestamp=1678684460.3844144) [('tile_y', [-1, 256]), ('tile_x', [-1, 8])],None,38
- No: 2 GFLOPS: 2.79/3.09 result: MeasureResult(costs=(0.09607420879999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8033630847930908, timestamp=1678684463.419324) [('tile_y', [-1, 2]), ('tile_x', [-1, 16])],None,41
- No: 3 GFLOPS: 11.51/11.51 result: MeasureResult(costs=(0.0233286196,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6122231483459473, timestamp=1678684464.0537403) [('tile_y', [-1, 4]), ('tile_x', [-1, 512])],None,92
- No: 4 GFLOPS: 8.81/11.51 result: MeasureResult(costs=(0.0304696042,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7071022987365723, timestamp=1678684466.0268316) [('tile_y', [-1, 1]), ('tile_x', [-1, 32])],None,50
- No: 5 GFLOPS: 1.36/11.51 result: MeasureResult(costs=(0.19755499,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.389230489730835, timestamp=1678684469.6100347) [('tile_y', [-1, 1]), ('tile_x', [-1, 2])],None,10
- No: 6 GFLOPS: 1.52/11.51 result: MeasureResult(costs=(0.17613902080000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.0493390560150146, timestamp=1678684473.9333296) [('tile_y', [-1, 1]), ('tile_x', [-1, 1])],None,0
- No: 7 GFLOPS: 14.42/14.42 result: MeasureResult(costs=(0.0186158342,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5975933074951172, timestamp=1678684474.4855804) [('tile_y', [-1, 32]), ('tile_x', [-1, 64])],None,65
- No: 8 GFLOPS: 2.34/14.42 result: MeasureResult(costs=(0.11476816100000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0721683502197266, timestamp=1678684476.583552) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
- No: 9 GFLOPS: 0.51/14.42 result: MeasureResult(costs=(0.5273584632,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.644187450408936, timestamp=1678684485.342885) [('tile_y', [-1, 128]), ('tile_x', [-1, 1])],None,7
- No: 10 GFLOPS: 2.13/14.42 result: MeasureResult(costs=(0.1262303024,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2271721363067627, timestamp=1678684487.6201863) [('tile_y', [-1, 2]), ('tile_x', [-1, 4])],None,21
+ No: 1 GFLOPS: 4.50/4.50 result: MeasureResult(costs=(0.05967109,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2330536842346191, timestamp=1678750593.585393) [('tile_y', [-1, 16]), ('tile_x', [-1, 16])],None,44
+ No: 2 GFLOPS: 11.28/11.28 result: MeasureResult(costs=(0.0237999648,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6429798603057861, timestamp=1678750595.485118) [('tile_y', [-1, 256]), ('tile_x', [-1, 256])],None,88
+ No: 3 GFLOPS: 0.87/11.28 result: MeasureResult(costs=(0.3071478688,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.141611099243164, timestamp=1678750600.6574879) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
+ No: 4 GFLOPS: 11.02/11.28 result: MeasureResult(costs=(0.0243695036,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6250784397125244, timestamp=1678750602.567924) [('tile_y', [-1, 2]), ('tile_x', [-1, 512])],None,91
+ No: 5 GFLOPS: 10.51/11.28 result: MeasureResult(costs=(0.025530940600000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6833343505859375, timestamp=1678750603.364809) [('tile_y', [-1, 8]), ('tile_x', [-1, 32])],None,53
+ No: 6 GFLOPS: 1.87/11.28 result: MeasureResult(costs=(0.14336039579999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.532989501953125, timestamp=1678750607.1755967) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
+ No: 7 GFLOPS: 11.46/11.46 result: MeasureResult(costs=(0.0234165186,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6113460063934326, timestamp=1678750607.8100495) [('tile_y', [-1, 256]), ('tile_x', [-1, 512])],None,98
+ No: 8 GFLOPS: 2.83/11.46 result: MeasureResult(costs=(0.0948549296,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7708613872528076, timestamp=1678750609.5896919) [('tile_y', [-1, 512]), ('tile_x', [-1, 16])],None,49
+ No: 9 GFLOPS: 10.18/11.46 result: MeasureResult(costs=(0.026369689599999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7020399570465088, timestamp=1678750610.4075158) [('tile_y', [-1, 512]), ('tile_x', [-1, 128])],None,79
+ No: 10 GFLOPS: 3.97/11.46 result: MeasureResult(costs=(0.06769754219999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3485820293426514, timestamp=1678750611.752353) [('tile_y', [-1, 64]), ('tile_x', [-1, 16])],None,46
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 5d8238879c..6e754ba1ac 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -311,7 +311,7 @@ standard deviation.
.. code-block:: none
- {'mean': 511.0277086200017, 'median': 510.5724576499995, 'std': 1.9596086703937687}
+ {'mean': 550.6390348200011, 'median': 546.4105448499993, 'std': 15.583710586461082}
@@ -545,30 +545,31 @@ 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: 16.06/ 23.55 GFLOPS | Progress: (4/20) | 10.56 s
[Task 1/25] Current/Best: 14.89/ 23.55 GFLOPS | Progress: (8/20) | 14.52 s
[Task 1/25] Current/Best: 11.12/ 23.55 GFLOPS | Progress: (12/20) | 16.52 s
[Task 1/25] Current/Best: 7.11/ 23.55 GFLOPS | Progress: (16/20) | 20.48 s
[Task 1/25] Current/Best: 12.93/ 23.55 GFLOPS | Progress: (20/20) | 23.18 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 14.74/ 15.05 GFLOPS | Progress: (4/20) | 4.72 s
[Task 2/25] Current/Best: 12.25/ 20.51 GFLOPS | Progress: (8/20) | 6.17 s
[Task 2/25] Current/Best: 12.09/ 20.51 GFLOPS | Progress: (12/20) | 7.83 s
[Task 2/25] Current/Best: 14.55/ 20.51 GFLOPS | Progress: (16/20) | 9.49 s
[Task 2/25] Current/Best: 11.69/ 21.86 GFLOPS | Progress: (20/20) | 11.46 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 16.56/ 20.38 GFLOPS | Progress: (4/20) | 5.02 s
[Task 3/25] Current/Best: 5.63/ 20.38 GFLOPS | Progress: (8/20) | 7.59 s
[Task 3/25] Current/Best: 11.59/ 23.04 GFLOPS | Progress: (12/20) | 11.15 s
[Task 3/25] Current/Best: 11.36/ 23.04 GFLOPS | Progress: (16/20) | 13.14 s
[Task 3/25] Current/Best: 11.88/ 23.04 GFLOPS | Progress: (20/20) | 15.53 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 14.00/ 14.33 GFLOPS | Progress: (4/20) | 6.20 s
[Task 4/25] Current/Best: 16.36/ 18.17 GFLOPS | Progress: (8/20) | 9.07 s
[Task 4/25] Current/Best: 17.88/ 18.22 GFLOPS | Progress: (12/20) | 10.78 s
[Task 4/25] Current/Best: 11.35/ 20.80 GFLOPS | Progress: (16/20) | 13.55 s
[Task 4/25] Current/Best: 5.37/ 20.80 GFLOPS | Progress: (20/20) | 16.14 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 14.35/ 14.35 GFLOPS | Progress: (4/20) | 4.77 s
[Task 5/25] Current/Best: 5.38/ 14.35 GFLOPS | Progress: (8/20) | 7.03 s
[Task 5/25] Current/Best: 14.31/ 14.55 GFLOPS | Progress: (12/20) | 9.46 s
[Task 5/25] Current/Best: 11.29/ 14.55 GFLOPS | Progress: (16/20) | 13.15 s
[Task 5/25] Current/Best: 5.36/ 14.55 GFLOPS | Progress: (20/20) | 16.08 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 14.54/ 14.54 GFLOPS | Progress: (4/20) | 5.87 s
[Task 6/25] Current/Best: 15.63/ 17.99 GFLOPS | Progress: (8/20) | 7.99 s
[Task 6/25] Current/Best: 4.13/ 18.47 GFLOPS | Progress: (12/20) | 11.01 s
[Task 6/25] Current/Best: 19.82/ 22.01 GFLOPS | Progress: (16/20) | 15.05 s
[Task 6/25] Current/Best: 15.13/ 22.01 GFLOPS | Progress: (20/20) | 17.07 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 12.65/ 17.95 GFLOPS | Progress: (4/20) | 5.27 s
[Task 7/25] Current/Best: 20.43/ 20.43 GFLOPS | Progress: (8/20) | 7.15 s
[Task 7/25] Current/Best: 4.70/ 20.43 GFLOPS | Progress: (12/20) | 10.03 s
[Task 7/25] Current/Best: 6.45/ 20.43 GFLOPS | Progress: (16/20) | 12.35 s
[Task 7/25] Current/Best: 19.63/ 20.43 GFLOPS | Progress: (20/20) | 14.76 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 12.53/ 18.50 GFLOPS | Progress: (4/20) | 5.03 s
[Task 8/25] Current/Best: 13.68/ 18.50 GFLOPS | Progress: (8/20) | 8.29 s
[Task 8/25] Current/Best: 10.80/ 18.50 GFLOPS | Progress: (12/20) | 15.61 s
[Task 8/25] Current/Best: 3.30/ 18.50 GFLOPS | Progress: (16/20) | 19.57 s
[Task 8/25] Current/Best: 12.06/ 18.50 GFLOPS | Progress: (20/20) | 25.41 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 10.55/ 17.08 GFLOPS | Progress: (4/20) | 5.24 s
[Task 9/25] Current/Best: 11.86/ 18.35 GFLOPS | Progress: (8/20) | 8.32 s
[Task 9/25] Current/Best: 18.11/ 18.35 GFLOPS | Progress: (12/20) | 11.17 s
[Task 9/25] Current/Best: 12.59/ 18.35 GFLOPS | Progress: (16/20) | 14.03 s
[Task 9/25] Current/Best: 13.13/ 18.35 GFLOPS | Progress: (20/20) | 16.88 s Done.
-
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 14.61/ 19.91 GFLOPS | Progress: (4/20) | 4.70 s
[Task 10/25] Current/Best: 13.49/ 20.88 GFLOPS | Progress: (8/20) | 7.63 s
[Task 10/25] Current/Best: 20.69/ 20.88 GFLOPS | Progress: (12/20) | 10.77 s
[Task 10/25] Current/Best: 16.03/ 20.88 GFLOPS | Progress: (16/20) | 13.74 s
[Task 10/25] Current/Best: 11.78/ 20.88 GFLOPS | Progress: (20/20) | 17.26 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 22.59/ 23.73 GFLOPS | Progress: (4/20) | 5.68 s
[Task 11/25] Current/Best: 12.70/ 23.73 GFLOPS | Progress: (8/20) | 8.40 s
[Task 11/25] Current/Best: 18.17/ 23.73 GFLOPS | Progress: (12/20) | 11.53 s
[Task 11/25] Current/Best: 16.20/ 23.73 GFLOPS | Progress: (16/20) | 13.89 s
[Task 11/25] Current/Best: 21.83/ 23.73 GFLOPS | Progress: (20/20) | 16.19 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 15.56/ 15.56 GFLOPS | Progress: (4/20) | 5.85 s
[Task 12/25] Current/Best: 3.14/ 15.56 GFLOPS | Progress: (8/20) | 8.81 s
[Task 12/25] Current/Best: 14.36/ 15.56 GFLOPS | Progress: (12/20) | 11.21 s
[Task 12/25] Current/Best: 18.27/ 18.27 GFLOPS | Progress: (16/20) | 13.80 s
[Task 12/25] Current/Best: 8.54/ 18.27 GFLOPS | Progress: (20/20) | 16.84 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 10.22/ 10.22 GFLOPS | Progress: (4/20) | 6.84 s
[Task 13/25] Current/Best: 6.03/ 17.34 GFLOPS | Progress: (8/20) | 10.41 s
[Task 13/25] Current/Best: 13.53/ 19.00 GFLOPS | Progress: (12/20) | 12.87 s
[Task 13/25] Current/Best: 3.11/ 19.00 GFLOPS | Progress: (16/20) | 16.89 s
[Task 13/25] Current/Best: 16.11/ 19.00 GFLOPS | Progress: (20/20) | 20.64 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 1.59/ 7.35 GFLOPS | Progress: (4/20) | 11.53 s
[Task 14/25] Current/Best: 8.32/ 17.03 GFLOPS | Progress: (8/20) | 15.22 s
[Task 14/25] Current/Best: 3.41/ 20.82 GFLOPS | Progress: (12/20) | 19.55 s
[Task 14/25] Current/Best: 13.08/ 20.82 GFLOPS | Progress: (16/20) | 22.50 s
[Task 14/25] Current/Best: 9.52/ 20.82 GFLOPS | Progress: (20/20) | 28.08 s Done.
-
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 7.15/ 20.93 GFLOPS | Progress: (4/20) | 6.34 s
[Task 15/25] Current/Best: 8.58/ 20.93 GFLOPS | Progress: (8/20) | 13.51 s
[Task 15/25] Current/Best: 14.29/ 20.93 GFLOPS | Progress: (12/20) | 20.92 s
[Task 15/25] Current/Best: 11.68/ 22.91 GFLOPS | Progress: (16/20) | 22.50 s
[Task 15/25] Current/Best: 3.13/ 22.91 GFLOPS | Progress: (20/20) | 25.76 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 10.49/ 20.41 GFLOPS | Progress: (4/20) | 4.92 s
[Task 16/25] Current/Best: 19.04/ 20.85 GFLOPS | Progress: (8/20) | 6.39 s
[Task 16/25] Current/Best: 9.47/ 20.85 GFLOPS | Progress: (12/20) | 10.64 s
[Task 16/25] Current/Best: 13.69/ 20.85 GFLOPS | Progress: (16/20) | 13.35 s
[Task 16/25] Current/Best: 10.03/ 20.85 GFLOPS | Progress: (20/2
0) | 16.72 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 18.15/ 18.15 GFLOPS | Progress: (4/20) | 5.64 s
[Task 17/25] Current/Best: 9.65/ 20.76 GFLOPS | Progress: (8/20) | 8.15 s
[Task 17/25] Current/Best: 11.67/ 20.76 GFLOPS | Progress: (12/20) | 10.75 s
[Task 17/25] Current/Best: 5.13/ 20.76 GFLOPS | Progress: (16/20) | 13.72 s
[Task 17/25] Current/Best: 20.70/ 21.09 GFLOPS | Progress: (20/20) | 16.44 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 12.05/ 19.12 GFLOPS | Progress: (4/20) | 6.98 s
[Task 18/25] Current/Best: 9.50/ 19.12 GFLOPS | Progress: (8/20) | 15.46 s
[Task 18/25] Current/Best: 13.79/ 19.12 GFLOPS | Progress: (12/20) | 19.40 s
[Task 18/25] Current/Best: 1.57/ 19.12 GFLOPS | Progress: (16/20) | 23.62 s
[Task 18/25] Current/Best: 4.15/ 19.12 GFLOPS | Progress: (20/20) | 28.60 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 2.69/ 20.80 GFLOPS | Progress: (4/20) | 5.83 s
[Task 19/25] Current/Best: 9.28/ 20.80 GFLOPS | Progress: (8/20) | 8.67 s
[Task 19/25] Current/Best: 10.39/ 20.80 GFLOPS | Progress: (12/20) | 13.58 s
[Task 19/25] Current/Best: 15.65/ 20.80 GFLOPS | Progress: (16/20) | 17.39 s
[Task 19/25] Current/Best: 9.44/ 20.80 GFLOPS | Progress: (20/20) | 20.29 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 8.12/ 15.41 GFLOPS | Progress: (4/20) | 5.33 s
[Task 20/25] Current/Best: 5.97/ 17.38 GFLOPS | Progress: (8/20) | 7.66 s
[Task 20/25] Current/Best: 7.18/ 18.13 GFLOPS | Progress: (12/20) | 10.37 s
[Task 20/25] Current/Best: 9.94/ 18.13 GFLOPS | Progress: (16/20) | 14.70 s
[Task 20/25] Current/Best: 12.07/ 18.13 GFLOPS | Progress: (20/20) | 17.05 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.95/ 16.08 GFLOPS | Progress: (4/20) | 5.68 s
[Task 21/25] Current/Best: 19.38/ 20.64 GFLOPS | Progress: (8/20) | 7.19 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 6.88/ 13.97 GFLOPS | Progress: (4/20) | 14.98 s
[Task 1/25] Current/Best: 14.11/ 14.18 GFLOPS | Progress: (8/20) | 20.61 s
[Task 1/25] Current/Best: 3.36/ 20.97 GFLOPS | Progress: (12/20) | 23.58 s
[Task 1/25] Current/Best: 9.26/ 20.97 GFLOPS | Progress: (16/20) | 26.47 s
[Task 1/25] Current/Best: 9.01/ 20.97 GFLOPS | Progress: (20/20) | 30.34 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 5.12/ 13.63 GFLOPS | Progress: (4/20) | 5.08 s
[Task 2/25] Current/Best: 8.47/ 16.78 GFLOPS | Progress: (8/20) | 7.01 s
[Task 2/25] Current/Best: 6.53/ 16.78 GFLOPS | Progress: (12/20) | 8.74 s
[Task 2/25] Current/Best: 7.92/ 16.78 GFLOPS | Progress: (16/20) | 10.77 s
[Task 2/25] Current/Best: 10.26/ 16.78 GFLOPS | Progress: (20/20) | 12.78 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 11.41/ 23.22 GFLOPS | Progress: (4/20) | 6.79 s
[Task 3/25] Current/Best: 12.29/ 23.22 GFLOPS | Progress: (8/20) | 9.52 s
[Task 3/25] Current/Best: 5.93/ 23.22 GFLOPS | Progress: (12/20) | 12.03 s
[Task 3/25] Current/Best: 13.69/ 23.22 GFLOPS | Progress: (16/20) | 14.40 s
[Task 3/25] Current/Best: 6.19/ 23.22 GFLOPS | Progress: (20/20) | 17.10 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 17.38/ 17.38 GFLOPS | Progress: (4/20) | 5.04 s
[Task 4/25] Current/Best: 12.93/ 17.38 GFLOPS | Progress: (8/20) | 6.82 s
[Task 4/25] Current/Best: 11.67/ 17.38 GFLOPS | Progress: (12/20) | 17.96 s
[Task 4/25] Current/Best: 9.96/ 17.38 GFLOPS | Progress: (16/20) | 23.66 s
[Task 4/25] Current/Best: 7.27/ 17.38 GFLOPS | Progress: (20/20) | 25.91 s
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 16.16/ 16.39 GFLOPS | Progress: (4/20) | 4.89 s
[Task 5/25] Current/Best: 9.60/ 16.39 GFLOPS | Progress: (8/20) | 7.06 s
[Task 5/25] Current/Best: 16.30/ 17.47 GFLOPS | Progress: (12/20) | 8.90 s
[Task 5/25] Current/Best: 13.61/ 23.02 GFLOPS | Progress: (16/20) | 10.98 s
[Task 5/25] Current/Best: 5.29/ 23.02 GFLOPS | Progress: (20/20)
| 13.64 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 3.26/ 11.22 GFLOPS | Progress: (4/20) | 7.49 s Done.
+
[Task 6/25] Current/Best: 14.05/ 14.18 GFLOPS | Progress: (8/20) | 11.09 s
[Task 6/25] Current/Best: 3.47/ 14.18 GFLOPS | Progress: (12/20) | 14.61 s
[Task 6/25] Current/Best: 15.69/ 15.69 GFLOPS | Progress: (16/20) | 18.25 s
[Task 6/25] Current/Best: 6.84/ 18.08 GFLOPS | Progress: (20/20) | 20.57 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 6.34/ 11.67 GFLOPS | Progress: (4/20) | 5.58 s
[Task 7/25] Current/Best: 12.30/ 14.04 GFLOPS | Progress: (8/20) | 9.25 s
[Task 7/25] Current/Best: 1.59/ 14.04 GFLOPS | Progress: (12/20) | 13.39 s
[Task 7/25] Current/Best: 5.58/ 19.05 GFLOPS | Progress: (16/20) | 16.36 s
[Task 7/25] Current/Best: 14.41/ 19.53 GFLOPS | Progress: (20/20) | 18.55 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 8.34/ 14.87 GFLOPS | Progress: (4/20) | 13.94 s
[Task 8/25] Current/Best: 12.72/ 14.87 GFLOPS | Progress: (8/20) | 16.67 s
[Task 8/25] Current/Best: 15.83/ 15.83 GFLOPS | Progress: (12/20) | 19.11 s
[Task 8/25] Current/Best: 12.36/ 17.49 GFLOPS | Progress: (16/20) | 22.31 s
[Task 8/25] Current/Best: 3.39/ 17.49 GFLOPS | Progress: (20/20) | 29.62 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 9.24/ 20.66 GFLOPS | Progress: (4/20) | 8.00 s
[Task 9/25] Current/Best: 11.95/ 20.66 GFLOPS | Progress: (8/20) | 10.48 s
[Task 9/25] Current/Best: 8.18/ 20.78 GFLOPS | Progress: (12/20) | 12.18 s
[Task 9/25] Current/Best: 10.44/ 20.78 GFLOPS | Progress: (16/20) | 14.18 s
[Task 9/25] Current/Best: 18.30/ 20.78 GFLOPS | Progress: (20/20) | 16.21 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 13.73/ 15.29 GFLOPS | Progress: (4/20) | 5.02 s
[Task 10/25] Current/Best: 4.42/ 15.29 GFLOPS | Progress: (8/20) | 8.77 s
[Task 10/25] Current/Best: 17.72/ 22.73 GFLOPS | Progress: (12/20) | 10.84 s
[Task 10/25] Current/Best: 11.84/ 22.73 GFLOPS | Progress: (16/20) | 12.57 s
[Task 10/25] Current/Best: 15.79/ 22.73 GFLOPS | Progress: (20/20) | 15.73 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 7.79/ 20.71 GFLOPS | Progress: (4/20) | 5.53 s
[Task 11/25] Current/Best: 18.20/ 22.65 GFLOPS | Progress: (8/20) | 7.68 s
[Task 11/25] Current/Best: 11.50/ 22.65 GFLOPS | Progress: (12/20) | 10.79 s
[Task 11/25] Current/Best: 9.88/ 22.65 GFLOPS | Progress: (16/20) | 14.32 s
[Task 11/25] Current/Best: 14.43/ 22.65 GFLOPS | Progress: (20/20) | 16.61 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 9.84/ 13.82 GFLOPS | Progress: (4/20) | 6.97 s
[Task 12/25] Current/Best: 16.00/ 18.49 GFLOPS | Progress: (8/20) | 9.26 s
[Task 12/25] Current/Best: 13.35/ 18.49 GFLOPS | Progress: (12/20) | 11.73 s
[Task 12/25] Current/Best: 9.70/ 19.01 GFLOPS | Progress: (16/20) | 15.57 s
[Task 12/25] Current/Best: 15.77/ 19.01 GFLOPS | Progress: (20/20) | 17.75 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 17.55/ 19.46 GFLOPS | Progress: (4/20) | 5.76 s
[Task 13/25] Current/Best: 14.97/ 19.46 GFLOPS | Progress: (8/20) | 7.95 s
[Task 13/25] Current/Best: 10.14/ 19.46 GFLOPS | Progress: (12/20) | 10.41 s
[Task 13/25] Current/Best: 16.42/ 19.46 GFLOPS | Progress: (16/20) | 14.32 s
[Task 13/25] Current/Best: 3.12/ 19.46 GFLOPS | Progress: (20/20) | 17.90 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 16.10/ 16.10 GFLOPS | Progress: (4/20) | 5.02 s
[Task 14/25] Current/Best: 7.35/ 16.10 GFLOPS | Progress: (8/20) | 8.90 s
[Task 14/25] Current/Best: 14.16/ 16.10 GFLOPS | Progress: (12/20) | 11.76 s
[Task 14/25] Current/Best: 5.42/ 16.10 GFLOPS | Progress: (16/20) | 14.35 s
[Task 14/25] Current/Best: 9.46/ 16.10 GFLOPS | Progress: (20/20) | 19.48 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 11.37/ 19.11 GFLOPS | Progress: (4/20) | 4.68 s
[Task 15/25] Current/Best: 7.13/ 21.25 GFLOPS | Progress: (8/20) | 9.19 s
[Task 15/25] Current/Best: 17.06/ 21.25 GFLOPS | Progress: (12/20) | 12.28 s Done.
+
[Task 15/25] Current/Best: 9.16/ 21.25 GFLOPS | Progress: (16/20) | 18.47 s
[Task 15/25] Current/Best: 18.32/ 21.25 GFLOPS | Progress: (20/20) | 22.25 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 15.25/ 21.15 GFLOPS | Progress: (4/20) | 4.70 s
[Task 16/25] Current/Best: 16.37/ 21.15 GFLOPS | Progress: (8/20) | 8.12 s
[Task 16/25] Current/Best: 9.39/ 21.15 GFLOPS | Progress: (12/20) | 9.76 s
[Task 16/25] Current/Best: 15.13/ 21.15 GFLOPS | Progress: (16/20) | 11.67 s
[Task 16/25] Current/Best: 13.02/ 21.15 GFLOPS | Progress: (20/20) | 13.19 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 20.24/ 20.24 GFLOPS | Progress: (4/20) | 5.76 s
[Task 17/25] Current/Best: 9.46/ 20.24 GFLOPS | Progress: (8/20) | 8.92 s
[Task 17/25] Current/Best: 11.87/ 20.24 GFLOPS | Progress: (12/20) | 11.96 s
[Task 17/25] Current/Best: 20.62/ 20.62 GFLOPS | Progress: (16/20) | 15.99 s
[Task 17/25] Current/Best: 17.95/ 21.00 GFLOPS | Progress: (20/20) | 18.82 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 2.88/ 18.16 GFLOPS | Progress: (4/20) | 5.46 s
[Task 18/25] Current/Best: 13.46/ 18.16 GFLOPS | Progress: (8/20) | 8.59 s
[Task 18/25] Current/Best: 15.13/ 18.16 GFLOPS | Progress: (12/20) | 12.61 s
[Task 18/25] Current/Best: 10.27/ 18.16 GFLOPS | Progress: (16/20) | 16.18 s
[Task 18/25] Current/Best: 9.13/ 20.35 GFLOPS | Progress: (20/20) | 22.01 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 10.37/ 18.57 GFLOPS | Progress: (4/20) | 5.56 s
[Task 19/25] Current/Best: 4.44/ 18.57 GFLOPS | Progress: (8/20) | 13.41 s
[Task 19/25] Current/Best: 13.86/ 21.45 GFLOPS | Progress: (12/20) | 16.75 s
[Task 19/25] Current/Best: 10.81/ 21.45 GFLOPS | Progress: (16/20) | 21.96 s
[Task 19/25] Current/Best: 5.05/ 21.45 GFLOPS | Progress: (20/20) | 25.51 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 10.71/ 13.76 GFLOPS | Progress: (4/20) | 7.05 s
[Task 20/25] Current/Best: 17.40/ 17.40 GFLOPS | Progress: (8/20) | 10.36 s
[Task 20/25] Current/Best: 17.71/ 17.71 GFLOPS | Progress: (12/20) | 12.23 s
[Task 20/25] Current/Best: 17.14/ 17.71 GFLOPS | Progress: (16/20) | 15.73 s
[Task 20/25] Current/Best: 10.42/ 17.71 GFLOPS | Progress: (20/20) | 19.15 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 8.69/ 8.69 GFLOPS | Progress: (4/20) | 5.30 s
[Task 21/25] Current/Best: 16.47/ 18.88 GFLOPS | Progress: (8/20) | 11.76 s
[Task 21/25] Current/Best: 18.74/ 18.88 GFLOPS | Progress: (12/20) | 13.76 s Done.
+
[Task 21/25] Current/Best: 10.41/ 18.88 GFLOPS | Progress: (16/20) | 17.37 s
[Task 21/25] Current/Best: 7.43/ 18.88 GFLOPS | Progress: (20/20) | 22.27 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 8.91/ 19.30 GFLOPS | Progress: (4/20) | 5.71 s
[Task 22/25] Current/Best: 17.81/ 21.90 GFLOPS | Progress: (8/20) | 7.49 s
[Task 22/25] Current/Best: 5.36/ 21.90 GFLOPS | Progress: (12/20) | 9.44 s
[Task 22/25] Current/Best: 20.70/ 21.90 GFLOPS | Progress: (16/20) | 11.45 s
[Task 22/25] Current/Best: 7.67/ 21.90 GFLOPS | Progress: (20/20) | 13.90 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 3.07/ 17.93 GFLOPS | Progress: (4/20) | 6.79 s
[Task 23/25] Current/Best: 13.08/ 22.58 GFLOPS | Progress: (8/20) | 9.80 s
[Task 23/25] Current/Best: 20.02/ 22.58 GFLOPS | Progress: (12/20) | 12.54 s
[Task 23/25] Current/Best: 10.10/ 22.58 GFLOPS | Progress: (16/20) | 16.51 s
[Task 23/25] Current/Best: 19.74/ 24.05 GFLOPS | Progress: (20/20) | 19.18 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 2.08/ 6.95 GFLOPS | Progress: (4/20) | 13.80 s
[Task 24/25] Current/Best: 8.09/ 8.09 GFLOPS | Progress: (8/20) | 26.19 s
[Task 24/25] Current/Best: 4.22/ 8.09 GFLOPS | Progress: (12/20) | 38.83 s
[Task 24/25] Current/Best: 7.90/ 8.09 GFLOPS | Progress: (16/20) | 49.50 s
[Task 24/25] Current/Best: 10.34/ 10.34 GFLOPS | Progress: (20/20) | 62.50 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
Done.
-
[Task 21/25] Current/Best: 7.62/ 20.64 GFLOPS | Progress: (12/20) | 11.45 s
[Task 21/25] Current/Best: 17.38/ 20.64 GFLOPS | Progress: (16/20) | 14.52 s
[Task 21/25] Current/Best: 5.41/ 20.64 GFLOPS | Progress: (20/20) | 18.27 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 12.20/ 16.83 GFLOPS | Progress: (4/20) | 4.49 s
[Task 22/25] Current/Best: 11.76/ 16.83 GFLOPS | Progress: (8/20) | 7.25 s
[Task 22/25] Current/Best: 11.72/ 18.57 GFLOPS | Progress: (12/20) | 9.24 s
[Task 22/25] Current/Best: 5.27/ 20.78 GFLOPS | Progress: (16/20) | 12.56 s
[Task 22/25] Current/Best: 14.03/ 20.78 GFLOPS | Progress: (20/20) | 15.69 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 20.10/ 20.10 GFLOPS | Progress: (4/20) | 4.97 s
[Task 23/25] Current/Best: 23.26/ 23.26 GFLOPS | Progress: (8/20) | 7.87 s
[Task 23/25] Current/Best: 1.55/ 23.26 GFLOPS | Progress: (12/20) | 12.65 s
[Task 23/25] Current/Best: 22.14/ 23.26 GFLOPS | Progress: (16/20) | 15.27 s
[Task 23/25] Current/Best: 19.29/ 23.26 GFLOPS | Progress: (20/20) | 17.50 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 4.90/ 8.19 GFLOPS | Progress: (4/20) | 13.74 s
[Task 24/25] Current/Best: 1.08/ 8.19 GFLOPS | Progress: (8/20) | 25.89 s
[Task 24/25] Current/Best: 3.87/ 8.66 GFLOPS | Progress: (12/20) | 36.83 s
[Task 24/25] Current/Best: 1.50/ 8.66 GFLOPS | Progress: (16/20) | 40.11 s
[Task 24/25] Current/Best: 7.03/ 9.16 GFLOPS | Progress: (20/20) | 50.75 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
[Task 25/25] Current/Best: 6.38/ 6.38 GFLOPS | Progress: (4/20) | 15.14 s
[Task 25/25] Current/Best: 1.55/ 9.67 GFLOPS | Progress: (8/20) | 26.10 s
[Task 25/25] Current/Best: 1.55/ 9.67 GFLOPS | Progress: (12/20) | 28.82 s
[Task 25/25] Current/Best: 8.46/ 9.67 GFLOPS | Progress: (16/20) | 30.27 s
[Task 25/25] Current/Best: 9.90/ 9.90 GFLOPS | Progress: (20/20) | 41.23 s
+ Done.
+
[Task 25/25] Current/Best: 5.12/ 7.36 GFLOPS | Progress: (4/20) | 5.59 s
[Task 25/25] Current/Best: 5.83/ 8.88 GFLOPS | Progress: (8/20) | 10.55 s
[Task 25/25] Current/Best: 9.24/ 9.24 GFLOPS | Progress: (12/20) | 21.24 s
[Task 25/25] Current/Best: 3.03/ 9.24 GFLOPS | Progress: (16/20) | 32.22 s
[Task 25/25] Current/Best: 7.77/ 9.24 GFLOPS | Progress: (20/20) | 43.18 s
@@ -665,7 +666,7 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
class='n02123045 tabby, tabby cat' with probability=0.621104
- class='n02123159 tiger cat' with probability=0.356379
+ class='n02123159 tiger cat' with probability=0.356378
class='n02124075 Egyptian cat' with probability=0.019712
class='n02129604 tiger, Panthera tigris' with probability=0.001215
class='n04040759 radiator' with probability=0.000262
@@ -722,8 +723,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 411.58318418000135, 'median': 410.62697249999474, 'std': 3.2463555895772456}
- unoptimized: {'mean': 511.0277086200017, 'median': 510.5724576499995, 'std': 1.9596086703937687}
+ optimized: {'mean': 426.34954656999753, 'median': 425.6728701499924, 'std': 3.6461599603240114}
+ unoptimized: {'mean': 550.6390348200011, 'median': 546.4105448499993, 'std': 15.583710586461082}
@@ -746,7 +747,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 12 minutes 34.020 seconds)
+ **Total running time of the script:** ( 13 minutes 15.632 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 33e7a4e221..1d63e57273 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -274,7 +274,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.344e-07 secs/op
+ 1.219e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index d20af8232b..9fab83fcae 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -270,7 +270,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x250a2f60)), stage(b, placeholder(b, 0xf1445b0)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T. [...]
+ [stage(a, placeholder(a, 0x2379ba40)), stage(b, placeholder(b, 0xd6b9d50)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T. [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index cecbe561c9..b4835ca144 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,24 +5,24 @@
Computation times
=================
-**16:19.758** total execution time for **tutorial** files:
+**16:41.930** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 12:34.020 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 13:15.632 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:30.840 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:22.267 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:00.705 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 00:58.402 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:36.382 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:37.142 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:35.404 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:25.867 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.385 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.563 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.852 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.864 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.170 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.194 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.000 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index 181a2b899b..65917ce81d 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -389,7 +389,7 @@ compile and run this new schedule with the parallel operation applied:
.. code-block:: none
- parallel: 0.000010
+ parallel: 0.000008
@@ -498,10 +498,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 6.761279998954706e-06 1.0
- naive 6.642e-06 0.9823583701646513
- parallel 1.0234400000000001e-05 1.5136778837116995
- vector 2.45438e-05 3.630052298350974
+ numpy 7.415540001147746e-06 1.0
+ naive 6.6986e-06 0.9033192456602241
+ parallel 8.3874e-06 1.1310572121115703
+ vector 2.4674099999999998e-05 3.3273504014786575
@@ -922,7 +922,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.018208
+ Numpy running time: 0.018502
@@ -980,7 +980,7 @@ optimizations.
.. code-block:: none
- none: 3.384123
+ none: 3.183745
@@ -1080,7 +1080,7 @@ schedule.
.. code-block:: none
- blocking: 0.305746
+ blocking: 0.300639
@@ -1164,7 +1164,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.340842
+ vectorization: 0.338995
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1230,7 +1230,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.113210
+ loop permutation: 0.120766
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1321,7 +1321,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.107266
+ array packing: 0.108812
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1404,7 +1404,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.110337
+ block caching: 0.111143
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1478,7 +1478,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.145685
+ parallelization: 0.146442
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1548,13 +1548,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.3841233051 1.0
- blocking 0.30574600900000004 0.09034718343129797
- vectorization 0.3408423891 0.1007180762551819
- loop permutation 0.1132099054 0.03345324481214631
- array packing 0.10726575680000001 0.031696763719674906
- block caching 0.1103372088 0.0326043701285109
- parallelization 0.1456849141 0.04304952892243832
+ none 3.1837453195 1.0
+ blocking 0.3006391989 0.09442941213250523
+ vectorization 0.3389950156 0.10647680061709158
+ loop permutation 0.1207660036 0.03793205532500509
+ array packing 0.10881237490000002 0.03417747463452534
+ block caching 0.1111432627 0.034909595946404026
+ parallelization 0.14644234039999998 0.045996876541305264
@@ -1594,11 +1594,6 @@ 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 0.705 seconds)
-
-
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
.. only:: html
diff --git a/docs/commit_hash b/docs/commit_hash
index 23f536c5f7..efda7c71b4 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-06fabe4c5a34db8ce33327b7022f63b7539c07e8
+ff12a2032352d8376cc902ca996e1121d405b3e1
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index a2dde9e99b..07c29bb9c8 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -590,7 +590,7 @@ class:['truck 0.9266'] left:471 top:83 right:689 bottom:169
class:['bicycle 0.9984'] left:111 top:113 right:577 bottom:447
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 19.115 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 21.306 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 81c0f440b8..46b735a3d0 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -511,7 +511,7 @@ Tensorflow is also required since it’s used as the default backend of keras.</
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Relay top-1 id: 285, class name: Egyptian cat
1/1 [==============================] - ETA: 0s
-1/1 [==============================] - 1s 941ms/step
+1/1 [==============================] - 1s 990ms/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 4c6aaf63b5..1e9b229ab7 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -444,7 +444,7 @@
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip4fd6033c-3c45-4b1b-84d2-4c6ffd33b2f3 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.zip3a91dd03-5f9a-467a-88e2-d4a308d99f35 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 e5da254035..59bf93e684 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -454,12 +454,12 @@ Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdo
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
0%| | 0.00/41.5M [00:00<?, ?B/s]
- 19%|#9 | 7.99M/41.5M [00:00<00:00, 76.7MB/s]
- 39%|###8 | 16.0M/41.5M [00:00<00:00, 53.5MB/s]
- 58%|#####7 | 24.0M/41.5M [00:00<00:00, 57.7MB/s]
- 77%|#######7 | 32.0M/41.5M [00:00<00:00, 65.6MB/s]
- 93%|#########3| 38.6M/41.5M [00:00<00:00, 66.6MB/s]
-100%|##########| 41.5M/41.5M [00:00<00:00, 64.6MB/s]
+ 19%|#9 | 7.99M/41.5M [00:00<00:00, 66.9MB/s]
+ 39%|###8 | 16.0M/41.5M [00:00<00:00, 64.7MB/s]
+ 58%|#####7 | 24.0M/41.5M [00:00<00:00, 62.1MB/s]
+ 77%|#######7 | 32.0M/41.5M [00:00<00:00, 64.9MB/s]
+ 96%|#########6| 40.0M/41.5M [00:00<00:00, 66.3MB/s]
+100%|##########| 41.5M/41.5M [00:00<00:00, 67.2MB/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 8630e65f70..0a9ef122c3 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -437,10 +437,11 @@ 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]
- 28%|##7 | 12.4M/44.7M [00:00<00:00, 130MB/s]
- 55%|#####5 | 24.8M/44.7M [00:00<00:00, 98.4MB/s]
- 82%|########1 | 36.6M/44.7M [00:00<00:00, 108MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 101MB/s]
+ 18%|#7 | 7.99M/44.7M [00:00<00:00, 78.1MB/s]
+ 49%|####9 | 22.0M/44.7M [00:00<00:00, 117MB/s]
+ 75%|#######4 | 33.3M/44.7M [00:00<00:00, 92.8MB/s]
+ 95%|#########5| 42.6M/44.7M [00:00<00:00, 85.4MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 87.5MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_tensorflow.html b/docs/how_to/compile_models/from_tensorflow.html
index 51a1a58302..2c2277e2aa 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -654,7 +654,7 @@ 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 23.459 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.775 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 7d2fd42fa2..0973568a00 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>06:37.193</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>06:42.775</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -354,43 +354,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></td>
-<td><p>01:23.459</p></td>
+<td><p>01:23.775</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></td>
-<td><p>01:19.115</p></td>
+<td><p>01:21.306</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>00:56.187</p></td>
+<td><p>00:57.512</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>00:37.473</p></td>
+<td><p>00:37.996</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></td>
-<td><p>00:31.830</p></td>
+<td><p>00:32.531</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></td>
-<td><p>00:31.078</p></td>
+<td><p>00:31.196</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>00:27.340</p></td>
+<td><p>00:27.306</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></td>
-<td><p>00:26.181</p></td>
+<td><p>00:25.628</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>00:21.810</p></td>
+<td><p>00:22.771</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>00:02.719</p></td>
+<td><p>00:02.753</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno.html b/docs/how_to/deploy_models/deploy_model_on_adreno.html
index 4461875728..30cc959e38 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno.html
@@ -925,7 +925,7 @@ Top5 predictions:
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 2752.8736 2751.4635 2763.1592 2750.2746 3.7295
+ 2540.5413 2541.5224 2542.5263 2537.8086 1.8890
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-model-on-adreno-py">
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 2dd3aca171..5514e3cf4f 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -667,7 +667,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.8604 15.8693 16.0653 15.6673 0.1238
+ 16.2988 16.1252 16.8254 15.7942 0.3921
</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 5ff6702efa..d7dd3edf9a 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -459,23 +459,32 @@ be unstable.</p>
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 5%|4 | 8.12M/170M [00:00<00:02, 83.0MB/s]
- 9%|9 | 16.0M/170M [00:00<00:01, 80.9MB/s]
- 14%|#4 | 24.0M/170M [00:00<00:02, 74.1MB/s]
- 21%|##1 | 35.8M/170M [00:00<00:01, 92.4MB/s]
- 26%|##6 | 44.8M/170M [00:00<00:01, 74.1MB/s]
- 31%|### | 52.4M/170M [00:00<00:01, 74.1MB/s]
- 35%|###5 | 59.8M/170M [00:00<00:01, 75.1MB/s]
- 42%|####2 | 72.0M/170M [00:00<00:01, 76.8MB/s]
- 52%|#####1 | 88.0M/170M [00:01<00:00, 91.2MB/s]
- 61%|###### | 103M/170M [00:01<00:00, 108MB/s]
- 67%|######7 | 114M/170M [00:01<00:00, 103MB/s]
- 73%|#######2 | 124M/170M [00:01<00:00, 85.2MB/s]
- 78%|#######8 | 133M/170M [00:01<00:00, 84.5MB/s]
- 83%|########3 | 141M/170M [00:01<00:00, 84.6MB/s]
- 89%|########9 | 152M/170M [00:01<00:00, 90.9MB/s]
- 94%|#########4| 160M/170M [00:01<00:00, 82.5MB/s]
-100%|##########| 170M/170M [00:02<00:00, 87.1MB/s]
+ 5%|4 | 7.99M/170M [00:00<00:03, 50.8MB/s]
+ 9%|8 | 14.8M/170M [00:00<00:02, 60.2MB/s]
+ 12%|#2 | 20.8M/170M [00:00<00:02, 59.5MB/s]
+ 16%|#5 | 26.6M/170M [00:00<00:03, 48.8MB/s]
+ 19%|#8 | 31.5M/170M [00:00<00:03, 40.3MB/s]
+ 21%|## | 35.6M/170M [00:00<00:03, 38.8MB/s]
+ 24%|##3 | 40.0M/170M [00:00<00:03, 36.9MB/s]
+ 28%|##8 | 48.0M/170M [00:01<00:02, 43.9MB/s]
+ 33%|###2 | 56.0M/170M [00:01<00:02, 51.6MB/s]
+ 38%|###7 | 64.0M/170M [00:01<00:02, 48.6MB/s]
+ 42%|####2 | 72.0M/170M [00:01<00:01, 53.9MB/s]
+ 47%|####7 | 80.0M/170M [00:01<00:01, 52.3MB/s]
+ 52%|#####1 | 88.0M/170M [00:01<00:01, 44.7MB/s]
+ 56%|#####5 | 94.3M/170M [00:02<00:01, 45.4MB/s]
+ 60%|###### | 102M/170M [00:02<00:01, 41.8MB/s]
+ 63%|######2 | 107M/170M [00:02<00:01, 40.4MB/s]
+ 67%|######6 | 114M/170M [00:02<00:01, 47.3MB/s]
+ 71%|####### | 120M/170M [00:02<00:01, 41.3MB/s]
+ 75%|#######5 | 128M/170M [00:02<00:00, 43.9MB/s]
+ 80%|######## | 136M/170M [00:03<00:00, 46.0MB/s]
+ 85%|########4 | 144M/170M [00:03<00:00, 49.8MB/s]
+ 88%|########8 | 150M/170M [00:03<00:00, 45.2MB/s]
+ 91%|#########1| 155M/170M [00:03<00:00, 40.8MB/s]
+ 94%|#########4| 160M/170M [00:03<00:00, 39.9MB/s]
+ 99%|#########8| 168M/170M [00:03<00:00, 49.0MB/s]
+100%|##########| 170M/170M [00:03<00:00, 46.3MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode=& [...]
@@ -573,7 +582,7 @@ 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 33.082 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 39.018 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 0eee4bc108..6796278029 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -500,8 +500,9 @@ 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, 80.1MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 102MB/s]
+ 47%|####6 | 6.30M/13.6M [00:00<00:00, 44.1MB/s]
+ 78%|#######7 | 10.5M/13.6M [00:00<00:00, 43.6MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 45.7MB/s]
</pre></div>
</div>
</div>
@@ -592,7 +593,7 @@ 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)
- 90.3305 90.1795 95.6701 90.0084 0.5895
+ 90.3555 90.3418 92.0020 90.0462 0.2361
</pre></div>
</div>
<div class="admonition note">
@@ -631,7 +632,7 @@ 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 16.788 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 17.863 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 9da86a0efe..cce0ff5a5e 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -585,7 +585,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 118.7670 118.6938 121.5630 117.8118 0.4996
+ 119.9148 119.8763 122.1049 119.2164 0.4122
</pre></div>
</div>
<div class="admonition note">
@@ -613,7 +613,7 @@ network for ARM CPU</span></a>.</p></li>
</ul>
</div></blockquote>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 34.564 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 31.118 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 8f453f6d71..c7b1fabf8c 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -526,7 +526,7 @@ for calibration. But the accuracy might be impacted.</p>
DeprecationWarning,
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 36.031 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 36.012 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 8c760bbcf6..704ed86fbf 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -468,23 +468,23 @@ to your device.</p>
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
0%| | 0/132723 [00:00<?, ?KB/s]
- 1%|1 | 1989/132723 [00:00<00:06, 19888.69KB/s]
- 5%|4 | 6158/132723 [00:00<00:03, 32711.42KB/s]
- 10%|# | 13445/132723 [00:00<00:02, 51047.11KB/s]
- 16%|#5 | 20863/132723 [00:00<00:01, 60176.73KB/s]
- 22%|##2 | 29504/132723 [00:00<00:01, 69633.26KB/s]
- 29%|##8 | 38185/132723 [00:00<00:01, 75471.14KB/s]
- 35%|###5 | 46783/132723 [00:00<00:01, 78904.17KB/s]
- 42%|####1 | 55443/132723 [00:00<00:00, 81352.12KB/s]
- 48%|####8 | 64109/132723 [00:00<00:00, 83006.98KB/s]
- 55%|#####4 | 72804/132723 [00:01<00:00, 84221.48KB/s]
- 61%|######1 | 81458/132723 [00:01<00:00, 84928.31KB/s]
- 68%|######7 | 90085/132723 [00:01<00:00, 85333.86KB/s]
- 74%|#######4 | 98726/132723 [00:01<00:00, 85658.75KB/s]
- 81%|######## | 107389/132723 [00:01<00:00, 85947.97KB/s]
- 87%|########7 | 116015/132723 [00:01<00:00, 86034.50KB/s]
- 94%|#########3| 124648/132723 [00:01<00:00, 86122.47KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 78304.85KB/s]
+ 4%|4 | 5376/132723 [00:00<00:02, 53756.72KB/s]
+ 9%|9 | 12535/132723 [00:00<00:01, 64244.31KB/s]
+ 14%|#4 | 18960/132723 [00:00<00:01, 63847.55KB/s]
+ 21%|## | 27659/132723 [00:00<00:01, 72946.38KB/s]
+ 26%|##6 | 34958/132723 [00:00<00:01, 70871.06KB/s]
+ 33%|###2 | 43677/132723 [00:00<00:01, 76301.14KB/s]
+ 39%|###8 | 51327/132723 [00:00<00:01, 56104.09KB/s]
+ 45%|####5 | 60042/132723 [00:00<00:01, 63950.61KB/s]
+ 52%|#####1 | 68705/132723 [00:01<00:00, 69964.92KB/s]
+ 58%|#####8 | 77495/132723 [00:01<00:00, 74891.88KB/s]
+ 65%|######5 | 86287/132723 [00:01<00:00, 78562.92KB/s]
+ 72%|#######1 | 95105/132723 [00:01<00:00, 81320.90KB/s]
+ 78%|#######8 | 103826/132723 [00:01<00:00, 83026.80KB/s]
+ 85%|########4 | 112531/132723 [00:01<00:00, 84204.05KB/s]
+ 91%|#########1| 121325/132723 [00:01<00:00, 85303.44KB/s]
+ 98%|#########8| 130124/132723 [00:01<00:00, 86097.96KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 75710.28KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -523,7 +523,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 45.518 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 48.579 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 4211899536..967c1ebfdb 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>15:22.218</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>15:27.211</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -354,39 +354,39 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>03:45.518</p></td>
+<td><p>03:48.579</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>03:33.082</p></td>
+<td><p>03:39.018</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>02:34.564</p></td>
+<td><p>02:31.118</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>01:36.031</p></td>
+<td><p>01:36.012</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>01:16.788</p></td>
+<td><p>01:17.863</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_adreno.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno.py</span></code>)</p></td>
-<td><p>00:57.082</p></td>
+<td><p>00:54.802</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></td>
-<td><p>00:42.485</p></td>
+<td><p>00:43.500</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:28.484</p></td>
+<td><p>00:28.466</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:28.179</p></td>
+<td><p>00:27.847</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index a84629a361..29ec24082a 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -624,7 +624,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
<span class="n">module</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">params</span></a> <span class="o">=</span> <span class="n">get_mobilenet</span><span class="p">()</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip41bcb284-c50e-4de0-8317-aec415239e34 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.zipe688c588-85b8-486a-b730-9ee5303afc08 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 1c6ef2908d..5b6b54a4b7 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:54.048</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:55.422</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:50.204</p></td>
+<td><p>00:51.480</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>00:02.745</p></td>
+<td><p>00:02.816</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></td>
-<td><p>00:01.089</p></td>
+<td><p>00:01.117</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index 157c12163e..62572eb27b 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -531,10 +531,10 @@ profile the execution time of each passes.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 22478us [22478us] (48.55%; 48.55%)
-FoldScaleAxis: 23818us [24us] (51.45%; 51.45%)
- FoldConstant: 23794us [1761us] (51.40%; 99.90%)
- InferType: 22033us [22033us] (47.59%; 92.60%)
+InferType: 22340us [22340us] (48.11%; 48.11%)
+FoldScaleAxis: 24100us [8us] (51.89%; 51.89%)
+ FoldConstant: 24092us [1742us] (51.88%; 99.97%)
+ InferType: 22350us [22350us] (48.13%; 92.77%)
</pre></div>
</div>
</div>
@@ -556,10 +556,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 22158us [22158us] (48.39%; 48.39%)
-FoldScaleAxis: 23635us [6us] (51.61%; 51.61%)
- FoldConstant: 23629us [1778us] (51.60%; 99.97%)
- InferType: 21851us [21851us] (47.72%; 92.48%)
+InferType: 22486us [22486us] (48.14%; 48.14%)
+FoldScaleAxis: 24222us [9us] (51.86%; 51.86%)
+ FoldConstant: 24213us [1817us] (51.84%; 99.96%)
+ InferType: 22396us [22396us] (47.95%; 92.49%)
</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 ef67ee4d1d..00c430bceb 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -580,7 +580,7 @@ latency of convolution.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Convolution: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 36.923774 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 39.969024 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 ef13d5a80c..efcc533cfd 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -862,7 +862,7 @@ be able to run on our build server</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"conv2d with tensor core: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">* [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 13.355830 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 13.256323 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 8d0ab1509b..2839592c1b 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -477,8 +477,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
<span class="nb">print</span><span class="p">(</span><span class="s2">"Baseline: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018244
-Baseline: 3.427344
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018360
+Baseline: 3.272373
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -537,7 +537,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt1: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.299723
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.310553
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -594,7 +594,7 @@ vastly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt2: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.331611
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.344205
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -649,7 +649,7 @@ the access pattern for A matrix is more cache friendly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt3: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.116872
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.113700
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -726,7 +726,7 @@ flattening.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt4: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109720
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.107906
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -804,7 +804,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt5: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111627
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.110479
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -884,7 +884,7 @@ class Module:
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.146855
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147061
</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 11f6c679ce..915ea4f467 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:35.023</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:34.632</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></td>
-<td><p>00:32.293</p></td>
+<td><p>00:32.048</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>00:01.612</p></td>
+<td><p>00:01.539</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>00:01.118</p></td>
+<td><p>00:01.045</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 c2bb6fac14..86f2ec281e 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>09:53.843</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>09:52.900</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -354,27 +354,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></td>
-<td><p>06:04.967</p></td>
+<td><p>06:02.638</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-x86-py"><span class="std std-ref">Auto-scheduling a Neural Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_x86.py</span></code>)</p></td>
-<td><p>01:42.003</p></td>
+<td><p>01:42.460</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>01:07.337</p></td>
+<td><p>01:07.681</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>00:31.824</p></td>
+<td><p>00:32.030</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>00:14.197</p></td>
+<td><p>00:14.221</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></td>
-<td><p>00:13.515</p></td>
+<td><p>00:13.869</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 4ef8e9f413..6d2648d268 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
@@ -510,479 +510,146 @@ class Module:
@T.prim_func
def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- blockIdx_x = T.launch_thread("blockIdx.x", 56)
- conv2d_nchw = T.allocate([7], "float32", "local")
- pad_temp_shared = T.allocate([336], "float32", "shared")
- kernel_shared = T.allocate([3072], "float32", "shared")
- threadIdx_x = T.launch_thread("threadIdx.x", 64)
- conv2d_nchw_1 = T.Buffer((7,), data=conv2d_nchw, scope="local", align=16)
+ blockIdx_x = T.launch_thread("blockIdx.x", 32)
+ conv2d_nchw = T.allocate([14], "float32", "local")
+ pad_temp_shared = T.allocate([6272], "float32", "shared")
+ kernel_shared = T.allocate([2048], "float32", "shared")
+ threadIdx_x = T.launch_thread("threadIdx.x", 56)
+ conv2d_nchw_1 = T.Buffer((14,), data=conv2d_nchw, scope="local", align=32)
conv2d_nchw_1[0] = T.float32(0)
+ conv2d_nchw_1[7] = T.float32(0)
conv2d_nchw_1[1] = T.float32(0)
+ conv2d_nchw_1[8] = T.float32(0)
conv2d_nchw_1[2] = T.float32(0)
+ conv2d_nchw_1[9] = T.float32(0)
conv2d_nchw_1[3] = T.float32(0)
+ conv2d_nchw_1[10] = T.float32(0)
conv2d_nchw_1[4] = T.float32(0)
+ conv2d_nchw_1[11] = T.float32(0)
conv2d_nchw_1[5] = T.float32(0)
+ conv2d_nchw_1[12] = T.float32(0)
conv2d_nchw_1[6] = T.float32(0)
- for rc_outer_outer, ry_outer_outer in T.grid(32, 3):
- cse_var_4: T.int32 = rc_outer_outer * 784
- cse_var_3: T.int32 = ry_outer_outer * 7
- cse_var_2: T.int32 = rc_outer_outer * 144
+ conv2d_nchw_1[13] = T.float32(0)
+ for rc_outer_outer, ry_outer_outer, rx_outer_outer in T.grid(4, 3, 3):
+ cse_var_2: T.int32 = rc_outer_outer * 1152
cse_var_1: T.int32 = ry_outer_outer * 3
+ pad_temp_shared_1 = T.Buffer((6272,), data=pad_temp_shared, scope="shared")
+ for ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer in range(112):
+ cse_var_3: T.int32 = ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer * 56
+ threadIdx_x_1 = T.launch_thread("threadIdx.x", 56)
+ data_1 = T.Buffer((25088,), data=data.data)
+ pad_temp_shared_1[cse_var_3 + threadIdx_x_1] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 7 + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7 and ry_outer_outer + (threadIdx_x_1 // 7 + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7 < 8 and 1 <= rx_outer_outer + threadIdx_x_1 % 7 and rx_outer_outer + threadIdx_x_1 % 7 < 8, data_1[rc_outer_outer * 6272 + cse_var_3 + ry_outer_outer * 7 + threadIdx_x_1 + rx_outer_outer - 8], T.float32(0))
threadIdx_x_1 = T.env_thread("threadIdx.x")
- pad_temp_shared_1 = T.Buffer((336,), data=pad_temp_shared, scope="shared")
- data_1 = T.Buffer((25088,), data=data.data)
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 <= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 64] = T.if_then_else(1 <= (threadIdx_x_1 + 1) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 1) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 64) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 128] = T.if_then_else(1 <= (threadIdx_x_1 + 2) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 2) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 128) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 192] = T.if_then_else(1 <= ry_outer_outer + (threadIdx_x_1 // 3 + 1) % 7 and ry_outer_outer + (threadIdx_x_1 // 3 + 1) % 7 < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 440], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- pad_temp_shared_1[threadIdx_x_1 + 256] = T.if_then_else(1 <= (threadIdx_x_1 + 4) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 4) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 256) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
- with T.launch_thread(threadIdx_x_1, 64):
- if T.likely(threadIdx_x_1 < 16):
- pad_temp_shared_1[threadIdx_x_1 + 320] = T.if_then_else((threadIdx_x_1 + 5) // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 320) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
- threadIdx_x_2 = T.env_thread("threadIdx.x")
- kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope="shared")
+ kernel_shared_1 = T.Buffer((2048,), data=kernel_shared, scope="shared")
kernel_1 = T.Buffer((2359296,), data=kernel.data)
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 64) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 64) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 128) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 128) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 192] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 18432]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 256) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 256) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 320) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 320) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 384] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 36864]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 448) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 448) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 512) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 512) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 576] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 55296]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 640) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 640) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 704) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 704) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 768] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 73728]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 832) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 832) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 896) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 896) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 960] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 92160]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1024) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1024) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1088) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1088) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1152] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 110592]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1216) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1216) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1280) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1280) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 129024]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1408) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1408) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1472) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1472) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1536] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 147456]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1600) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1600) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1664) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1664) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1728] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 165888]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1792) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1792) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1856) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1856) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 1920] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 184320]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 1984) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 1984) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2048) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2048) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2112] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 202752]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2176) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2176) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2240) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2240) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2304] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 221184]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2368) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2368) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2432) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2432) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2496] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 239616]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2560) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2560) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2624) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2624) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 258048]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2752) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2752) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2816) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2816) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[threadIdx_x_2 + 2880] = kernel_1[blockIdx_x // 7 * 294912 + threadIdx_x_2 // 48 * 4608 + cse_var_2 + threadIdx_x_2 % 48 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 276480]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 2944) // 48 * 48 + (threadIdx_x_2 + 16) % 48 // 3 * 3 + (threadIdx_x_2 + 1) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 2944) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
- with T.launch_thread(threadIdx_x_2, 64):
- kernel_shared_1[(threadIdx_x_2 + 3008) // 48 * 48 + (threadIdx_x_2 + 32) % 48 // 3 * 3 + (threadIdx_x_2 + 2) % 3] = kernel_1[blockIdx_x // 7 * 294912 + (threadIdx_x_2 + 3008) // 48 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) % 48 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[0] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[21] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[42] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[63] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[3] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[24] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[45] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[66] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[6] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[27] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[48] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[69] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[9] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[30] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[51] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[72] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[12] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[33] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[54] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[75] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[15] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[36] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[57] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[78] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[18] * kernel_shared_1[threadIdx_x * 48]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[39] * kernel_shared_1[threadIdx_x * 48 + 3]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[60] * kernel_shared_1[threadIdx_x * 48 + 6]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[81] * kernel_shared_1[threadIdx_x * 48 + 9]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[1] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[22] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[43] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[64] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[4] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[25] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[46] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[67] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[7] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[28] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[49] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[70] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[10] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[31] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[52] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[73] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[13] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[34] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[55] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[76] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[16] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[37] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[58] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[79] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[19] * kernel_shared_1[threadIdx_x * 48 + 1]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[40] * kernel_shared_1[threadIdx_x * 48 + 4]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[61] * kernel_shared_1[threadIdx_x * 48 + 7]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[82] * kernel_shared_1[threadIdx_x * 48 + 10]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[2] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[23] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[44] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[65] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[5] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[26] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[47] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[68] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[8] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[29] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[50] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[71] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[11] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[32] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[53] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[74] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[14] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[35] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[56] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[77] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[17] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[38] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[59] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[80] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[20] * kernel_shared_1[threadIdx_x * 48 + 2]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[41] * kernel_shared_1[threadIdx_x * 48 + 5]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[62] * kernel_shared_1[threadIdx_x * 48 + 8]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[83] * kernel_shared_1[threadIdx_x * 48 + 11]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[84] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[105] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[126] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[147] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[87] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[108] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[129] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[150] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[90] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[111] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[132] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[153] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[93] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[114] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[135] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[156] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[96] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[117] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[138] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[159] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[99] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[120] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[141] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[162] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[102] * kernel_shared_1[threadIdx_x * 48 + 12]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[123] * kernel_shared_1[threadIdx_x * 48 + 15]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[144] * kernel_shared_1[threadIdx_x * 48 + 18]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[165] * kernel_shared_1[threadIdx_x * 48 + 21]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[85] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[106] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[127] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[148] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[88] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[109] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[130] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[151] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[91] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[112] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[133] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[154] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[94] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[115] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[136] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[157] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[97] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[118] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[139] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[160] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[100] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[121] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[142] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[163] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[103] * kernel_shared_1[threadIdx_x * 48 + 13]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[124] * kernel_shared_1[threadIdx_x * 48 + 16]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[145] * kernel_shared_1[threadIdx_x * 48 + 19]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[166] * kernel_shared_1[threadIdx_x * 48 + 22]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[86] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[107] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[128] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[149] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[89] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[110] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[131] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[152] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[92] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[113] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[134] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[155] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[95] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[116] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[137] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[158] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[98] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[119] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[140] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[161] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[101] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[122] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[143] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[164] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[104] * kernel_shared_1[threadIdx_x * 48 + 14]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[125] * kernel_shared_1[threadIdx_x * 48 + 17]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[146] * kernel_shared_1[threadIdx_x * 48 + 20]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[167] * kernel_shared_1[threadIdx_x * 48 + 23]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[168] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[189] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[210] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[231] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[171] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[192] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[213] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[234] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[174] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[195] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[216] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[237] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[177] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[198] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[219] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[240] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[180] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[201] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[222] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[243] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[183] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[204] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[225] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[246] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[186] * kernel_shared_1[threadIdx_x * 48 + 24]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[207] * kernel_shared_1[threadIdx_x * 48 + 27]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[228] * kernel_shared_1[threadIdx_x * 48 + 30]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[249] * kernel_shared_1[threadIdx_x * 48 + 33]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[169] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[190] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[211] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[232] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[172] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[193] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[214] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[235] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[175] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[196] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[217] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[238] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[178] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[199] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[220] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[241] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[181] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[202] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[223] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[244] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[184] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[205] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[226] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[247] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[187] * kernel_shared_1[threadIdx_x * 48 + 25]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[208] * kernel_shared_1[threadIdx_x * 48 + 28]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[229] * kernel_shared_1[threadIdx_x * 48 + 31]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[250] * kernel_shared_1[threadIdx_x * 48 + 34]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[170] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[191] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[212] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[233] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[173] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[194] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[215] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[236] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[176] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[197] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[218] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[239] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[179] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[200] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[221] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[242] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[182] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[203] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[224] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[245] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[185] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[206] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[227] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[248] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[188] * kernel_shared_1[threadIdx_x * 48 + 26]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[209] * kernel_shared_1[threadIdx_x * 48 + 29]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[230] * kernel_shared_1[threadIdx_x * 48 + 32]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[251] * kernel_shared_1[threadIdx_x * 48 + 35]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[252] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[273] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[294] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[315] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[255] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[276] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[297] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[318] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[258] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[279] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[300] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[321] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[261] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[282] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[303] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[324] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[264] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[285] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[306] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[327] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[267] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[288] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[309] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[330] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[270] * kernel_shared_1[threadIdx_x * 48 + 36]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[291] * kernel_shared_1[threadIdx_x * 48 + 39]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[312] * kernel_shared_1[threadIdx_x * 48 + 42]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[333] * kernel_shared_1[threadIdx_x * 48 + 45]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[253] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[274] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[295] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[316] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[256] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[277] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[298] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[319] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[259] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[280] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[301] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[322] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[262] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[283] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[304] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[325] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[265] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[286] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[307] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[328] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[268] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[289] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[310] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[331] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[271] * kernel_shared_1[threadIdx_x * 48 + 37]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[292] * kernel_shared_1[threadIdx_x * 48 + 40]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[313] * kernel_shared_1[threadIdx_x * 48 + 43]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[334] * kernel_shared_1[threadIdx_x * 48 + 46]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[254] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[275] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[296] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[317] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[257] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[278] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[299] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[320] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[260] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[281] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[302] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[323] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[263] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[284] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[305] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[326] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[266] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[287] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[308] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[329] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[269] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[290] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[311] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[332] * kernel_shared_1[threadIdx_x * 48 + 47]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[272] * kernel_shared_1[threadIdx_x * 48 + 38]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[293] * kernel_shared_1[threadIdx_x * 48 + 41]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[314] * kernel_shared_1[threadIdx_x * 48 + 44]
- conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[335] * kernel_shared_1[threadIdx_x * 48 + 47]
- for i2_inner in range(7):
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 56] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 504]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 112] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 112) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 112) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 168] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 168) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 360]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 224] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 224) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 96) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 280] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 280) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 216]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 336] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 336) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 80) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 392] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 392) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 72]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 448] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 448) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 576]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 504] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 504) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 120) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 560] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 560) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 432]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 616] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 616) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 104) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 672] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 672) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 288]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 728] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 728) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 88) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 784] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 784) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 144]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 840] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 840) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 648]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 896] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 32256]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 952] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 952) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 504]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1008] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1008) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 112) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1064] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1064) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 360]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1120] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1120) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 96) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1176] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1176) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 216]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1232] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1232) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 80) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1288] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1288) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 72]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1344] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1344) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 576]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1400] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1400) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 120) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1456] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1456) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 432]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1512] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1512) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 104) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1568] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1568) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 288]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1624] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1624) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 88) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1680] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1680) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 144]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1736] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1736) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 648]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1792] = kernel_1[blockIdx_x * 73728 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 64512]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1848] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1848) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 504]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1904] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1904) // 128 * 4608 + cse_var_2 + (threadIdx_x_1 + 112) % 128 * 9 + cse_var_1 + rx_outer_outer]
+ with T.launch_thread(threadIdx_x_1, 56):
+ kernel_shared_1[threadIdx_x_1 + 1960] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 1960) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 360]
+ with T.launch_thread(threadIdx_x_1, 56):
+ if T.likely(threadIdx_x_1 < 32):
+ kernel_shared_1[threadIdx_x_1 + 2016] = kernel_1[blockIdx_x * 73728 + (threadIdx_x_1 + 2016) // 128 * 4608 + cse_var_2 + threadIdx_x_1 * 9 + cse_var_1 + rx_outer_outer + 864]
+ for rc_outer_inner in range(64):
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 49] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[7] = conv2d_nchw_1[7] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 49] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 1] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 50] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[8] = conv2d_nchw_1[8] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 50] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 2] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 51] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[9] = conv2d_nchw_1[9] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 51] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 3] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 3] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 52] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[10] = conv2d_nchw_1[10] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 52] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 4] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 4] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[4] = conv2d_nchw_1[4] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 53] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[11] = conv2d_nchw_1[11] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 53] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 5] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 5] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[5] = conv2d_nchw_1[5] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[12] = conv2d_nchw_1[12] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 54] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 6] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 6] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 128]
+ conv2d_nchw_1[6] = conv2d_nchw_1[6] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 1]
+ conv2d_nchw_1[13] = conv2d_nchw_1[13] + pad_temp_shared_1[rc_outer_inner * 98 + threadIdx_x % 7 * 7 + 55] * kernel_shared_1[threadIdx_x // 7 * 256 + rc_outer_inner * 2 + 129]
+ for i1_inner, i3_inner in T.grid(2, 7):
compute_1 = T.Buffer((25088,), data=compute.data)
bias_1 = T.Buffer((512,), data=bias.data)
- compute_1[blockIdx_x // 7 * 3136 + threadIdx_x * 49 + i2_inner * 7 + blockIdx_x % 7] = T.max(conv2d_nchw_1[i2_inner] + bias_1[blockIdx_x // 7 * 64 + threadIdx_x], T.float32(0))
+ compute_1[blockIdx_x * 784 + threadIdx_x // 7 * 98 + i1_inner * 49 + threadIdx_x % 7 * 7 + i3_inner] = T.max(conv2d_nchw_1[i1_inner * 7 + i3_inner] + bias_1[blockIdx_x * 16 + threadIdx_x // 7 * 2 + i1_inner], T.float32(0))
</pre></div>
</div>
</div>
@@ -1016,7 +683,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.360 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.355 ms
</pre></div>
</div>
</div>
@@ -1045,35 +712,35 @@ conv2d_nchw_nn_o_i, conv2d_nchw_nn_i = s[conv2d_nchw].split(conv2d_nchw_nn, fact
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
-conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
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=64)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
-conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=7)
-conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
+conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
+conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
-conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
conv2d_nchw_xx_o_o_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=4)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+conv2d_nchw_rc_o_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=64)
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=3)
+conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
-compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
-compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
-compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
@@ -1094,14 +761,14 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
-s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 64)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -1119,417 +786,108 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[7];
- __shared__ float pad_temp_shared[336];
- __shared__ float kernel_shared[3072];
+extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[14];
+ __shared__ float pad_temp_shared[6272];
+ __shared__ float kernel_shared[2048];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[7] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[8] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[9] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[10] = 0.000000e+00f;
conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[11] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[12] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
+ conv2d_nchw[13] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 4; ++rc_outer_outer) {
for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) && ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) - 8)] : 0.0000 [...]
- pad_temp_shared[(((int)threadIdx.x) + 64)] = (((((1 <= ((((((int)threadIdx.x) + 1) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 1) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 64) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (( [...]
- pad_temp_shared[(((int)threadIdx.x) + 128)] = (((((1 <= ((((((int)threadIdx.x) + 2) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 2) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 128) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
- pad_temp_shared[(((int)threadIdx.x) + 192)] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 3) + 1) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 3) + 1) % 7)) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % [...]
- pad_temp_shared[(((int)threadIdx.x) + 256)] = (((((1 <= ((((((int)threadIdx.x) + 4) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 4) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 256) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
- if (((int)threadIdx.x) < 16) {
- pad_temp_shared[(((int)threadIdx.x) + 320)] = (((((((((int)threadIdx.x) + 5) / 3) + ry_outer_outer) < 8) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 320) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 2) % 3)) - 8)] : 0.000000e+00f);
+ for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
+ __syncthreads();
+ for (int ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer = 0; ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer < 112; ++ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) {
+ pad_temp_shared[((ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer * 56) + ((int)threadIdx.x))] = (((((1 <= (ry_outer_outer + (((((int)threadIdx.x) / 7) + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7))) && ((ry_outer_outer + (((((int)threadIdx.x) / 7) + ax0_ax1_fused_ax2_fused_ax3_fused_outer_outer) % 7)) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer [...]
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 504)];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 112) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 112) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 168) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 360)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 224) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 96) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 280) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 336) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 80) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 392) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 448) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 576)];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 504) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 120) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 560) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 432)];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 616) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 104) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 672) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 288)];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 728) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 88) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 784) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 840) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 648)];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 952) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 504)];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1008) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 112) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1064) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 360)];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1120) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 96) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1176) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 216)];
+ kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1232) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 80) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1288) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1344) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 576)];
+ kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1400) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 120) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1456) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 432)];
+ kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1512) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 104) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1568) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 288)];
+ kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1624) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 88) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1680) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 144)];
+ kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1736) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 648)];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((((int)blockIdx.x) * 73728) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1848) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 504)];
+ kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1904) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((((int)threadIdx.x) + 112) & 127) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 1960) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 360)];
+ if (((int)threadIdx.x) < 32) {
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[(((((((((int)blockIdx.x) * 73728) + (((((int)threadIdx.x) + 2016) >> 7) * 4608)) + (rc_outer_outer * 1152)) + (((int)threadIdx.x) * 9)) + (ry_outer_outer * 3)) + rx_outer_outer) + 864)];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 64; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7))] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 49)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 49)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 50)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 50)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 51)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 51)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 3)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 52)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 52)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 4)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 53)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 53)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 5)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 54)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2))]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 6)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 128)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 55)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 1)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 98) + ((((int)threadIdx.x) % 7) * 7)) + 55)] * kernel_shared[((((((int)threadIdx.x) / 7) * 256) + (rc_outer_inner * 2)) + 129)]));
+ }
}
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 64) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 64) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 128) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 128) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 18432)];
- kernel_shared[(((((((int)threadIdx.x) + 256) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 256) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 320) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 320) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((((((int)threadIdx.x) + 448) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 448) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 512) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 512) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 55296)];
- kernel_shared[(((((((int)threadIdx.x) + 640) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 640) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 704) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 704) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((((((int)threadIdx.x) + 832) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 832) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 896) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 896) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 92160)];
- kernel_shared[(((((((int)threadIdx.x) + 1024) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1024) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1088) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1088) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((((((int)threadIdx.x) + 1216) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1216) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1280) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1280) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
- kernel_shared[(((((((int)threadIdx.x) + 1408) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1408) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1472) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1472) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((((((int)threadIdx.x) + 1600) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1600) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1664) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1664) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 165888)];
- kernel_shared[(((((((int)threadIdx.x) + 1792) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1792) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 1856) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1856) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((((((int)threadIdx.x) + 1984) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1984) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2048) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2048) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 202752)];
- kernel_shared[(((((((int)threadIdx.x) + 2176) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2176) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2240) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2240) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((((((int)threadIdx.x) + 2368) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2368) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2432) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2432) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 239616)];
- kernel_shared[(((((((int)threadIdx.x) + 2560) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2560) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2624) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2624) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((((((int)threadIdx.x) + 2752) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2752) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 2816) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2816) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 276480)];
- kernel_shared[(((((((int)threadIdx.x) + 2944) / 48) * 48) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2944) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 16) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((((((int)threadIdx.x) + 3008) / 48) * 48) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))] = kernel[(((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3008) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 32) % 48) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[9] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[12] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[18] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[231] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[234] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[216] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[237] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[219] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[240] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[222] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[243] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[225] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[246] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[228] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[249] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[232] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[235] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[217] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[238] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[220] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[241] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[223] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[244] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[226] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[247] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[229] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[250] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[233] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[236] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[218] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[239] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[221] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[242] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[224] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[245] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[227] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[248] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[230] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[251] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[252] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[273] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[294] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[315] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[255] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[276] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[297] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[318] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[258] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[279] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[300] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[321] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[261] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[282] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[303] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[324] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[264] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[285] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[306] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[327] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[267] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[288] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[309] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[330] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[270] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[291] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[312] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[333] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[253] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[274] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[295] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[316] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[256] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[277] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[298] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[319] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[259] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[280] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[301] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[322] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[262] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[283] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[304] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[325] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[265] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[286] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[307] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[328] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[268] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[289] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[310] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[331] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[271] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[292] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[313] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[334] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[254] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[275] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[296] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[317] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[257] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[278] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[299] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[320] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[260] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[281] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[302] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[323] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[263] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[284] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[305] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[326] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[266] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[287] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[308] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[329] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[269] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[290] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[311] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[332] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[272] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[293] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[314] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[335] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
}
}
- for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[(((((((int)blockIdx.x) / 7) * 3136) + (((int)threadIdx.x) * 49)) + (i2_inner * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[(((((int)blockIdx.x) / 7) * 64) + ((int)threadIdx.x))]), 0.000000e+00f);
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+ for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
+ compute[(((((((int)blockIdx.x) * 784) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[(((((int)blockIdx.x) * 16) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ }
}
}
</pre></div>
@@ -1564,7 +922,7 @@ In the example below we resume the status and do more 5 trials.</p>
Get devices for measurement successfully!
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 4.967 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 2.638 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 7df5cce1b3..fbaadb60f9 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -921,7 +921,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 7.9149 7.9138 7.9185 7.9123 0.0026
+ 7.9235 7.9218 7.9282 7.9203 0.0034
</pre></div>
</div>
</div>
@@ -943,7 +943,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 7.337 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 7.681 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 93df1de1fc..8ccc14bb69 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -940,7 +940,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 750.5247 748.0445 756.7973 746.7324 4.4676
+ 759.8578 760.0620 760.0696 759.4420 0.2941
</pre></div>
</div>
</div>
@@ -962,7 +962,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 42.003 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 42.460 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 a008e8577e..3db94e0a50 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -637,12 +637,12 @@ class Module:
@T.prim_func
def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
T.func_attr({"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True})
- for i0_outer_i1_outer_fused in T.parallel(256):
- compute_1 = T.allocate([256], "float32", "global")
- compute_2 = T.Buffer((256,), data=compute_1)
- for i_outer_inner in range(4):
- for i_inner_init in range(4):
- cse_var_1: T.int32 = i_outer_inner * 64 + i_inner_init * 16
+ for i0_outer_i1_outer_fused in T.parallel(16):
+ compute_1 = T.allocate([4096], "float32", "global")
+ compute_2 = T.Buffer((4096,), data=compute_1)
+ for i_outer_inner, nb_j_inner in T.grid(2, 2):
+ for i_inner_init in range(64):
+ cse_var_1: T.int32 = i_outer_inner * 2048 + i_inner_init * 32 + nb_j_inner * 16
compute_2[cse_var_1] = T.float32(0)
compute_2[cse_var_1 + 1] = T.float32(0)
compute_2[cse_var_1 + 2] = T.float32(0)
@@ -659,66 +659,52 @@ class Module:
compute_2[cse_var_1 + 13] = T.float32(0)
compute_2[cse_var_1 + 14] = T.float32(0)
compute_2[cse_var_1 + 15] = T.float32(0)
- for elem_idx, i_inner in T.grid(T.Let(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2], where={cse_var_2: i0_outer_i1_outer_fused % 32}), 4):
+ for elem_idx, i_inner in T.grid(T.Let(placeholder_5[cse_var_2 + 1] - placeholder_5[cse_var_2], where={cse_var_2: i0_outer_i1_outer_fused * 2 + nb_j_inner}), 64):
cse_var_2 = T.int32()
placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
- cse_var_3: T.int32 = i0_outer_i1_outer_fused % 32
+ cse_var_21: T.int32 = elem_idx * 16
+ cse_var_20: T.int32 = i0_outer_i1_outer_fused * 2 + nb_j_inner
+ cse_var_19: T.int32 = i_outer_inner * 16384 + i_inner * 256
+ cse_var_18: T.int32 = i_outer_inner * 2048 + i_inner * 32 + nb_j_inner * 16
+ cse_var_17: T.int32 = cse_var_18 + 9
+ cse_var_16: T.int32 = cse_var_18 + 8
+ cse_var_15: T.int32 = cse_var_18 + 7
+ cse_var_14: T.int32 = cse_var_18 + 6
+ cse_var_13: T.int32 = cse_var_18 + 5
+ cse_var_12: T.int32 = cse_var_18 + 4
+ cse_var_11: T.int32 = cse_var_18 + 3
+ cse_var_10: T.int32 = cse_var_18 + 2
+ cse_var_9: T.int32 = cse_var_18 + 15
+ cse_var_8: T.int32 = cse_var_18 + 14
+ cse_var_7: T.int32 = cse_var_18 + 13
+ cse_var_6: T.int32 = cse_var_18 + 12
+ cse_var_5: T.int32 = cse_var_18 + 11
+ cse_var_4: T.int32 = cse_var_18 + 10
+ cse_var_3: T.int32 = cse_var_18 + 1
placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
placeholder_7 = T.Buffer((32768,), data=placeholder.data)
placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_4: T.int32 = i_outer_inner * 64 + i_inner * 16
- compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_5: T.int32 = i_outer_inner * 64 + i_inner * 16 + 1
- compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 1] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_6: T.int32 = i_outer_inner * 64 + i_inner * 16 + 2
- compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 2] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_7: T.int32 = i_outer_inner * 64 + i_inner * 16 + 3
- compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 3] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_8: T.int32 = i_outer_inner * 64 + i_inner * 16 + 4
- compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 4] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_9: T.int32 = i_outer_inner * 64 + i_inner * 16 + 5
- compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 5] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_10: T.int32 = i_outer_inner * 64 + i_inner * 16 + 6
- compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 6] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_11: T.int32 = i_outer_inner * 64 + i_inner * 16 + 7
- compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 7] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_12: T.int32 = i_outer_inner * 64 + i_inner * 16 + 8
- compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 8] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_13: T.int32 = i_outer_inner * 64 + i_inner * 16 + 9
- compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 9] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_14: T.int32 = i_outer_inner * 64 + i_inner * 16 + 10
- compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 10] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_15: T.int32 = i_outer_inner * 64 + i_inner * 16 + 11
- compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 11] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_16: T.int32 = i_outer_inner * 64 + i_inner * 16 + 12
- compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 12] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_17: T.int32 = i_outer_inner * 64 + i_inner * 16 + 13
- compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 13] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_18: T.int32 = i_outer_inner * 64 + i_inner * 16 + 14
- compute_2[cse_var_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 14] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- if T.likely(elem_idx < placeholder_5[cse_var_3 + 1] - placeholder_5[cse_var_3]):
- cse_var_19: T.int32 = i_outer_inner * 64 + i_inner * 16 + 15
- compute_2[cse_var_19] = compute_2[cse_var_19] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + 15] * T.max(placeholder_7[i0_outer_i1_outer_fused // 32 * 4096 + i_outer_inner * 1024 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
- for i0_inner in range(16):
- cse_var_20: T.int32 = i0_outer_i1_outer_fused // 32 * 8192 + i0_inner * 512 + i0_outer_i1_outer_fused % 32 * 16
+ compute_2[cse_var_18] = compute_2[cse_var_18] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_3] = compute_2[cse_var_3] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 1] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_10] = compute_2[cse_var_10] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 2] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_11] = compute_2[cse_var_11] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 3] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_12] = compute_2[cse_var_12] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 4] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_13] = compute_2[cse_var_13] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 5] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_14] = compute_2[cse_var_14] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 6] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_15] = compute_2[cse_var_15] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 7] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_16] = compute_2[cse_var_16] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 8] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_17] = compute_2[cse_var_17] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 9] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_4] = compute_2[cse_var_4] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 10] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_5] = compute_2[cse_var_5] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 11] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_6] = compute_2[cse_var_6] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 12] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_7] = compute_2[cse_var_7] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 13] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_8] = compute_2[cse_var_8] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 14] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ compute_2[cse_var_9] = compute_2[cse_var_9] + placeholder_6[placeholder_5[cse_var_20] * 16 + cse_var_21 + 15] * T.max(placeholder_7[cse_var_19 + placeholder_8[placeholder_5[cse_var_20] + elem_idx]], T.float32(0))
+ for i0_inner in range(128):
+ cse_var_22: T.int32 = i0_inner * 512 + i0_outer_i1_outer_fused * 32
compute_3 = T.Buffer((65536,), data=compute.data)
placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
- compute_3[cse_var_20:cse_var_20 + 16] = T.max(compute_2[i0_inner * 16:i0_inner * 16 + 16] + placeholder_5[cse_var_20:cse_var_20 + 16], T.Broadcast(T.float32(0), 16))
+ compute_3[cse_var_22:cse_var_22 + 32] = T.max(compute_2[i0_inner * 32:i0_inner * 32 + 32] + placeholder_5[cse_var_22:cse_var_22 + 32], T.Broadcast(T.float32(0), 32))
</pre></div>
</div>
</div>
@@ -752,7 +738,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 2.139 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.871 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 d7960aa74c..fa5ae07962 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:33.006</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:47.394</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></td>
-<td><p>00:32.972</p></td>
+<td><p>00:47.358</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></td>
-<td><p>00:00.020</p></td>
+<td><p>00:00.022</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></td>
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 de7cc7de36..8d1642ada8 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -695,7 +695,7 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9435978
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4477522
No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -818,7 +818,7 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 1, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6333769
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5170306
No: 3 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -941,9 +941,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8582861
-No: 4 GFLOPS: 103.89/103.89 result: MeasureResult(costs=(0.002228368358490566,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.504410028457642, timestamp=1678686052.69968) [('tile_f', [-1, 16, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,6987874
-No: 5 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2301592
+No: 4 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1065,8 +1064,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9866495
-No: 6 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9435764
+No: 5 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1188,8 +1187,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5987806
-No: 7 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 32, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2557412
+No: 6 GFLOPS: 17.60/17.60 result: MeasureResult(costs=(0.013150684777777777,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.790144443511963, timestamp=1678752180.9416506) [('tile_f', [-1, 2, 2, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2560249
+No: 7 GFLOPS: 0.00/17.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1311,8 +1311,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3677119
-No: 8 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8628384
+No: 8 GFLOPS: 0.00/17.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1434,8 +1434,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6613715
-No: 9 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1638929
+No: 9 GFLOPS: 0.00/17.60 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1557,8 +1557,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5742039
-No: 10 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 256, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4206012
+No: 10 GFLOPS: 149.26/149.26 result: MeasureResult(costs=(0.0015510194563106796,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5596656799316406, timestamp=1678752182.7189288) [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4883131
+No: 11 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1680,9 +1681,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 2, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5190827
-No: 11 GFLOPS: 42.24/103.89 result: MeasureResult(costs=(0.00548004262962963,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.892000436782837, timestamp=1678686056.4503279) [('tile_f', [-1, 1, 32, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,656390
-No: 12 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3648223
+No: 12 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1804,8 +1804,26 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 16, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7580699
-No: 13 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 32, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7272364
+No: 13 GFLOPS: 0.00/149.26 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, 4, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9154477
+No: 14 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1927,8 +1945,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 256]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8257258
-No: 14 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7382068
+No: 15 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2050,8 +2068,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2499208
-No: 15 GFLOPS: 0.00/103.89 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 32, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9552881
+No: 16 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2173,9 +2191,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 8, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8491318
-No: 16 GFLOPS: 232.07/232.07 result: MeasureResult(costs=(0.0009975514785276072,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9048430919647217, timestamp=1678686058.5952504) [('tile_f', [-1, 1, 8, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7089621
-No: 17 GFLOPS: 0.00/232.07 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 32, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6150584
+No: 17 GFLOPS: 1.31/149.26 result: MeasureResult(costs=(0.177029724,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.406627178192139, timestamp=1678752198.4004164) [('tile_f', [-1, 64, 1, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5616962
+No: 18 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2297,9 +2315,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 8, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4579057
-No: 18 GFLOPS: 91.97/232.07 result: MeasureResult(costs=(0.0025171067560975607,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.0822193622589111, timestamp=1678686059.8699977) [('tile_f', [-1, 1, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1246310
-No: 19 GFLOPS: 0.00/232.07 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 4, 128]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9865015
+No: 19 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -2421,130 +2438,160 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,636036
-No: 20 GFLOPS: 0.00/232.07 result: Traceback (most recent call last):
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
- func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
- func = build(s, args, target=target, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8895135
+No: 20 GFLOPS: 0.00/149.26 result: Traceback (most recent call last):
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 742, in __call__
+ yield remote, remote.load_module(os.path.split(build_result.filename)[1])
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 706, in run_through_rpc
+ costs = time_f(*args).results
+ File "/workspace/python/tvm/runtime/module.py", line 357, in evaluator
+ blob = feval(*args)
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/./packed_func.pxi", line 262, in tvm._ffi._cy3.core.FuncCall
+ File "tvm/_ffi/_cython/./packed_func.pxi", line 251, in tvm._ffi._cy3.core.FuncCall3
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
+ 4: TVMFuncCall
at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:402
- 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
- at ../src/driver/driver_api.cc:388
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:283
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1621
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+ 3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../include/tvm/runtime/packed_func.h:1217
+ 2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+ at ../src/runtime/rpc/rpc_module.cc:129
+ 1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1012
+ 0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
+ at ../src/runtime/rpc/rpc_endpoint.cc:804
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 804
+TVMError:
+---------------------------------------------------------------
+An error occurred during the execution of TVM.
+For more information, please see: https://tvm.apache.org/docs/errors.html
+---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
+
+During handling of the above exception, another exception occurred:
Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1734
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1674
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1634
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1634
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1649
- 13: operator()
- at ../src/driver/driver_api.cc:402
- 12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
- at ../src/driver/driver_api.cc:388
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:283
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:451
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1753
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1697
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 706, in run_through_rpc
+ costs = time_f(*args).results
+ File "/usr/lib/python3.7/contextlib.py", line 130, in __exit__
+ self.gen.throw(type, value, traceback)
+ File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 746, in __call__
+ remote.remove(build_result.filename)
+ File "/workspace/python/tvm/rpc/client.py", line 144, in remove
+ self._remote_funcs["remove"] = self.get_function("tvm.rpc.server.remove")
+ File "/workspace/python/tvm/rpc/client.py", line 72, in get_function
+ return self._sess.get_function(name)
+ File "/workspace/python/tvm/runtime/module.py", line 171, in get_function
+ self.handle, c_str(name), ctypes.c_int(query_imports), ctypes.byref(ret_handle)
+ File "/workspace/python/tvm/_ffi/base.py", line 348, in check_call
+ raise get_last_ffi_error()
+tvm._ffi.base.TVMError: Traceback (most recent call last):
+ 52: 0xffffffffffffffff
+ 51: _start
+ 50: __libc_start_main
+ 49: _Py_UnixMain
+ 48: 0x0000000000650da0
+ 47: 0x0000000000650afa
+ 46: _PyFunction_FastCallDict
+ 45: _PyEval_EvalCodeWithName
+ 44: _PyEval_EvalFrameDefault
+ 43: _PyFunction_FastCallKeywords
+ 42: _PyEval_EvalCodeWithName
+ 41: _PyEval_EvalFrameDefault
+ 40: _PyMethodDef_RawFastCallKeywords
+ 39: 0x0000000000546369
+ 38: _PyEval_EvalCodeWithName
+ 37: _PyEval_EvalFrameDefault
+ 36: _PyFunction_FastCallKeywords
+ 35: _PyEval_EvalCodeWithName
+ 34: _PyEval_EvalFrameDefault
+ 33: _PyFunction_FastCallDict
+ 32: _PyEval_EvalCodeWithName
+ 31: _PyEval_EvalFrameDefault
+ 30: _PyObject_FastCallDict
+ 29: 0x00000000004c06e1
+ 28: _PyFunction_FastCallDict
+ 27: _PyEval_EvalFrameDefault
+ 26: _PyMethodDescr_FastCallKeywords
+ 25: 0x00000000005dcb58
+ 24: 0x00000000005dc83f
+ 23: 0x00000000004ba127
+ 22: _PyEval_EvalFrameDefault
+ 21: _PyFunction_FastCallKeywords
+ 20: _PyEval_EvalFrameDefault
+ 19: _PyFunction_FastCallKeywords
+ 18: _PyEval_EvalFrameDefault
+ 17: _PyFunction_FastCallKeywords
+ 16: _PyEval_EvalCodeWithName
+ 15: _PyEval_EvalFrameDefault
+ 14: 0x0000000000537c30
+ 13: _PyObject_FastCallKeywords
+ 12: 0x00007f3eb5baafa2
+ 11: _ctypes_callproc
+ 10: ffi_call
+ 9: ffi_call_unix64
+ 8: TVMModGetFunction
+ at ../src/runtime/c_runtime_api.cc:408
+ 7: tvm::runtime::ModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)
+ at ../src/runtime/module.cc:66
+ 6: tvm::runtime::RPCModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, tvm::runtime::ObjectPtr<tvm::runtime::Object> const&)
+ at ../src/runtime/rpc/rpc_module.cc:185
+ 5: tvm::runtime::RPCClientSession::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
+ at ../src/runtime/rpc/rpc_endpoint.cc:1007
+ 4: tvm::runtime::TVMRetValue tvm::runtime::RPCEndpoint::SysCallRemote<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(tvm::runtime::RPCCode, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
+ at ../src/runtime/rpc/rpc_endpoint.h:223
+ 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(int&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const
at ../include/tvm/runtime/packed_func.h:1621
2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
at ../include/tvm/runtime/packed_func.h:1217
1: Call
at ../include/tvm/runtime/packed_func.h:1213
0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 1, 128]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2213632
+ at ../src/runtime/rpc/rpc_endpoint.cc:684
+ File "../src/runtime/rpc/rpc_endpoint.cc", line 684
+TVMError:
+---------------------------------------------------------------
+An error occurred during the execution of TVM.
+For more information, please see: https://tvm.apache.org/docs/errors.html
+---------------------------------------------------------------
+ Check failed: (code == RPCCode::kReturn) is false: code=1
+
+Traceback (most recent call last):
+ 52: 0xffffffffffffffff
+ 51: _start
+ 50: __libc_start_main
+ 49: _Py_UnixMain
+ 48: 0x0000000000650da0
+ 47: 0x0000000000650afa
+ 46: _PyFunction_FastCallDict
+ 45: _PyEval_EvalCodeWithName
+ 44: _PyEval_EvalFrameDefault
+ 43: _PyFunction_FastCallKeywords
+ 42: _PyEval_EvalCodeWithName
+ 41: _PyEval_EvalFrameDefault
+ 40: _PyMethodDef_RawFastCallKeywords
+ 39: 0x0000000000546369
+ 38: _PyEval_EvalCodeWithName
+ 37: _PyEval_EvalFrameDefault
+ 36: _PyFunction_FastCallKeywords
+ 35: _PyEval_EvalCodeWithName
+ 34: _PyEval_EvalFrameDefault
+ 33: _PyFunction_FastCallDict
+ 32: _PyEval_EvalCodeWithName
+ 31: _PyEval_EvalFrameDefault
+ 30: _PyObject_FastCallDict
+ 29: 0x00000000004c06e1
+ 28: _PyFunction_FastCallDict
+ 27: _PyEval_EvalFrameDefault
+ 26: _PyMethodDescr_FastCallKeywords
+ 25: 0x00000000005dcb58
+ 24: 0x00000000005dc83f
+ 23: 0x00000000004ba127
+ 22: _PyEval_EvalFrameDefault
+ 21: _PyFunction_FastCallKeywords
+ 20: _PyEval_EvalFrameDefault
+ 19: _PyFunction_FastCall [('tile_f', [-1, 8, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3105708
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2583,9 +2630,9 @@ and measure running time.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Finish loading 20 records
Best config:
-[('tile_f', [-1, 1, 8, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7089621
+[('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4883131
Finish loading 20 records
-Time cost of this operator: 0.001352
+Time cost of this operator: 0.001962
</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 9cacea3c8f..4917e13e50 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -648,10 +648,10 @@ 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 374.6 98.916 (1, 2, 10, 10, 3) 2 1 [374.6]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.114 0.822 (1, 6, 10, 10) 1 1 [3.114]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.992 0.262 (1, 1, 10, 10, 3) 1 1 [0.992]
-Total_time - 378.706 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 313.8 98.742 (1, 2, 10, 10, 3) 2 1 [313.8]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.041 0.957 (1, 6, 10, 10) 1 1 [3.041]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.958 0.301 (1, 1, 10, 10, 3) 1 1 [0.958]
+Total_time - 317.799 - - - - -
</pre></div>
</div>
</div>
@@ -703,13 +703,13 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.4 97.358 (1, 6, 10, 10, 1) 2 1 [100.4]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.762 1.709 (1, 6, 10, 10) 1 1 [1.762]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.962 0.933 (1, 1, 10, 10, 3) 1 1 [0.962]
-Total_time - 103.124 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 102.6 97.423 (1, 6, 10, 10, 1) 2 1 [102.6]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.745 1.657 (1, 6, 10, 10) 1 1 [1.745]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 0.92 (1, 1, 10, 10, 3) 1 1 [0.969]
+Total_time - 105.314 - - - - -
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 21.943 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.989 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_autotune.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_pytorch.html b/docs/how_to/work_with_microtvm/micro_pytorch.html
index 6bf04d4e9b..e3125d0120 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -459,7 +459,7 @@ download a cat image and preprocess it to use as the model input.</p>
Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
0%| | 0.00/3.42M [00:00<?, ?B/s]
-100%|##########| 3.42M/3.42M [00:00<00:00, 43.3MB/s]
+100%|##########| 3.42M/3.42M [00:00<00:00, 40.4MB/s]
/workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
return LooseVersion(torch_ver) > ver
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -585,7 +585,7 @@ via the host <cite>main.cc`</cite> or if a Zephyr emulated board is selected as
Torch top-1 id: 282, class name: tiger cat
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 18.283 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 18.623 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-pytorch-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/12b9ecc04c41abaa12022061771821d1/micro_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index 2b728e1180..896f7d8ccb 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -528,7 +528,7 @@ take about <strong>2 minutes</strong> to download the Stanford Cars, while COCO
<a href="https://docs.python.org/3/library/shutil.html#shutil.move" title="shutil.move" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">move</span></a><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-typ [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpm32a5n_m/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmp03odc2we/images/random'
</pre></div>
</div>
</div>
@@ -588,8 +588,8 @@ objects to other stuff? We can display some examples from our datasets using <co
<span class="n">plt</span><span class="o">.</span><span class="n">axis</span><span class="p">(</span><span class="s2">"off"</span><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpm32a5n_m/images/target contains 8144 images
-/tmp/tmpm32a5n_m/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp03odc2we/images/target contains 8144 images
+/tmp/tmp03odc2we/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -701,13 +701,13 @@ the time on our validation set).</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 46s - loss: 0.2104 - accuracy: 0.9252 - val_loss: 0.1579 - val_accuracy: 0.9494 - 46s/epoch - 141ms/step
+328/328 - 47s - loss: 0.2161 - accuracy: 0.9268 - val_loss: 0.1325 - val_accuracy: 0.9539 - 47s/epoch - 144ms/step
Epoch 2/3
-328/328 - 43s - loss: 0.0956 - accuracy: 0.9635 - val_loss: 0.1417 - val_accuracy: 0.9566 - 43s/epoch - 131ms/step
+328/328 - 43s - loss: 0.1036 - accuracy: 0.9620 - val_loss: 0.1488 - val_accuracy: 0.9513 - 43s/epoch - 132ms/step
Epoch 3/3
-328/328 - 43s - loss: 0.0641 - accuracy: 0.9759 - val_loss: 0.1162 - val_accuracy: 0.9622 - 43s/epoch - 131ms/step
+328/328 - 43s - loss: 0.0646 - accuracy: 0.9777 - val_loss: 0.1283 - val_accuracy: 0.9588 - 43s/epoch - 132ms/step
-<keras.callbacks.History object at 0x7f16b577bc90>
+<keras.callbacks.History object at 0x7f5a9577ed10>
</pre></div>
</div>
</div>
@@ -971,7 +971,7 @@ as intended.</p>
<p>From here, we could modify the model to read live images from the camera - we have another
Arduino tutorial for how to do that <a class="reference external" href="https://github.com/guberti/tvm-arduino-demos/tree/master/examples/person_detection">on GitHub</a>. Alternatively, we could also
<a class="reference external" href="https://tvm.apache.org/docs/how_to/work_with_microtvm/micro_autotune.html">use TVM’s autotuning capabilities</a> to dramatically improve the model’s performance.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 18.505 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 46.174 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 d69fa4304d..ab6064b206 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>07:16.046</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>07:43.444</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 82%" />
@@ -354,23 +354,23 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">5. Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>04:18.505</p></td>
+<td><p>04:46.174</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">6. Model Tuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>01:21.943</p></td>
+<td><p>01:20.989</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_pytorch.html#sphx-glr-how-to-work-with-microtvm-micro-pytorch-py"><span class="std std-ref">4. microTVM PyTorch Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_pytorch.py</span></code>)</p></td>
-<td><p>01:18.283</p></td>
+<td><p>01:18.623</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_aot.html#sphx-glr-how-to-work-with-microtvm-micro-aot-py"><span class="std std-ref">3. microTVM Ahead-of-Time (AOT) Compilation</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:10.076</p></td>
+<td><p>00:10.297</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">2. microTVM TFLite Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:07.239</p></td>
+<td><p>00:07.361</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">7. Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index 44c304f9f4..a91be9df44 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:45.562</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:45.955</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,15 +354,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:33.531</p></td>
+<td><p>00:33.652</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>00:10.369</p></td>
+<td><p>00:10.463</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>00:01.656</p></td>
+<td><p>00:01.834</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 967f8aca64..ced9454a64 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -540,7 +540,7 @@ The following example customizes CUDA lowering rule for <code class="code docuti
<a href="../../reference/api/python/ir.html#tvm.ir.register_intrin_lowering" title="tvm.ir.register_intrin_lowering" class="sphx-glr-backref-module-tvm-ir sphx-glr-backref-type-py-function"><span class="n">register_intrin_lowering</span></a><span class="p">(</span><span class="s2">"tir.exp"</span><span class="p">,</span> <span class="n">target</span><span class="o">=</span><span class="s2">"cuda"</span><span class="p">,</span> <span class="n">f</span><span class="o">= [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f1563938d40>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f5934085d40>
</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 6f50d4ed4b..8b0f1d15ff 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:07.402</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:09.103</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,23 +354,23 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></td>
-<td><p>00:04.893</p></td>
+<td><p>00:06.511</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>00:01.151</p></td>
+<td><p>00:01.228</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>00:00.573</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>00:00.555</p></td>
+<td><p>00:00.556</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>00:00.119</p></td>
+<td><p>00:00.117</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>
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index 1515c4b747..e285b427a5 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -234,7 +234,17 @@
<p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal" href="index.html">Installing TVM</a><ul class="current">
-<li class="toctree-l2"><a class="reference internal" href="from_source.html">Install from Source</a></li>
+<li class="toctree-l2 current"><a class="reference internal" href="from_source.html">Install from Source</a><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#developers-get-source-from-github">Developers: Get Source from Github</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#build-the-shared-library">Build the Shared Library</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#python-package-installation">Python Package Installation</a></li>
+<li class="toctree-l3 current"><a class="reference internal" href="from_source.html#install-contrib-libraries">Install Contrib Libraries</a><ul class="current">
+<li class="toctree-l4 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a></li>
+</ul>
+</li>
+<li class="toctree-l3"><a class="reference internal" href="from_source.html#enable-c-tests">Enable C++ Tests</a></li>
+</ul>
+</li>
<li class="toctree-l2"><a class="reference internal" href="docker.html">Docker Images</a></li>
<li class="toctree-l2 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#conditions">Conditions</a></li>
diff --git a/docs/reference/api/doxygen/detail_2broadcast_8h_source.html b/docs/reference/api/doxygen/detail_2broadcast_8h_source.html
index 7284b565a2..6122be4cee 100644
--- a/docs/reference/api/doxygen/detail_2broadcast_8h_source.html
+++ b/docs/reference/api/doxygen/detail_2broadcast_8h_source.html
@@ -66,7 +66,7 @@ $(function() {
<div class="title">broadcast.h</div> </div>
</div><!--header-->
<div class="contents">
-<a href="detail_2broadcast_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * o [...]
+<a href="detail_2broadcast_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * o [...]
<div class="ttc" id="namespacetvm_html_aac2abc149c1a47944c37b560181b15c0"><div class="ttname"><a href="namespacetvm.html#aac2abc149c1a47944c37b560181b15c0">tvm::min</a></div><div class="ttdeci">PrimExpr min(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">take minimum of two values </div></div>
<div class="ttc" id="namespacetvm_html"><div class="ttname"><a href="namespacetvm.html">tvm</a></div><div class="ttdoc">runtime implementation for LibTorch/TorchScript. </div><div class="ttdef"><b>Definition:</b> analyzer.h:36</div></div>
<div class="ttc" id="classtvm_1_1tir_1_1Var_html"><div class="ttname"><a href="classtvm_1_1tir_1_1Var.html">tvm::tir::Var</a></div><div class="ttdoc">a named variable in TIR </div><div class="ttdef"><b>Definition:</b> var.h:88</div></div>
diff --git a/docs/reference/api/doxygen/group__norm_8h_source.html b/docs/reference/api/doxygen/group__norm_8h_source.html
index e7d11042fc..1149e211bc 100644
--- a/docs/reference/api/doxygen/group__norm_8h_source.html
+++ b/docs/reference/api/doxygen/group__norm_8h_source.html
@@ -67,7 +67,7 @@ $(function() {
</div><!--header-->
<div class="contents">
<a href="group__norm_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or more [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_a12a94f9c1103f5e05fb9819e7176bf4c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a12a94f9c1103f5e05fb9819e7176bf4c">tvm::topi::MakeTupleSumReducer</a></div><div class="ttdeci">FCommReduce MakeTupleSumReducer()</div><div class="ttdoc">Create communitive reducer summing over tuples. </div><div class="ttdef"><b>Definition:</b> reduction.h:576</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a12a94f9c1103f5e05fb9819e7176bf4c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a12a94f9c1103f5e05fb9819e7176bf4c">tvm::topi::MakeTupleSumReducer</a></div><div class="ttdeci">FCommReduce MakeTupleSumReducer()</div><div class="ttdoc">Create communitive reducer summing over tuples. </div><div class="ttdef"><b>Definition:</b> reduction.h:580</div></div>
<div class="ttc" id="namespacetvm_1_1tir_html_a1a071208bbbab6b220cf46f5cdccdd86"><div class="ttname"><a href="namespacetvm_1_1tir.html#a1a071208bbbab6b220cf46f5cdccdd86">tvm::tir::make_const</a></div><div class="ttdeci">PrimExpr make_const(DataType t, ValueType value, Span span=Span())</div><div class="ttdoc">Make a const value with certain data type. </div><div class="ttdef"><b>Definition:</b> op.h:954</div></div>
<div class="ttc" id="namespacetvm_html"><div class="ttname"><a href="namespacetvm.html">tvm</a></div><div class="ttdoc">runtime implementation for LibTorch/TorchScript. </div><div class="ttdef"><b>Definition:</b> analyzer.h:36</div></div>
<div class="ttc" id="namespacetvm_1_1te_html"><div class="ttname"><a href="namespacetvm_1_1te.html">tvm::te</a></div><div class="ttdoc">Tensor expression language DSL. </div><div class="ttdef"><b>Definition:</b> extracted_task.h:33</div></div>
diff --git a/docs/reference/api/doxygen/layer__norm_8h_source.html b/docs/reference/api/doxygen/layer__norm_8h_source.html
index 4bcd5c012a..5eee2cb351 100644
--- a/docs/reference/api/doxygen/layer__norm_8h_source.html
+++ b/docs/reference/api/doxygen/layer__norm_8h_source.html
@@ -67,7 +67,7 @@ $(function() {
</div><!--header-->
<div class="contents">
<a href="layer__norm_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or more [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_a12a94f9c1103f5e05fb9819e7176bf4c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a12a94f9c1103f5e05fb9819e7176bf4c">tvm::topi::MakeTupleSumReducer</a></div><div class="ttdeci">FCommReduce MakeTupleSumReducer()</div><div class="ttdoc">Create communitive reducer summing over tuples. </div><div class="ttdef"><b>Definition:</b> reduction.h:576</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a12a94f9c1103f5e05fb9819e7176bf4c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a12a94f9c1103f5e05fb9819e7176bf4c">tvm::topi::MakeTupleSumReducer</a></div><div class="ttdeci">FCommReduce MakeTupleSumReducer()</div><div class="ttdoc">Create communitive reducer summing over tuples. </div><div class="ttdef"><b>Definition:</b> reduction.h:580</div></div>
<div class="ttc" id="namespacetvm_1_1tir_html_a1a071208bbbab6b220cf46f5cdccdd86"><div class="ttname"><a href="namespacetvm_1_1tir.html#a1a071208bbbab6b220cf46f5cdccdd86">tvm::tir::make_const</a></div><div class="ttdeci">PrimExpr make_const(DataType t, ValueType value, Span span=Span())</div><div class="ttdoc">Make a const value with certain data type. </div><div class="ttdef"><b>Definition:</b> op.h:954</div></div>
<div class="ttc" id="namespacetvm_html"><div class="ttname"><a href="namespacetvm.html">tvm</a></div><div class="ttdoc">runtime implementation for LibTorch/TorchScript. </div><div class="ttdef"><b>Definition:</b> analyzer.h:36</div></div>
<div class="ttc" id="namespacetvm_1_1te_html"><div class="ttname"><a href="namespacetvm_1_1te.html">tvm::te</a></div><div class="ttdoc">Tensor expression language DSL. </div><div class="ttdef"><b>Definition:</b> extracted_task.h:33</div></div>
diff --git a/docs/reference/api/doxygen/nn_2pooling_8h_source.html b/docs/reference/api/doxygen/nn_2pooling_8h_source.html
index 2561c7ed15..fbbd1598a8 100644
--- a/docs/reference/api/doxygen/nn_2pooling_8h_source.html
+++ b/docs/reference/api/doxygen/nn_2pooling_8h_source.html
@@ -66,14 +66,14 @@ $(function() {
<div class="title">pooling.h</div> </div>
</div><!--header-->
<div class="contents">
-<a href="nn_2pooling_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or more [...]
+<a href="nn_2pooling_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or more [...]
<div class="ttc" id="namespacetvm_1_1topi_html_a938350880b154670bea57cd9ce69d490"><div class="ttname"><a href="namespacetvm_1_1topi.html#a938350880b154670bea57cd9ce69d490">tvm::topi::kCommReduceIdx</a></div><div class="ttdeci">constexpr auto kCommReduceIdx</div><div class="ttdef"><b>Definition:</b> tags.h:35</div></div>
<div class="ttc" id="namespacetvm_1_1topi_1_1nn_html_aca7c280684bfa7f8eb16a4a2ae0891f4"><div class="ttname"><a href="namespacetvm_1_1topi_1_1nn.html#aca7c280684bfa7f8eb16a4a2ae0891f4">tvm::topi::nn::pool1d</a></div><div class="ttdeci">Tensor pool1d(const Tensor &x, const Array< PrimExpr > &kernel_size, const Array< PrimExpr > &stride_size, const Array< PrimExpr > &dilation_size, const Array< PrimExpr > &padding_size, PoolType pool_type, bool ce [...]
<div class="ttc" id="namespacetvm_html_a3b37fa55ea93d6868751a2441996b072"><div class="ttname"><a href="namespacetvm.html#a3b37fa55ea93d6868751a2441996b072">tvm::min_value</a></div><div class="ttdeci">PrimExpr min_value(const DataType &dtype, Span span=Span())</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a4b434e701bc9835e2a7de8f0fadebea5"><div class="ttname"><a href="namespacetvm_1_1topi.html#a4b434e701bc9835e2a7de8f0fadebea5">tvm::topi::MakeArgmaxReducer</a></div><div class="ttdeci">FCommReduce MakeArgmaxReducer(bool select_last_index=false)</div><div class="ttdef"><b>Definition:</b> reduction.h:495</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a4b434e701bc9835e2a7de8f0fadebea5"><div class="ttname"><a href="namespacetvm_1_1topi.html#a4b434e701bc9835e2a7de8f0fadebea5">tvm::topi::MakeArgmaxReducer</a></div><div class="ttdeci">FCommReduce MakeArgmaxReducer(bool select_last_index=false)</div><div class="ttdef"><b>Definition:</b> reduction.h:499</div></div>
<div class="ttc" id="namespacetvm_html_ada5ad8338d3144221d8f16380e6c4855"><div class="ttname"><a href="namespacetvm.html#ada5ad8338d3144221d8f16380e6c4855">tvm::indexmod</a></div><div class="ttdeci">PrimExpr indexmod(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">compute the remainder floor(a / b) where a and b are non-negative. </div></div>
<div class="ttc" id="namespacetvm_1_1tir_html_a1a071208bbbab6b220cf46f5cdccdd86"><div class="ttname"><a href="namespacetvm_1_1tir.html#a1a071208bbbab6b220cf46f5cdccdd86">tvm::tir::make_const</a></div><div class="ttdeci">PrimExpr make_const(DataType t, ValueType value, Span span=Span())</div><div class="ttdoc">Make a const value with certain data type. </div><div class="ttdef"><b>Definition:</b> op.h:954</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_ae488679377c78cd5411b7df11c297673"><div class="ttname"><a href="namespacetvm_1_1topi.html#ae488679377c78cd5411b7df11c297673">tvm::topi::min</a></div><div class="ttdeci">Tensor min(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that finds the minimum of elements over a given axis. </div><div class="ttdef"><b>Definition:</b> reduction.h:410 [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_ae488679377c78cd5411b7df11c297673"><div class="ttname"><a href="namespacetvm_1_1topi.html#ae488679377c78cd5411b7df11c297673">tvm::topi::min</a></div><div class="ttdeci">Tensor min(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that finds the minimum of elements over a given axis. </div><div class="ttdef"><b>Definition:</b> reduction.h:414 [...]
<div class="ttc" id="namespacetvm_html"><div class="ttname"><a href="namespacetvm.html">tvm</a></div><div class="ttdoc">runtime implementation for LibTorch/TorchScript. </div><div class="ttdef"><b>Definition:</b> analyzer.h:36</div></div>
<div class="ttc" id="namespacetvm_1_1topi_1_1nn_html_a3fb74f1c3df6edf17c9a3f1e122f84fe"><div class="ttname"><a href="namespacetvm_1_1topi_1_1nn.html#a3fb74f1c3df6edf17c9a3f1e122f84fe">tvm::topi::nn::adaptive_pool_impl</a></div><div class="ttdeci">Tensor adaptive_pool_impl(const Tensor &x, const Array< PrimExpr > &output_size, PoolType pool_type, const std::vector< int > &axes)</div><div class="ttdoc">Perform adaptive pooling on N dimensional data. </div><div class [...]
<div class="ttc" id="namespacetvm_1_1te_html"><div class="ttname"><a href="namespacetvm_1_1te.html">tvm::te</a></div><div class="ttdoc">Tensor expression language DSL. </div><div class="ttdef"><b>Definition:</b> extracted_task.h:33</div></div>
@@ -103,7 +103,7 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1topi_1_1nn_html_af2bcf6bad18e028ce14ef854974629c0"><div class="ttname"><a href="namespacetvm_1_1topi_1_1nn.html#af2bcf6bad18e028ce14ef854974629c0">tvm::topi::nn::global_pool</a></div><div class="ttdeci">Tensor global_pool(const Tensor &x, PoolType pool_type, const std::string &layout="NCHW")</div><div class="ttdoc">Perform global pooling on height and width dimension of data. It decides the height and width dimensi...</div><div class [...]
<div class="ttc" id="namespacetvm_1_1te_html_aae384e9b73c2271905486e4a74b69265"><div class="ttname"><a href="namespacetvm_1_1te.html#aae384e9b73c2271905486e4a74b69265">tvm::te::reduce_axis</a></div><div class="ttdeci">IterVar reduce_axis(Range dom, std::string name="rv")</div><div class="ttdoc">Create a new IterVar for reduction operations. </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a3305d377f96cd20c23032eeada2756d5"><div class="ttname"><a href="namespacetvm_1_1topi.html#a3305d377f96cd20c23032eeada2756d5">tvm::topi::pad</a></div><div class="ttdeci">tvm::te::Tensor pad(const tvm::te::Tensor &t, const tvm::Array< tvm::PrimExpr > &pad_before, tvm::Array< tvm::PrimExpr > pad_after=tvm::Array< tvm::PrimExpr >(), PrimExpr pad_value=PrimExpr(), std::string name="T_pad", std::string tag=kElem [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_a59e9137c0e02da820906a44c6c9616b7"><div class="ttname"><a href="namespacetvm_1_1topi.html#a59e9137c0e02da820906a44c6c9616b7">tvm::topi::argmax</a></div><div class="ttdeci">Tensor argmax(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false, bool select_last_index=false)</div><div class="ttdoc">Creates an operation that finds the indices of the maximum values over a given axis. </div><div cl [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a59e9137c0e02da820906a44c6c9616b7"><div class="ttname"><a href="namespacetvm_1_1topi.html#a59e9137c0e02da820906a44c6c9616b7">tvm::topi::argmax</a></div><div class="ttdeci">Tensor argmax(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false, bool select_last_index=false)</div><div class="ttdoc">Creates an operation that finds the indices of the maximum values over a given axis. </div><div cl [...]
<div class="ttc" id="namespacetvm_1_1te_html_ae0c71f84710b436cbe0b32289d0838f4"><div class="ttname"><a href="namespacetvm_1_1te.html#ae0c71f84710b436cbe0b32289d0838f4">tvm::te::var</a></div><div class="ttdeci">Var var(std::string name_hint, DataType t=DataType::Int(32))</div><div class="ttdoc">Construct a new Var expression. </div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_a6b097149e69ea03fe3b812a3f5f7fcd9"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#a6b097149e69ea03fe3b812a3f5f7fcd9">tvm::runtime::Array::end</a></div><div class="ttdeci">iterator end() const</div><div class="ttdef"><b>Definition:</b> array.h:390</div></div>
<div class="ttc" id="classtvm_1_1te_1_1Tensor_html"><div class="ttname"><a href="classtvm_1_1te_1_1Tensor.html">tvm::te::Tensor</a></div><div class="ttdoc">Tensor structure representing a possible input, or intermediate computation result. </div><div class="ttdef"><b>Definition:</b> tensor.h:102</div></div>
diff --git a/docs/reference/api/doxygen/reduction_8h_source.html b/docs/reference/api/doxygen/reduction_8h_source.html
index f38fdc5265..51cb33d6d7 100644
--- a/docs/reference/api/doxygen/reduction_8h_source.html
+++ b/docs/reference/api/doxygen/reduction_8h_source.html
@@ -66,28 +66,28 @@ $(function() {
<div class="title">reduction.h</div> </div>
</div><!--header-->
<div class="contents">
-<a href="reduction_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or more c [...]
+<a href="reduction_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> * or more c [...]
<div class="ttc" id="namespacetvm_1_1relay_html_af40ca6124bc2e88f2323eeb79d326cc0"><div class="ttname"><a href="namespacetvm_1_1relay.html#af40ca6124bc2e88f2323eeb79d326cc0">tvm::relay::Span</a></div><div class="ttdeci">tvm::Span Span</div><div class="ttdef"><b>Definition:</b> base.h:65</div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_a1a7727b86efaf35c58a5198ab1c139c8"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#a1a7727b86efaf35c58a5198ab1c139c8">tvm::runtime::Array::reserve</a></div><div class="ttdeci">void reserve(int64_t n)</div><div class="ttdoc">Make sure the list has the capacity of at least n. </div><div class="ttdef"><b>Definition:</b> array.h:569</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a50b46549c2a66bedd305f5e7db0d6d89"><div class="ttname"><a href="namespacetvm_1_1topi.html#a50b46549c2a66bedd305f5e7db0d6d89">tvm::topi::CommReduceIdx</a></div><div class="ttdeci">Tensor CommReduceIdx(const Tensor &data, const Array< Integer > &axis, FCommReduce func, bool keepdims, bool atleast1d)</div><div class="ttdoc">Create an index reduction operation. </div><div class="ttdef"><b>Definition:</b> reduction.h:205</div></div>
<div class="ttc" id="classtvm_1_1tir_1_1CommReducer_html"><div class="ttname"><a href="classtvm_1_1tir_1_1CommReducer.html">tvm::tir::CommReducer</a></div><div class="ttdoc">Managed reference to CommReducerNode. </div><div class="ttdef"><b>Definition:</b> expr.h:1085</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a1d3988f52c26cc29967cefa057b79372"><div class="ttname"><a href="namespacetvm_1_1topi.html#a1d3988f52c26cc29967cefa057b79372">tvm::topi::MakeCommReducer</a></div><div class="ttdeci">FCommReduce MakeCommReducer(FCombine fcombine, FIdentity fidentity, std::string name="reduce")</div><div class="ttdoc">Create a commutative reducer for a reduction. </div><div class="ttdef"><b>Definition:</b> reduction.h:268</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a0055fc9fc2a61a472efd124c80db2c08"><div class="ttname"><a href="namespacetvm_1_1topi.html#a0055fc9fc2a61a472efd124c80db2c08">tvm::topi::collapse_sum</a></div><div class="ttdeci">Tensor collapse_sum(const Tensor &data, Array< PrimExpr > target_shape)</div><div class="ttdef"><b>Definition:</b> reduction.h:331</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a0055fc9fc2a61a472efd124c80db2c08"><div class="ttname"><a href="namespacetvm_1_1topi.html#a0055fc9fc2a61a472efd124c80db2c08">tvm::topi::collapse_sum</a></div><div class="ttdeci">Tensor collapse_sum(const Tensor &data, Array< PrimExpr > target_shape)</div><div class="ttdef"><b>Definition:</b> reduction.h:335</div></div>
<div class="ttc" id="namespacetvm_html_aac2abc149c1a47944c37b560181b15c0"><div class="ttname"><a href="namespacetvm.html#aac2abc149c1a47944c37b560181b15c0">tvm::min</a></div><div class="ttdeci">PrimExpr min(PrimExpr a, PrimExpr b, Span span=Span())</div><div class="ttdoc">take minimum of two values </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a938350880b154670bea57cd9ce69d490"><div class="ttname"><a href="namespacetvm_1_1topi.html#a938350880b154670bea57cd9ce69d490">tvm::topi::kCommReduceIdx</a></div><div class="ttdeci">constexpr auto kCommReduceIdx</div><div class="ttdef"><b>Definition:</b> tags.h:35</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a72ff2d600f5610f32a39b8b5522a4137"><div class="ttname"><a href="namespacetvm_1_1topi.html#a72ff2d600f5610f32a39b8b5522a4137">tvm::topi::FCombine</a></div><div class="ttdeci">std::function< Array< PrimExpr >(Array< Var > lhs, Array< Var > rhs)> FCombine</div><div class="ttdoc">A combiner function for a reduction. </div><div class="ttdef"><b>Definition:</b> reduction.h:254</div></div>
<div class="ttc" id="classtvm_1_1tir_1_1Reduce_html"><div class="ttname"><a href="classtvm_1_1tir_1_1Reduce.html">tvm::tir::Reduce</a></div><div class="ttdoc">Managed reference to ReduceNode. </div><div class="ttdef"><b>Definition:</b> expr.h:1149</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a12a94f9c1103f5e05fb9819e7176bf4c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a12a94f9c1103f5e05fb9819e7176bf4c">tvm::topi::MakeTupleSumReducer</a></div><div class="ttdeci">FCommReduce MakeTupleSumReducer()</div><div class="ttdoc">Create communitive reducer summing over tuples. </div><div class="ttdef"><b>Definition:</b> reduction.h:576</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a12a94f9c1103f5e05fb9819e7176bf4c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a12a94f9c1103f5e05fb9819e7176bf4c">tvm::topi::MakeTupleSumReducer</a></div><div class="ttdeci">FCommReduce MakeTupleSumReducer()</div><div class="ttdoc">Create communitive reducer summing over tuples. </div><div class="ttdef"><b>Definition:</b> reduction.h:580</div></div>
<div class="ttc" id="namespacetvm_html_a3b37fa55ea93d6868751a2441996b072"><div class="ttname"><a href="namespacetvm.html#a3b37fa55ea93d6868751a2441996b072">tvm::min_value</a></div><div class="ttdeci">PrimExpr min_value(const DataType &dtype, Span span=Span())</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a4b434e701bc9835e2a7de8f0fadebea5"><div class="ttname"><a href="namespacetvm_1_1topi.html#a4b434e701bc9835e2a7de8f0fadebea5">tvm::topi::MakeArgmaxReducer</a></div><div class="ttdeci">FCommReduce MakeArgmaxReducer(bool select_last_index=false)</div><div class="ttdef"><b>Definition:</b> reduction.h:495</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a4b434e701bc9835e2a7de8f0fadebea5"><div class="ttname"><a href="namespacetvm_1_1topi.html#a4b434e701bc9835e2a7de8f0fadebea5">tvm::topi::MakeArgmaxReducer</a></div><div class="ttdeci">FCommReduce MakeArgmaxReducer(bool select_last_index=false)</div><div class="ttdef"><b>Definition:</b> reduction.h:499</div></div>
<div class="ttc" id="namespacetvm_1_1tir_html_a1a071208bbbab6b220cf46f5cdccdd86"><div class="ttname"><a href="namespacetvm_1_1tir.html#a1a071208bbbab6b220cf46f5cdccdd86">tvm::tir::make_const</a></div><div class="ttdeci">PrimExpr make_const(DataType t, ValueType value, Span span=Span())</div><div class="ttdoc">Make a const value with certain data type. </div><div class="ttdef"><b>Definition:</b> op.h:954</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_ae488679377c78cd5411b7df11c297673"><div class="ttname"><a href="namespacetvm_1_1topi.html#ae488679377c78cd5411b7df11c297673">tvm::topi::min</a></div><div class="ttdeci">Tensor min(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that finds the minimum of elements over a given axis. </div><div class="ttdef"><b>Definition:</b> reduction.h:410 [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_ae488679377c78cd5411b7df11c297673"><div class="ttname"><a href="namespacetvm_1_1topi.html#ae488679377c78cd5411b7df11c297673">tvm::topi::min</a></div><div class="ttdeci">Tensor min(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that finds the minimum of elements over a given axis. </div><div class="ttdef"><b>Definition:</b> reduction.h:414 [...]
<div class="ttc" id="namespacetvm_html"><div class="ttname"><a href="namespacetvm.html">tvm</a></div><div class="ttdoc">runtime implementation for LibTorch/TorchScript. </div><div class="ttdef"><b>Definition:</b> analyzer.h:36</div></div>
<div class="ttc" id="namespacetvm_1_1te_html"><div class="ttname"><a href="namespacetvm_1_1te.html">tvm::te</a></div><div class="ttdoc">Tensor expression language DSL. </div><div class="ttdef"><b>Definition:</b> extracted_task.h:33</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a86c0ad6fe6717f649f3fba88a3dc4b29"><div class="ttname"><a href="namespacetvm_1_1topi.html#a86c0ad6fe6717f649f3fba88a3dc4b29">tvm::topi::MakeReduceAxes</a></div><div class="ttdeci">Array< IterVar > MakeReduceAxes(const std::vector< int > &real_axis, const Tensor &data)</div><div class="ttdoc">Enumerate the axes for a reduce op. </div><div class="ttdef"><b>Definition:</b> reduction.h:89</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_ad58b3ba5122294bd1eb045b5792c3976"><div class="ttname"><a href="namespacetvm_1_1topi.html#ad58b3ba5122294bd1eb045b5792c3976">tvm::topi::MakeReduceTargetShape</a></div><div class="ttdeci">Array< PrimExpr > MakeReduceTargetShape(const std::vector< int > &real_axis, const Tensor &data, bool keepdims, bool atleast1d)</div><div class="ttdoc">Calculate the target shape for a reduce op. </div><div class="ttdef"><b>Definition:</b [...]
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_aa026b914ee05f81b6c20130b8905f257"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#aa026b914ee05f81b6c20130b8905f257">tvm::runtime::Array::push_back</a></div><div class="ttdeci">void push_back(const T &item)</div><div class="ttdoc">push a new item to the back of the list </div><div class="ttdef"><b>Definition:</b> array.h:457</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a988ca437c8085900c96ff750521af96f"><div class="ttname"><a href="namespacetvm_1_1topi.html#a988ca437c8085900c96ff750521af96f">tvm::topi::MakeArgminReducer</a></div><div class="ttdeci">FCommReduce MakeArgminReducer(bool select_last_index=false)</div><div class="ttdef"><b>Definition:</b> reduction.h:434</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a988ca437c8085900c96ff750521af96f"><div class="ttname"><a href="namespacetvm_1_1topi.html#a988ca437c8085900c96ff750521af96f">tvm::topi::MakeArgminReducer</a></div><div class="ttdeci">FCommReduce MakeArgminReducer(bool select_last_index=false)</div><div class="ttdef"><b>Definition:</b> reduction.h:438</div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_abee7c35e8c15e2e61afe35852dfcb252"><div class="ttname"><a href="namespacetvm_1_1topi.html#abee7c35e8c15e2e61afe35852dfcb252">tvm::topi::sum</a></div><div class="ttdeci">Tensor sum(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that sums array elements over a given axis. </div><div class="ttdef"><b>Definition:</b> reduction.h:326</div></div>
<div class="ttc" id="constant__utils_8h_html"><div class="ttname"><a href="constant__utils_8h.html">constant_utils.h</a></div><div class="ttdoc">Utility functions for handling constants in TVM expressions. </div></div>
<div class="ttc" id="classtvm_1_1Range_html"><div class="ttname"><a href="classtvm_1_1Range.html">tvm::Range</a></div><div class="ttdoc">Range constainer. </div><div class="ttdef"><b>Definition:</b> expr.h:715</div></div>
@@ -108,28 +108,28 @@ $(function() {
<div class="ttc" id="namespacetvm_1_1topi_html_af62dd10dd04c1fbf820581b14498de6e"><div class="ttname"><a href="namespacetvm_1_1topi.html#af62dd10dd04c1fbf820581b14498de6e">tvm::topi::ProdOp</a></div><div class="ttdeci">PrimExpr ProdOp(PrimExpr source, Array< IterVar > axis, Array< PrimExpr > init={}, Span span=Span())</div><div class="ttdoc">Wrap tvm::prod to ensure we get the correct overload. </div><div class="ttdef"><b>Definition:</b> reduction.h:308</div></div>
<div class="ttc" id="namespacetvm_1_1te_html_aae384e9b73c2271905486e4a74b69265"><div class="ttname"><a href="namespacetvm_1_1te.html#aae384e9b73c2271905486e4a74b69265">tvm::te::reduce_axis</a></div><div class="ttdeci">IterVar reduce_axis(Range dom, std::string name="rv")</div><div class="ttdoc">Create a new IterVar for reduction operations. </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_a7b1acf424786ee187f0f19a725b85d8c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a7b1acf424786ee187f0f19a725b85d8c">tvm::topi::kCommReduce</a></div><div class="ttdeci">constexpr auto kCommReduce</div><div class="ttdef"><b>Definition:</b> tags.h:34</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a59e9137c0e02da820906a44c6c9616b7"><div class="ttname"><a href="namespacetvm_1_1topi.html#a59e9137c0e02da820906a44c6c9616b7">tvm::topi::argmax</a></div><div class="ttdeci">Tensor argmax(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false, bool select_last_index=false)</div><div class="ttdoc">Creates an operation that finds the indices of the maximum values over a given axis. </div><div cl [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_a59e9137c0e02da820906a44c6c9616b7"><div class="ttname"><a href="namespacetvm_1_1topi.html#a59e9137c0e02da820906a44c6c9616b7">tvm::topi::argmax</a></div><div class="ttdeci">Tensor argmax(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false, bool select_last_index=false)</div><div class="ttdoc">Creates an operation that finds the indices of the maximum values over a given axis. </div><div cl [...]
<div class="ttc" id="namespacetvm_1_1te_html_ae0c71f84710b436cbe0b32289d0838f4"><div class="ttname"><a href="namespacetvm_1_1te.html#ae0c71f84710b436cbe0b32289d0838f4">tvm::te::var</a></div><div class="ttdeci">Var var(std::string name_hint, DataType t=DataType::Int(32))</div><div class="ttdoc">Construct a new Var expression. </div></div>
<div class="ttc" id="classtvm_1_1runtime_1_1Array_html_a6b097149e69ea03fe3b812a3f5f7fcd9"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html#a6b097149e69ea03fe3b812a3f5f7fcd9">tvm::runtime::Array::end</a></div><div class="ttdeci">iterator end() const</div><div class="ttdef"><b>Definition:</b> array.h:390</div></div>
<div class="ttc" id="classtvm_1_1te_1_1Tensor_html"><div class="ttname"><a href="classtvm_1_1te_1_1Tensor.html">tvm::te::Tensor</a></div><div class="ttdoc">Tensor structure representing a possible input, or intermediate computation result. </div><div class="ttdef"><b>Definition:</b> tensor.h:102</div></div>
<div class="ttc" id="operation_8h_html"><div class="ttname"><a href="operation_8h.html">operation.h</a></div><div class="ttdoc">Operation node can generate one or multiple Tensors. </div></div>
<div class="ttc" id="namespacetvm_html_adeeaff4fb29f75a9da8ff4d67723c693"><div class="ttname"><a href="namespacetvm.html#adeeaff4fb29f75a9da8ff4d67723c693">tvm::all</a></div><div class="ttdeci">PrimExpr all(PrimExpr source, Array< tir::IterVar > axis, Array< PrimExpr > init={}, Span span=Span())</div><div class="ttdoc">logical And of source expression over axis </div></div>
<div class="ttc" id="classtvm_1_1tir_1_1Select_html"><div class="ttname"><a href="classtvm_1_1tir_1_1Select.html">tvm::tir::Select</a></div><div class="ttdoc">Managed reference to SelectNode. </div><div class="ttdef"><b>Definition:</b> expr.h:609</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_a4bc269a40cbdbac3b8b764950820dc8c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a4bc269a40cbdbac3b8b764950820dc8c">tvm::topi::prod</a></div><div class="ttdeci">Tensor prod(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates product operation over given axis. </div><div class="ttdef"><b>Definition:</b> reduction.h:568</div></div>
+<div class="ttc" id="namespacetvm_1_1topi_html_a4bc269a40cbdbac3b8b764950820dc8c"><div class="ttname"><a href="namespacetvm_1_1topi.html#a4bc269a40cbdbac3b8b764950820dc8c">tvm::topi::prod</a></div><div class="ttdeci">Tensor prod(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates product operation over given axis. </div><div class="ttdef"><b>Definition:</b> reduction.h:572</div></div>
<div class="ttc" id="topi_2transform_8h_html"><div class="ttname"><a href="topi_2transform_8h.html">transform.h</a></div><div class="ttdoc">Transform op constructors. </div></div>
<div class="ttc" id="namespacetvm_html_a4f1398024c0af23699447ef910b654b8"><div class="ttname"><a href="namespacetvm.html#a4f1398024c0af23699447ef910b654b8">tvm::max_value</a></div><div class="ttdeci">PrimExpr max_value(const DataType &dtype, Span span=Span())</div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_afb48d90f345698b1b3417bafa1911504"><div class="ttname"><a href="namespacetvm_1_1topi.html#afb48d90f345698b1b3417bafa1911504">tvm::topi::any</a></div><div class="ttdeci">Tensor any(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that computes the logical OR of elements over a given axis. </div><div class="ttdef"><b>Definition:</b> reduction [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_afb48d90f345698b1b3417bafa1911504"><div class="ttname"><a href="namespacetvm_1_1topi.html#afb48d90f345698b1b3417bafa1911504">tvm::topi::any</a></div><div class="ttdeci">Tensor any(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that computes the logical OR of elements over a given axis. </div><div class="ttdef"><b>Definition:</b> reduction [...]
<div class="ttc" id="namespacetvm_1_1topi_html_af0e52ef3c0d8e11bf493d5163033cd0d"><div class="ttname"><a href="namespacetvm_1_1topi.html#af0e52ef3c0d8e11bf493d5163033cd0d">tvm::topi::FReduce</a></div><div class="ttdeci">std::function< PrimExpr(PrimExpr source, const Array< IterVar > &axis, Array< PrimExpr > init, Span span)> FReduce</div><div class="ttdoc">The operation to use for CommReduce. </div><div class="ttdef"><b>Definition:</b> reduction.h:47</div></div>
<div class="ttc" id="tags_8h_html"><div class="ttname"><a href="tags_8h.html">tags.h</a></div><div class="ttdoc">External function interface to rocBLAS libraries. </div></div>
<div class="ttc" id="namespacetvm_1_1te_html_afe4f57aeb3dd5ae9c0b58135e14d67ca"><div class="ttname"><a href="namespacetvm_1_1te.html#afe4f57aeb3dd5ae9c0b58135e14d67ca">tvm::te::compute</a></div><div class="ttdeci">Tensor compute(Array< PrimExpr > shape, FCompute fcompute, std::string name="tensor", std::string tag="", Map< String, ObjectRef > attrs={})</div><div class="ttdoc">Construct a new tensor by computing over shape, using the computation rule: resul [...]
<div class="ttc" id="namespacetvm_1_1topi_html_aa45cdc15f72e867eff29c74b2dffd185"><div class="ttname"><a href="namespacetvm_1_1topi.html#aa45cdc15f72e867eff29c74b2dffd185">tvm::topi::GetRealAxis</a></div><div class="ttdeci">std::vector< int > GetRealAxis(int ndim, const Array< Integer > &axis)</div><div class="ttdoc">Convert a reduction axis which could be empty or have negative elements into a real axis with valid d...</div><div class="ttdef"><b>Definition:</b> reduction [...]
-<div class="ttc" id="namespacetvm_1_1topi_html_ad299ebf7fc4294a1e8391fbfe268dfa5"><div class="ttname"><a href="namespacetvm_1_1topi.html#ad299ebf7fc4294a1e8391fbfe268dfa5">tvm::topi::all</a></div><div class="ttdeci">Tensor all(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that computes the logical AND of elements over a given axis. ...</div><div class="ttdef"><b>Definition:</b> reduc [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_ad299ebf7fc4294a1e8391fbfe268dfa5"><div class="ttname"><a href="namespacetvm_1_1topi.html#ad299ebf7fc4294a1e8391fbfe268dfa5">tvm::topi::all</a></div><div class="ttdeci">Tensor all(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false)</div><div class="ttdoc">Creates an operation that computes the logical AND of elements over a given axis. ...</div><div class="ttdef"><b>Definition:</b> reduc [...]
<div class="ttc" id="broadcast_8h_html"><div class="ttname"><a href="broadcast_8h.html">broadcast.h</a></div><div class="ttdoc">Broadcast op constructions. </div></div>
<div class="ttc" id="classtvm_1_1PrimExpr_html"><div class="ttname"><a href="classtvm_1_1PrimExpr.html">tvm::PrimExpr</a></div><div class="ttdoc">Reference to PrimExprNode. </div><div class="ttdef"><b>Definition:</b> expr.h:114</div></div>
<div class="ttc" id="ravel__unravel_8h_html"><div class="ttname"><a href="ravel__unravel_8h.html">ravel_unravel.h</a></div><div class="ttdoc">Index ravel and unraval operations. </div></div>
<div class="ttc" id="namespacetvm_1_1topi_html_ab7fb7a9f1651970c4ba04a48acdb695f"><div class="ttname"><a href="namespacetvm_1_1topi.html#ab7fb7a9f1651970c4ba04a48acdb695f">tvm::topi::DoCommReduce</a></div><div class="ttdeci">Tensor DoCommReduce(const Tensor &data, FReduce func, const Array< PrimExpr > &target_shape, const std::vector< int > &reduce_axes, const std::vector< int > &squeeze_axes, Span span=Span())</div><div class="ttdoc">Create a reduction [...]
<div class="ttc" id="namespacetvm_html_a32a87ae9eacafb2b5b71b28bcc9ef35e"><div class="ttname"><a href="namespacetvm.html#a32a87ae9eacafb2b5b71b28bcc9ef35e">tvm::prod</a></div><div class="ttdeci">PrimExpr prod(PrimExpr source, Array< tir::IterVar > axis, Array< PrimExpr > init={}, Span span=Span())</div><div class="ttdoc">product of source expression over axis </div></div>
-<div class="ttc" id="namespacetvm_1_1topi_html_aed8fdf7a1568bacd2b2d2dd53192c59e"><div class="ttname"><a href="namespacetvm_1_1topi.html#aed8fdf7a1568bacd2b2d2dd53192c59e">tvm::topi::argmin</a></div><div class="ttdeci">Tensor argmin(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false, bool select_last_index=false)</div><div class="ttdoc">Creates an operation that finds the indices of the minimum values over a given axis. </div><div cl [...]
+<div class="ttc" id="namespacetvm_1_1topi_html_aed8fdf7a1568bacd2b2d2dd53192c59e"><div class="ttname"><a href="namespacetvm_1_1topi.html#aed8fdf7a1568bacd2b2d2dd53192c59e">tvm::topi::argmin</a></div><div class="ttdeci">Tensor argmin(const Tensor &data, const Array< Integer > &axis, bool keepdims=false, bool atleast1d=false, bool select_last_index=false)</div><div class="ttdoc">Creates an operation that finds the indices of the minimum values over a given axis. </div><div cl [...]
<div class="ttc" id="namespacetvm_1_1topi_html_aec9d2c654a75e1be977d159b87a6b8f5"><div class="ttname"><a href="namespacetvm_1_1topi.html#aec9d2c654a75e1be977d159b87a6b8f5">tvm::topi::CommReduce</a></div><div class="ttdeci">Tensor CommReduce(const Tensor &data, const Array< Integer > &axis, FReduce func, bool keepdims, bool atleast1d)</div><div class="ttdoc">Create a reduction operation. </div><div class="ttdef"><b>Definition:</b> reduction.h:182</div></div>
</div><!-- fragment --></div><!-- contents -->
<!-- start footer part -->
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index e217ac8376..45f15d6631 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1621,7 +1621,7 @@ history states as starting point to perform Evolutionary Search).</p></li>
<dl class="py class">
<dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
<dd><p>The search policy that searches in a hierarchical search space defined by sketches.
The policy randomly samples programs from the space defined by sketches and use evolutionary
search to fine-tune them.</p>
@@ -1905,7 +1905,7 @@ Candidates:
<dl class="py function">
<dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
<dd><p>THIS API IS DEPRECATED.</p>
<p>Run auto scheduling search for a task.</p>
<dl class="field-list simple">
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index d8c6386cfa..e4a06d0443 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/06fabe4c5/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
</ul>
</aside>
</section>
@@ -151,7 +151,7 @@
<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
</ul>
</aside>
</section>
@@ -168,7 +168,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index e2b9112a8c..9c6cf262de 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/06fabe4c5/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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 13bcfe0180..cf89c08bac 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/06fabe4c5/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L357">runtime.ts:357</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L357">runtime.ts:357</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">code<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L355">runtime.ts:355</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -177,7 +177,7 @@
<div class="tsd-signature tsd-kind-icon">lanes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L359">runtime.ts:359</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -199,7 +199,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L376">runtime.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L376">runtime.ts:376</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -216,7 +216,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L367">runtime.ts:367</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L367">runtime.ts:367</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/dldevice.html b/docs/reference/api/typedoc/classes/dldevice.html
index c87bf82335..6a4cb4c8e1 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/06fabe4c5/web/src/runtime.ts#L299">runtime.ts:299</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L299">runtime.ts:299</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L297">runtime.ts:297</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -161,7 +161,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L295">runtime.ts:295</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -183,7 +183,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L320">runtime.ts:320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L320">runtime.ts:320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -205,7 +205,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L327">runtime.ts:327</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L327">runtime.ts:327</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/environment.html b/docs/reference/api/typedoc/classes/environment.html
index 786511a7c8..285f14a275 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/06fabe4c5/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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 96e7ba4791..eaa276a984 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/06fabe4c5/web/src/runtime.ts#L50">runtime.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L50">runtime.ts:50</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L47">runtime.ts:47</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L46">runtime.ts:46</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L45">runtime.ts:45</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">webGPUContext<span class="tsd-signature-symbol">:</span> <a href="webgpucontext.html" class="tsd-signature-type">WebGPUContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L48">runtime.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L48">runtime.ts:48</a></li>
</ul>
</aside>
</section>
@@ -203,7 +203,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L77">runtime.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L77">runtime.ts:77</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -226,7 +226,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L67">runtime.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L67">runtime.ts:67</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -243,7 +243,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L85">runtime.ts:85</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L85">runtime.ts:85</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <a href="cachedcallstack.html" class="tsd-signature-type">CachedCallStack</a></h4>
@@ -260,7 +260,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L96">runtime.ts:96</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L96">runtime.ts:96</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -283,7 +283,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L73">runtime.ts:73</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L73">runtime.ts:73</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/instance.html b/docs/reference/api/typedoc/classes/instance.html
index 8ef91bf63f..74b1ea377e 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -161,7 +161,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L844">runtime.ts:844</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L844">runtime.ts:844</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L834">runtime.ts:834</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L834">runtime.ts:834</a></li>
</ul>
</aside>
</section>
@@ -234,7 +234,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L833">runtime.ts:833</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L833">runtime.ts:833</a></li>
</ul>
</aside>
</section>
@@ -251,7 +251,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L973">runtime.ts:973</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L973">runtime.ts:973</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -296,7 +296,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L932">runtime.ts:932</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -318,7 +318,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L901">runtime.ts:901</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L901">runtime.ts:901</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -381,7 +381,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -412,7 +412,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -453,7 +453,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -491,7 +491,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L922">runtime.ts:922</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L922">runtime.ts:922</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -508,7 +508,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -552,7 +552,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L943">runtime.ts:943</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L943">runtime.ts:943</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -577,7 +577,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -609,7 +609,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -640,7 +640,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -672,7 +672,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -695,7 +695,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -729,7 +729,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L986">runtime.ts:986</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L986">runtime.ts:986</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -769,7 +769,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -817,7 +817,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -857,7 +857,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -900,7 +900,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -938,7 +938,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -990,7 +990,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1014,7 +1014,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1046,7 +1046,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1078,7 +1078,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1110,7 +1110,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1141,7 +1141,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L957">runtime.ts:957</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L957">runtime.ts:957</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/memory.html b/docs/reference/api/typedoc/classes/memory.html
index e03168d6da..25a8b3cf98 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/06fabe4c5/web/src/memory.ts#L40">memory.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L32">memory.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L33">memory.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L154">memory.ts:154</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L90">memory.ts:90</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L97">memory.ts:97</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L74">memory.ts:74</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L81">memory.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L104">memory.ts:104</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L132">memory.ts:132</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L145">memory.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L60">memory.ts:60</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L67">memory.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L53">memory.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L114">memory.ts:114</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L124">memory.ts:124</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/memory.ts#L175">memory.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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 132aa4cf47..1f907a93f2 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.html
@@ -119,7 +119,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L614">runtime.ts:614</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L614">runtime.ts:614</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L626">runtime.ts:626</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L626">runtime.ts:626</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -186,7 +186,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L653">runtime.ts:653</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L653">runtime.ts:653</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L641">runtime.ts:641</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L641">runtime.ts:641</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L687">runtime.ts:687</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L687">runtime.ts:687</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ndarray.html b/docs/reference/api/typedoc/classes/ndarray.html
index cbc817d211..f2383cae3a 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/06fabe4c5/web/src/runtime.ts#L401">runtime.ts:401</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L401">runtime.ts:401</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <a href="dldevice.html" class="tsd-signature-type">DLDevice</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L394">runtime.ts:394</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L394">runtime.ts:394</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -173,7 +173,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L390">runtime.ts:390</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L390">runtime.ts:390</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
<div class="tsd-signature tsd-kind-icon">ndim<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L388">runtime.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L388">runtime.ts:388</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
<div class="tsd-signature tsd-kind-icon">shape<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L392">runtime.ts:392</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L392">runtime.ts:392</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -225,7 +225,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L480">runtime.ts:480</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L480">runtime.ts:480</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -258,7 +258,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L524">runtime.ts:524</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L524">runtime.ts:524</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -290,7 +290,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L465">runtime.ts:465</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L465">runtime.ts:465</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -307,7 +307,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L458">runtime.ts:458</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L458">runtime.ts:458</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -339,7 +339,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L584">runtime.ts:584</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L584">runtime.ts:584</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -363,7 +363,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L553">runtime.ts:553</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L553">runtime.ts:553</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/packedfunccell.html b/docs/reference/api/typedoc/classes/packedfunccell.html
index 52270b2a9e..70b675e145 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -117,7 +117,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L248">runtime.ts:248</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L255">runtime.ts:255</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L255">runtime.ts:255</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -163,7 +163,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L264">runtime.ts:264</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L264">runtime.ts:264</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index 6666d88f25..f60ed38b39 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/06fabe4c5/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">unknown</span><span class="tsd-signat [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -201,7 +201,7 @@
<div class="tsd-signature tsd-kind-icon">key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
</ul>
</aside>
</section>
@@ -211,7 +211,7 @@
<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -242,7 +242,7 @@
<div class="tsd-signature tsd-kind-icon">socket<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">WebSocket</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
</ul>
</aside>
</section>
@@ -252,7 +252,7 @@
<div class="tsd-signature tsd-kind-icon">state<span class="tsd-signature-symbol">:</span> <a href="../enums/rpcserverstate.html" class="tsd-signature-type">RPCServerState</a><span class="tsd-signature-symbol"> = RPCServerState.InitHeader</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
</ul>
</aside>
</section>
@@ -262,7 +262,7 @@
<div class="tsd-signature tsd-kind-icon">url<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/classes/runtimecontext.html b/docs/reference/api/typedoc/classes/runtimecontext.html
index 2bb5b43f2c..3bc88cea6d 100644
--- a/docs/reference/api/typedoc/classes/runtimecontext.html
+++ b/docs/reference/api/typedoc/classes/runtimecontext.html
@@ -132,7 +132,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L148">runtime.ts:148</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L148">runtime.ts:148</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Item<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L143">runtime.ts:143</a></li>
</ul>
</aside>
</section>
@@ -182,7 +182,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Size<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L144">runtime.ts:144</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L144">runtime.ts:144</a></li>
</ul>
</aside>
</section>
@@ -192,7 +192,7 @@
<div class="tsd-signature tsd-kind-icon">array<wbr>Make<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L145">runtime.ts:145</a></li>
</ul>
</aside>
</section>
@@ -202,7 +202,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Sys<wbr>Lib<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L146">runtime.ts:146</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L146">runtime.ts:146</a></li>
</ul>
</aside>
</section>
@@ -219,7 +219,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L189">runtime.ts:189</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -263,7 +263,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L163">runtime.ts:163</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L163">runtime.ts:163</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -280,7 +280,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L208">runtime.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L208">runtime.ts:208</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
@@ -309,7 +309,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L157">runtime.ts:157</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -326,7 +326,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L167">runtime.ts:167</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L167">runtime.ts:167</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -343,7 +343,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L198">runtime.ts:198</a></li>
</ul>
</aside>
<h4 class="tsd-type-parameters-title">Type parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index af870a56bb..f1db07ebef 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/06fabe4c5/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -137,7 +137,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L235">runtime.ts:235</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">value<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L233">runtime.ts:233</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L233">runtime.ts:233</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmarray.html b/docs/reference/api/typedoc/classes/tvmarray.html
index e33c91ea27..b8e5a35a1d 100644
--- a/docs/reference/api/typedoc/classes/tvmarray.html
+++ b/docs/reference/api/typedoc/classes/tvmarray.html
@@ -133,7 +133,7 @@
<aside class="tsd-sources">
<p>Overrides <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#constructor">constructor</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L784">runtime.ts:784</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L784">runtime.ts:784</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -162,7 +162,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#ctx">ctx</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#dispose">dispose</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -197,7 +197,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L804">runtime.ts:804</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L804">runtime.ts:804</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -230,7 +230,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#gethandle">getHandle</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L730">runtime.ts:730</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L796">runtime.ts:796</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L796">runtime.ts:796</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -283,7 +283,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typeindex">typeIndex</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -306,7 +306,7 @@
<aside class="tsd-sources">
<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typekey">typeKey</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmobject.html b/docs/reference/api/typedoc/classes/tvmobject.html
index 8db042b857..70d8c7891f 100644
--- a/docs/reference/api/typedoc/classes/tvmobject.html
+++ b/docs/reference/api/typedoc/classes/tvmobject.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">ctx<span class="tsd-signature-symbol">:</span> <a href="runtimecontext.html" class="tsd-signature-type">RuntimeContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L703">runtime.ts:703</a></li>
</ul>
</aside>
</section>
@@ -175,7 +175,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L715">runtime.ts:715</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -192,7 +192,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L730">runtime.ts:730</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L738">runtime.ts:738</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -246,7 +246,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L758">runtime.ts:758</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/webgpucontext.html b/docs/reference/api/typedoc/classes/webgpucontext.html
index acd7a5ebe7..ade804b077 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/06fabe4c5/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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/06fabe4c5/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/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 e6ae77fa8d..5cf05240f2 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/06fabe4c5/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
</ul>
</aside>
</section>
@@ -116,7 +116,7 @@
<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
</ul>
</aside>
</section>
@@ -126,7 +126,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
</ul>
</aside>
</section>
@@ -136,7 +136,7 @@
<div class="tsd-signature tsd-kind-icon">Null<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
</ul>
</aside>
</section>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">TVMBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 12</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
</ul>
</aside>
</section>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">TVMDLTensor<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 7</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">TVMData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">TVMModule<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 9</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">TVMNDArray<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 13</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
</ul>
</aside>
</section>
@@ -196,7 +196,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
</ul>
</aside>
</section>
@@ -206,7 +206,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObjectRValue<wbr>Ref<wbr>Arg<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 14</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
</ul>
</aside>
</section>
@@ -216,7 +216,7 @@
<div class="tsd-signature tsd-kind-icon">TVMOpaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
</ul>
</aside>
</section>
@@ -226,7 +226,7 @@
<div class="tsd-signature tsd-kind-icon">TVMPacked<wbr>Func<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 10</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
</ul>
</aside>
</section>
@@ -236,7 +236,7 @@
<div class="tsd-signature tsd-kind-icon">TVMStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 11</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
</ul>
</aside>
</section>
@@ -246,7 +246,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/aynccallbackcode.html b/docs/reference/api/typedoc/enums/aynccallbackcode.html
index 2c7ecb6a6b..d77d38f023 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/06fabe4c5/web/src/runtime.ts#L812">runtime.ts:812</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L812">runtime.ts:812</a></li>
</ul>
</aside>
</section>
@@ -103,7 +103,7 @@
<div class="tsd-signature tsd-kind-icon">k<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L811">runtime.ts:811</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L811">runtime.ts:811</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/dldatatypecode.html b/docs/reference/api/typedoc/enums/dldatatypecode.html
index f3fd4be353..b5d408de2d 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/06fabe4c5/web/src/runtime.ts#L339">runtime.ts:339</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L339">runtime.ts:339</a></li>
</ul>
</aside>
</section>
@@ -105,7 +105,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L337">runtime.ts:337</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L337">runtime.ts:337</a></li>
</ul>
</aside>
</section>
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">Opaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L340">runtime.ts:340</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L340">runtime.ts:340</a></li>
</ul>
</aside>
</section>
@@ -125,7 +125,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L338">runtime.ts:338</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L338">runtime.ts:338</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/rpcserverstate.html b/docs/reference/api/typedoc/enums/rpcserverstate.html
index f58ebf1f56..c6303c9076 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/06fabe4c5/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
</ul>
</aside>
</section>
@@ -100,7 +100,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<wbr>Key<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Server<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Body<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Header<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">Wait<wbr>For<wbr>Callback<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/sizeof.html b/docs/reference/api/typedoc/enums/sizeof.html
index 96ed3678b7..dde7c5e5bd 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/06fabe4c5/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32 + I32</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">F32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">F64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">I32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
</ul>
</aside>
</section>
@@ -150,7 +150,7 @@
<div class="tsd-signature tsd-kind-icon">I64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
</ul>
</aside>
</section>
@@ -160,7 +160,7 @@
<div class="tsd-signature tsd-kind-icon">TVMValue<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
</ul>
</aside>
</section>
@@ -170,7 +170,7 @@
<div class="tsd-signature tsd-kind-icon">U16<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<div class="tsd-signature tsd-kind-icon">U8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/index.html b/docs/reference/api/typedoc/index.html
index 1ddc67d1a3..de295e9b80 100644
--- a/docs/reference/api/typedoc/index.html
+++ b/docs/reference/api/typedoc/index.html
@@ -182,7 +182,7 @@
<div class="tsd-signature tsd-kind-icon">FObject<wbr>Constructor<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, lib<span class="tsd-signature-symbol">: </span><a href="classes/ffilibrary.html" class="tsd-signature-type">FFILibrary</a>, ctx<span class="tsd-signature-symbol">: </span><a href="classes/runtimecontext.html" class="t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L778">runtime.ts:778</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L778">runtime.ts:778</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Alloc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>shape<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, ndim<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeCode<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeBits<span class="tsd [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -288,7 +288,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>Bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">num [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -332,7 +332,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>To<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>from<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, to<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-sig [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -376,7 +376,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>ToBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</sp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -420,7 +420,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -456,7 +456,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMBackend<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number< [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -508,7 +508,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCFunc<wbr>Set<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ret<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -556,7 +556,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCb<wbr>Arg<wbr>ToReturn<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, code<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span c [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -595,7 +595,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Call<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -651,7 +651,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -687,7 +687,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Get<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span cla [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -726,7 +726,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>List<wbr>Global<wbr>Names<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>outSize<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, outArray<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -765,7 +765,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Register<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, f<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, override<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</spa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -808,7 +808,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMGet<wbr>Last<wbr>Error<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -838,7 +838,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -874,7 +874,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Get<wbr>Function<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, funcName<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, queryImports<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">numbe [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -922,7 +922,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Import<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, dep<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-si [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -962,7 +962,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -998,7 +998,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Get<wbr>Type<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1037,7 +1037,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Index2<wbr>Key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_index<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, out_type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><spa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1076,7 +1076,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Key2<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1115,7 +1115,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMSynchronize<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>deviceType<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, deviceId<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signatur [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1157,7 +1157,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Alloc<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>size<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1193,7 +1193,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Free<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ptr<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1229,7 +1229,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Func<wbr>Create<wbr>FromCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resource<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1269,7 +1269,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>args<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1321,7 +1321,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<wbr>Finalizer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resourceHandle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1357,7 +1357,7 @@
<div class="tsd-signature tsd-kind-icon">GPUPointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1372,7 +1372,7 @@
<div class="tsd-signature tsd-kind-icon">Packed<wbr>Func<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">...</span>args<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol"> & </span><a href="interfaces/disp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L37">runtime.ts:37</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L37">runtime.ts:37</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1387,7 +1387,7 @@
<div class="tsd-signature tsd-kind-icon">Pointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1402,7 +1402,7 @@
<div class="tsd-signature tsd-kind-icon">Ptr<wbr>Offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1417,7 +1417,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Base<span class="tsd-signature-symbol">:</span> <a href="classes/tvmobject.html" class="tsd-signature-type">TVMObject</a><span class="tsd-signature-symbol"> | </span><a href="classes/ndarray.html" class="tsd-signature-type">NDArray</a><span class="tsd-signature-symbol"> | </span><a href="classes/module.html" class="tsd-signature-type">Module</a><span class="tsd-signature-symbol"> | </span><a href="index.html#packedfunc" class="t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L781">runtime.ts:781</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L781">runtime.ts:781</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1435,7 +1435,7 @@
<div class="tsd-signature tsd-kind-icon">RPC_<wbr>MAGIC<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">1045105</span><span class="tsd-signature-symbol"> = 1045105</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1457,7 +1457,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/support.ts#L25">support.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/support.ts#L25">support.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1489,7 +1489,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/support.ts#L39">support.ts:39</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/support.ts#L39">support.ts:39</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1518,7 +1518,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/support.ts#L52">support.ts:52</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/support.ts#L52">support.ts:52</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1555,7 +1555,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/compact.ts#L38">compact.ts:38</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/compact.ts#L38">compact.ts:38</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1586,7 +1586,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1608,7 +1608,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/environment.ts#L32">environment.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/environment.ts#L32">environment.ts:32</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1639,7 +1639,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/compact.ts#L24">compact.ts:24</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/compact.ts#L24">compact.ts:24</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1661,7 +1661,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L1749">runtime.ts:1749</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L1749">runtime.ts:1749</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1726,7 +1726,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/support.ts#L62">support.ts:62</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/support.ts#L62">support.ts:62</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1748,7 +1748,7 @@
<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<wbr>Code<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L343">runtime.ts:343</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L343">runtime.ts:343</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1757,7 +1757,7 @@
<div class="tsd-signature tsd-kind-icon">0<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "int"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L344">runtime.ts:344</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L344">runtime.ts:344</a></li>
</ul>
</aside>
</section>
@@ -1767,7 +1767,7 @@
<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "uint"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L345">runtime.ts:345</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L345">runtime.ts:345</a></li>
</ul>
</aside>
</section>
@@ -1777,7 +1777,7 @@
<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "float"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L346">runtime.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L346">runtime.ts:346</a></li>
</ul>
</aside>
</section>
@@ -1787,7 +1787,7 @@
<div class="tsd-signature tsd-kind-icon">3<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "handle"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L347">runtime.ts:347</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L347">runtime.ts:347</a></li>
</ul>
</aside>
</section>
@@ -1798,7 +1798,7 @@
<div class="tsd-signature tsd-kind-icon">Device<wbr>Enum<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L272">runtime.ts:272</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L272">runtime.ts:272</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1807,7 +1807,7 @@
<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "cpu"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L273">runtime.ts:273</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L273">runtime.ts:273</a></li>
</ul>
</aside>
</section>
@@ -1817,7 +1817,7 @@
<div class="tsd-signature tsd-kind-icon">15<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "webgpu"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L277">runtime.ts:277</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L277">runtime.ts:277</a></li>
</ul>
</aside>
</section>
@@ -1827,7 +1827,7 @@
<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "cuda"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L274">runtime.ts:274</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L274">runtime.ts:274</a></li>
</ul>
</aside>
</section>
@@ -1837,7 +1837,7 @@
<div class="tsd-signature tsd-kind-icon">4<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "opencl"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L275">runtime.ts:275</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L275">runtime.ts:275</a></li>
</ul>
</aside>
</section>
@@ -1847,7 +1847,7 @@
<div class="tsd-signature tsd-kind-icon">8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "metal"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L276">runtime.ts:276</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L276">runtime.ts:276</a></li>
</ul>
</aside>
</section>
@@ -1858,7 +1858,7 @@
<div class="tsd-signature tsd-kind-icon">Device<wbr>Str<wbr>ToEnum<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L280">runtime.ts:280</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L280">runtime.ts:280</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1867,7 +1867,7 @@
<div class="tsd-signature tsd-kind-icon">cl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L283">runtime.ts:283</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L283">runtime.ts:283</a></li>
</ul>
</aside>
</section>
@@ -1877,7 +1877,7 @@
<div class="tsd-signature tsd-kind-icon">cpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L281">runtime.ts:281</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L281">runtime.ts:281</a></li>
</ul>
</aside>
</section>
@@ -1887,7 +1887,7 @@
<div class="tsd-signature tsd-kind-icon">cuda<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L282">runtime.ts:282</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L282">runtime.ts:282</a></li>
</ul>
</aside>
</section>
@@ -1897,7 +1897,7 @@
<div class="tsd-signature tsd-kind-icon">metal<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L286">runtime.ts:286</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L286">runtime.ts:286</a></li>
</ul>
</aside>
</section>
@@ -1907,7 +1907,7 @@
<div class="tsd-signature tsd-kind-icon">opencl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L284">runtime.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L284">runtime.ts:284</a></li>
</ul>
</aside>
</section>
@@ -1917,7 +1917,7 @@
<div class="tsd-signature tsd-kind-icon">vulkan<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 7</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L285">runtime.ts:285</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L285">runtime.ts:285</a></li>
</ul>
</aside>
</section>
@@ -1927,7 +1927,7 @@
<div class="tsd-signature tsd-kind-icon">webgpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 15</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/runtime.ts#L287">runtime.ts:287</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/runtime.ts#L287">runtime.ts:287</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/interfaces/disposable.html b/docs/reference/api/typedoc/interfaces/disposable.html
index d49c3b9200..7edb0dc476 100644
--- a/docs/reference/api/typedoc/interfaces/disposable.html
+++ b/docs/reference/api/typedoc/interfaces/disposable.html
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">dispose<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/types.ts#L52">types.ts:52</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/types.ts#L52">types.ts:52</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/interfaces/functioninfo.html b/docs/reference/api/typedoc/interfaces/functioninfo.html
index e7030e59fd..9ac43329be 100644
--- a/docs/reference/api/typedoc/interfaces/functioninfo.html
+++ b/docs/reference/api/typedoc/interfaces/functioninfo.html
@@ -95,7 +95,7 @@
<div class="tsd-signature tsd-kind-icon">arg_<wbr>types<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
</ul>
</aside>
</section>
@@ -105,7 +105,7 @@
<div class="tsd-signature tsd-kind-icon">launch_<wbr>param_<wbr>tags<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
</ul>
</aside>
</section>
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">name<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/interfaces/libraryprovider.html b/docs/reference/api/typedoc/interfaces/libraryprovider.html
index 7d76d7beba..91b522bd34 100644
--- a/docs/reference/api/typedoc/interfaces/libraryprovider.html
+++ b/docs/reference/api/typedoc/interfaces/libraryprovider.html
@@ -112,7 +112,7 @@
<div class="tsd-signature tsd-kind-icon">imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/types.ts#L34">types.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/types.ts#L34">types.ts:34</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -127,7 +127,7 @@
<div class="tsd-signature tsd-kind-icon">start<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>inst<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">Instance</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/06fabe4c5/web/src/types.ts#L39">types.ts:39</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/ff12a2032/web/src/types.ts#L39">types.ts:39</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/searchindex.js b/docs/searchindex.js
index 25687e1211..8aac71a66f 100644
--- a/docs/searchindex.js
+++ b/docs/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
+Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
diff --git a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
index 398d6e9ab1..cd2679f504 100644
--- a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:31.022</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
+<p><strong>00:31.463</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 82%" />
@@ -354,7 +354,7 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-relay-vta-py"><span class="std std-ref">Auto-tuning a convolutional network on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_vta.py</span></code>)</p></td>
-<td><p>00:31.016</p></td>
+<td><p>00:31.456</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_alu_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-alu-vta-py"><span class="std std-ref">Auto-tuning a ALU fused op on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_alu_vta.py</span></code>)</p></td>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_classification.html b/docs/topic/vta/tutorials/frontend/deploy_classification.html
index eee73b2ab3..75247d5f2f 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_classification.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_classification.html
@@ -588,7 +588,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
-resnet18_v1 inference graph built in 32.96s!
+resnet18_v1 inference graph built in 34.04s!
</pre></div>
</div>
</div>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_detection.html b/docs/topic/vta/tutorials/frontend/deploy_detection.html
index 5d5c3b48ef..0ac92bb7bf 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_detection.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_detection.html
@@ -606,7 +606,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
-yolov3-tiny inference graph built in 22.67s!
+yolov3-tiny inference graph built in 22.89s!
</pre></div>
</div>
</div>
diff --git a/docs/topic/vta/tutorials/frontend/sg_execution_times.html b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
index 0c009707e9..c07352fbc3 100644
--- a/docs/topic/vta/tutorials/frontend/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-frontend-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>01:39.194</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
+<p><strong>01:40.731</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_classification.html#sphx-glr-topic-vta-tutorials-frontend-deploy-classification-py"><span class="std std-ref">Deploy Pretrained Vision Model from MxNet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_classification.py</span></code>)</p></td>
-<td><p>00:49.715</p></td>
+<td><p>00:50.884</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_detection.html#sphx-glr-topic-vta-tutorials-frontend-deploy-detection-py"><span class="std std-ref">Deploy Pretrained Vision Detection Model from Darknet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_detection.py</span></code>)</p></td>
-<td><p>00:49.479</p></td>
+<td><p>00:49.848</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/topic/vta/tutorials/optimize/sg_execution_times.html b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
index 9d3e16117c..e958eb0011 100644
--- a/docs/topic/vta/tutorials/optimize/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-optimize-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:03.153</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
+<p><strong>00:03.183</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="convolution_opt.html#sphx-glr-topic-vta-tutorials-optimize-convolution-opt-py"><span class="std std-ref">2D Convolution Optimization</span></a> (<code class="docutils literal notranslate"><span class="pre">convolution_opt.py</span></code>)</p></td>
-<td><p>00:02.694</p></td>
+<td><p>00:02.725</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="matrix_multiply_opt.html#sphx-glr-topic-vta-tutorials-optimize-matrix-multiply-opt-py"><span class="std std-ref">Matrix Multiply Blocking</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply_opt.py</span></code>)</p></td>
-<td><p>00:00.459</p></td>
+<td><p>00:00.458</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/topic/vta/tutorials/sg_execution_times.html b/docs/topic/vta/tutorials/sg_execution_times.html
index fbc3669ef3..f74e65e2a3 100644
--- a/docs/topic/vta/tutorials/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:00.756</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
+<p><strong>00:00.766</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -354,11 +354,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="matrix_multiply.html#sphx-glr-topic-vta-tutorials-matrix-multiply-py"><span class="std std-ref">Simple Matrix Multiply</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply.py</span></code>)</p></td>
-<td><p>00:00.392</p></td>
+<td><p>00:00.397</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="vta_get_started.html#sphx-glr-topic-vta-tutorials-vta-get-started-py"><span class="std std-ref">Get Started with VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">vta_get_started.py</span></code>)</p></td>
-<td><p>00:00.364</p></td>
+<td><p>00:00.369</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/tutorial/auto_scheduler_matmul_x86.html b/docs/tutorial/auto_scheduler_matmul_x86.html
index 893f2ab75c..02b0850dc0 100644
--- a/docs/tutorial/auto_scheduler_matmul_x86.html
+++ b/docs/tutorial/auto_scheduler_matmul_x86.html
@@ -574,7 +574,7 @@ class Module:
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 98.147 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 95.234 ms
</pre></div>
</div>
</div>
@@ -636,7 +636,6 @@ resume the status and do more 5 trials.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Resume search:
-.T
</pre></div>
</div>
</div>
@@ -647,7 +646,7 @@ automatically optimize a matrix multiplication, without the need to specify a
search template. It ends a series of examples that starts from the Tensor
Expression (TE) language that demonstrates how TVM can optimize computational
operations.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 30.840 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 22.267 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-auto-scheduler-matmul-x86-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../_downloads/eac4389b114db015e95cb3cdf8b86b83/auto_scheduler_matmul_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">auto_scheduler_matmul_x86.py</span></code></a></p>
diff --git a/docs/tutorial/autotvm_matmul_x86.html b/docs/tutorial/autotvm_matmul_x86.html
index 803afa6fb3..9cc8059ccf 100644
--- a/docs/tutorial/autotvm_matmul_x86.html
+++ b/docs/tutorial/autotvm_matmul_x86.html
@@ -685,16 +685,16 @@ reduce variance, we take 5 measurements and average them.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>waiting for device...
device available
Get devices for measurement successfully!
-No: 1 GFLOPS: 3.09/3.09 result: MeasureResult(costs=(0.0867567328,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6469478607177734, timestamp=1678684460.3844144) [('tile_y', [-1, 256]), ('tile_x', [-1, 8])],None,38
-No: 2 GFLOPS: 2.79/3.09 result: MeasureResult(costs=(0.09607420879999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8033630847930908, timestamp=1678684463.419324) [('tile_y', [-1, 2]), ('tile_x', [-1, 16])],None,41
-No: 3 GFLOPS: 11.51/11.51 result: MeasureResult(costs=(0.0233286196,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6122231483459473, timestamp=1678684464.0537403) [('tile_y', [-1, 4]), ('tile_x', [-1, 512])],None,92
-No: 4 GFLOPS: 8.81/11.51 result: MeasureResult(costs=(0.0304696042,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7071022987365723, timestamp=1678684466.0268316) [('tile_y', [-1, 1]), ('tile_x', [-1, 32])],None,50
-No: 5 GFLOPS: 1.36/11.51 result: MeasureResult(costs=(0.19755499,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.389230489730835, timestamp=1678684469.6100347) [('tile_y', [-1, 1]), ('tile_x', [-1, 2])],None,10
-No: 6 GFLOPS: 1.52/11.51 result: MeasureResult(costs=(0.17613902080000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.0493390560150146, timestamp=1678684473.9333296) [('tile_y', [-1, 1]), ('tile_x', [-1, 1])],None,0
-No: 7 GFLOPS: 14.42/14.42 result: MeasureResult(costs=(0.0186158342,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5975933074951172, timestamp=1678684474.4855804) [('tile_y', [-1, 32]), ('tile_x', [-1, 64])],None,65
-No: 8 GFLOPS: 2.34/14.42 result: MeasureResult(costs=(0.11476816100000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0721683502197266, timestamp=1678684476.583552) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
-No: 9 GFLOPS: 0.51/14.42 result: MeasureResult(costs=(0.5273584632,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.644187450408936, timestamp=1678684485.342885) [('tile_y', [-1, 128]), ('tile_x', [-1, 1])],None,7
-No: 10 GFLOPS: 2.13/14.42 result: MeasureResult(costs=(0.1262303024,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2271721363067627, timestamp=1678684487.6201863) [('tile_y', [-1, 2]), ('tile_x', [-1, 4])],None,21
+No: 1 GFLOPS: 4.50/4.50 result: MeasureResult(costs=(0.05967109,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2330536842346191, timestamp=1678750593.585393) [('tile_y', [-1, 16]), ('tile_x', [-1, 16])],None,44
+No: 2 GFLOPS: 11.28/11.28 result: MeasureResult(costs=(0.0237999648,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6429798603057861, timestamp=1678750595.485118) [('tile_y', [-1, 256]), ('tile_x', [-1, 256])],None,88
+No: 3 GFLOPS: 0.87/11.28 result: MeasureResult(costs=(0.3071478688,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.141611099243164, timestamp=1678750600.6574879) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
+No: 4 GFLOPS: 11.02/11.28 result: MeasureResult(costs=(0.0243695036,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6250784397125244, timestamp=1678750602.567924) [('tile_y', [-1, 2]), ('tile_x', [-1, 512])],None,91
+No: 5 GFLOPS: 10.51/11.28 result: MeasureResult(costs=(0.025530940600000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6833343505859375, timestamp=1678750603.364809) [('tile_y', [-1, 8]), ('tile_x', [-1, 32])],None,53
+No: 6 GFLOPS: 1.87/11.28 result: MeasureResult(costs=(0.14336039579999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.532989501953125, timestamp=1678750607.1755967) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
+No: 7 GFLOPS: 11.46/11.46 result: MeasureResult(costs=(0.0234165186,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6113460063934326, timestamp=1678750607.8100495) [('tile_y', [-1, 256]), ('tile_x', [-1, 512])],None,98
+No: 8 GFLOPS: 2.83/11.46 result: MeasureResult(costs=(0.0948549296,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7708613872528076, timestamp=1678750609.5896919) [('tile_y', [-1, 512]), ('tile_x', [-1, 16])],None,49
+No: 9 GFLOPS: 10.18/11.46 result: MeasureResult(costs=(0.026369689599999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7020399570465088, timestamp=1678750610.4075158) [('tile_y', [-1, 512]), ('tile_x', [-1, 128])],None,79
+No: 10 GFLOPS: 3.97/11.46 result: MeasureResult(costs=(0.06769754219999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3485820293426514, timestamp=1678750611.752353) [('tile_y', [-1, 64]), ('tile_x', [-1, 16])],None,46
</pre></div>
</div>
<p>With tuning completed, we can choose the configuration from the log file that
diff --git a/docs/tutorial/autotvm_relay_x86.html b/docs/tutorial/autotvm_relay_x86.html
index 353fec36b0..14bee4ea55 100644
--- a/docs/tutorial/autotvm_relay_x86.html
+++ b/docs/tutorial/autotvm_relay_x86.html
@@ -563,7 +563,7 @@ standard deviation.</p>
<span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">unoptimized</span></a><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{'mean': 511.0277086200017, 'median': 510.5724576499995, 'std': 1.9596086703937687}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{'mean': 550.6390348200011, 'median': 546.4105448499993, 'std': 15.583710586461082}
</pre></div>
</div>
</div>
@@ -715,178 +715,179 @@ depending on the specifics of the model and the target platform.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 1/25] Current/Best: 16.06/ 23.55 GFLOPS | Progress: (4/20) | 10.56 s
-[Task 1/25] Current/Best: 14.89/ 23.55 GFLOPS | Progress: (8/20) | 14.52 s
-[Task 1/25] Current/Best: 11.12/ 23.55 GFLOPS | Progress: (12/20) | 16.52 s
-[Task 1/25] Current/Best: 7.11/ 23.55 GFLOPS | Progress: (16/20) | 20.48 s
-[Task 1/25] Current/Best: 12.93/ 23.55 GFLOPS | Progress: (20/20) | 23.18 s Done.
+[Task 1/25] Current/Best: 6.88/ 13.97 GFLOPS | Progress: (4/20) | 14.98 s
+[Task 1/25] Current/Best: 14.11/ 14.18 GFLOPS | Progress: (8/20) | 20.61 s
+[Task 1/25] Current/Best: 3.36/ 20.97 GFLOPS | Progress: (12/20) | 23.58 s
+[Task 1/25] Current/Best: 9.26/ 20.97 GFLOPS | Progress: (16/20) | 26.47 s
+[Task 1/25] Current/Best: 9.01/ 20.97 GFLOPS | Progress: (20/20) | 30.34 s Done.
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 2/25] Current/Best: 14.74/ 15.05 GFLOPS | Progress: (4/20) | 4.72 s
-[Task 2/25] Current/Best: 12.25/ 20.51 GFLOPS | Progress: (8/20) | 6.17 s
-[Task 2/25] Current/Best: 12.09/ 20.51 GFLOPS | Progress: (12/20) | 7.83 s
-[Task 2/25] Current/Best: 14.55/ 20.51 GFLOPS | Progress: (16/20) | 9.49 s
-[Task 2/25] Current/Best: 11.69/ 21.86 GFLOPS | Progress: (20/20) | 11.46 s Done.
+[Task 2/25] Current/Best: 5.12/ 13.63 GFLOPS | Progress: (4/20) | 5.08 s
+[Task 2/25] Current/Best: 8.47/ 16.78 GFLOPS | Progress: (8/20) | 7.01 s
+[Task 2/25] Current/Best: 6.53/ 16.78 GFLOPS | Progress: (12/20) | 8.74 s
+[Task 2/25] Current/Best: 7.92/ 16.78 GFLOPS | Progress: (16/20) | 10.77 s
+[Task 2/25] Current/Best: 10.26/ 16.78 GFLOPS | Progress: (20/20) | 12.78 s Done.
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 3/25] Current/Best: 16.56/ 20.38 GFLOPS | Progress: (4/20) | 5.02 s
-[Task 3/25] Current/Best: 5.63/ 20.38 GFLOPS | Progress: (8/20) | 7.59 s
-[Task 3/25] Current/Best: 11.59/ 23.04 GFLOPS | Progress: (12/20) | 11.15 s
-[Task 3/25] Current/Best: 11.36/ 23.04 GFLOPS | Progress: (16/20) | 13.14 s
-[Task 3/25] Current/Best: 11.88/ 23.04 GFLOPS | Progress: (20/20) | 15.53 s Done.
+[Task 3/25] Current/Best: 11.41/ 23.22 GFLOPS | Progress: (4/20) | 6.79 s
+[Task 3/25] Current/Best: 12.29/ 23.22 GFLOPS | Progress: (8/20) | 9.52 s
+[Task 3/25] Current/Best: 5.93/ 23.22 GFLOPS | Progress: (12/20) | 12.03 s
+[Task 3/25] Current/Best: 13.69/ 23.22 GFLOPS | Progress: (16/20) | 14.40 s
+[Task 3/25] Current/Best: 6.19/ 23.22 GFLOPS | Progress: (20/20) | 17.10 s Done.
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 4/25] Current/Best: 14.00/ 14.33 GFLOPS | Progress: (4/20) | 6.20 s
-[Task 4/25] Current/Best: 16.36/ 18.17 GFLOPS | Progress: (8/20) | 9.07 s
-[Task 4/25] Current/Best: 17.88/ 18.22 GFLOPS | Progress: (12/20) | 10.78 s
-[Task 4/25] Current/Best: 11.35/ 20.80 GFLOPS | Progress: (16/20) | 13.55 s
-[Task 4/25] Current/Best: 5.37/ 20.80 GFLOPS | Progress: (20/20) | 16.14 s Done.
-
+[Task 4/25] Current/Best: 17.38/ 17.38 GFLOPS | Progress: (4/20) | 5.04 s
+[Task 4/25] Current/Best: 12.93/ 17.38 GFLOPS | Progress: (8/20) | 6.82 s
+[Task 4/25] Current/Best: 11.67/ 17.38 GFLOPS | Progress: (12/20) | 17.96 s
+[Task 4/25] Current/Best: 9.96/ 17.38 GFLOPS | Progress: (16/20) | 23.66 s
+[Task 4/25] Current/Best: 7.27/ 17.38 GFLOPS | Progress: (20/20) | 25.91 s
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 5/25] Current/Best: 14.35/ 14.35 GFLOPS | Progress: (4/20) | 4.77 s
-[Task 5/25] Current/Best: 5.38/ 14.35 GFLOPS | Progress: (8/20) | 7.03 s
-[Task 5/25] Current/Best: 14.31/ 14.55 GFLOPS | Progress: (12/20) | 9.46 s
-[Task 5/25] Current/Best: 11.29/ 14.55 GFLOPS | Progress: (16/20) | 13.15 s
-[Task 5/25] Current/Best: 5.36/ 14.55 GFLOPS | Progress: (20/20) | 16.08 s Done.
+[Task 5/25] Current/Best: 16.16/ 16.39 GFLOPS | Progress: (4/20) | 4.89 s
+[Task 5/25] Current/Best: 9.60/ 16.39 GFLOPS | Progress: (8/20) | 7.06 s
+[Task 5/25] Current/Best: 16.30/ 17.47 GFLOPS | Progress: (12/20) | 8.90 s
+[Task 5/25] Current/Best: 13.61/ 23.02 GFLOPS | Progress: (16/20) | 10.98 s
+[Task 5/25] Current/Best: 5.29/ 23.02 GFLOPS | Progress: (20/20) | 13.64 s Done.
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 6/25] Current/Best: 14.54/ 14.54 GFLOPS | Progress: (4/20) | 5.87 s
-[Task 6/25] Current/Best: 15.63/ 17.99 GFLOPS | Progress: (8/20) | 7.99 s
-[Task 6/25] Current/Best: 4.13/ 18.47 GFLOPS | Progress: (12/20) | 11.01 s
-[Task 6/25] Current/Best: 19.82/ 22.01 GFLOPS | Progress: (16/20) | 15.05 s
-[Task 6/25] Current/Best: 15.13/ 22.01 GFLOPS | Progress: (20/20) | 17.07 s Done.
+[Task 6/25] Current/Best: 3.26/ 11.22 GFLOPS | Progress: (4/20) | 7.49 s Done.
+
+[Task 6/25] Current/Best: 14.05/ 14.18 GFLOPS | Progress: (8/20) | 11.09 s
+[Task 6/25] Current/Best: 3.47/ 14.18 GFLOPS | Progress: (12/20) | 14.61 s
+[Task 6/25] Current/Best: 15.69/ 15.69 GFLOPS | Progress: (16/20) | 18.25 s
+[Task 6/25] Current/Best: 6.84/ 18.08 GFLOPS | Progress: (20/20) | 20.57 s Done.
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 7/25] Current/Best: 12.65/ 17.95 GFLOPS | Progress: (4/20) | 5.27 s
-[Task 7/25] Current/Best: 20.43/ 20.43 GFLOPS | Progress: (8/20) | 7.15 s
-[Task 7/25] Current/Best: 4.70/ 20.43 GFLOPS | Progress: (12/20) | 10.03 s
-[Task 7/25] Current/Best: 6.45/ 20.43 GFLOPS | Progress: (16/20) | 12.35 s
-[Task 7/25] Current/Best: 19.63/ 20.43 GFLOPS | Progress: (20/20) | 14.76 s Done.
+[Task 7/25] Current/Best: 6.34/ 11.67 GFLOPS | Progress: (4/20) | 5.58 s
+[Task 7/25] Current/Best: 12.30/ 14.04 GFLOPS | Progress: (8/20) | 9.25 s
+[Task 7/25] Current/Best: 1.59/ 14.04 GFLOPS | Progress: (12/20) | 13.39 s
+[Task 7/25] Current/Best: 5.58/ 19.05 GFLOPS | Progress: (16/20) | 16.36 s
+[Task 7/25] Current/Best: 14.41/ 19.53 GFLOPS | Progress: (20/20) | 18.55 s Done.
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 8/25] Current/Best: 12.53/ 18.50 GFLOPS | Progress: (4/20) | 5.03 s
-[Task 8/25] Current/Best: 13.68/ 18.50 GFLOPS | Progress: (8/20) | 8.29 s
-[Task 8/25] Current/Best: 10.80/ 18.50 GFLOPS | Progress: (12/20) | 15.61 s
-[Task 8/25] Current/Best: 3.30/ 18.50 GFLOPS | Progress: (16/20) | 19.57 s
-[Task 8/25] Current/Best: 12.06/ 18.50 GFLOPS | Progress: (20/20) | 25.41 s Done.
+[Task 8/25] Current/Best: 8.34/ 14.87 GFLOPS | Progress: (4/20) | 13.94 s
+[Task 8/25] Current/Best: 12.72/ 14.87 GFLOPS | Progress: (8/20) | 16.67 s
+[Task 8/25] Current/Best: 15.83/ 15.83 GFLOPS | Progress: (12/20) | 19.11 s
+[Task 8/25] Current/Best: 12.36/ 17.49 GFLOPS | Progress: (16/20) | 22.31 s
+[Task 8/25] Current/Best: 3.39/ 17.49 GFLOPS | Progress: (20/20) | 29.62 s Done.
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 9/25] Current/Best: 10.55/ 17.08 GFLOPS | Progress: (4/20) | 5.24 s
-[Task 9/25] Current/Best: 11.86/ 18.35 GFLOPS | Progress: (8/20) | 8.32 s
-[Task 9/25] Current/Best: 18.11/ 18.35 GFLOPS | Progress: (12/20) | 11.17 s
-[Task 9/25] Current/Best: 12.59/ 18.35 GFLOPS | Progress: (16/20) | 14.03 s
-[Task 9/25] Current/Best: 13.13/ 18.35 GFLOPS | Progress: (20/20) | 16.88 s Done.
+[Task 9/25] Current/Best: 9.24/ 20.66 GFLOPS | Progress: (4/20) | 8.00 s
+[Task 9/25] Current/Best: 11.95/ 20.66 GFLOPS | Progress: (8/20) | 10.48 s
+[Task 9/25] Current/Best: 8.18/ 20.78 GFLOPS | Progress: (12/20) | 12.18 s
+[Task 9/25] Current/Best: 10.44/ 20.78 GFLOPS | Progress: (16/20) | 14.18 s
+[Task 9/25] Current/Best: 18.30/ 20.78 GFLOPS | Progress: (20/20) | 16.21 s Done.
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 10/25] Current/Best: 14.61/ 19.91 GFLOPS | Progress: (4/20) | 4.70 s
-[Task 10/25] Current/Best: 13.49/ 20.88 GFLOPS | Progress: (8/20) | 7.63 s
-[Task 10/25] Current/Best: 20.69/ 20.88 GFLOPS | Progress: (12/20) | 10.77 s
-[Task 10/25] Current/Best: 16.03/ 20.88 GFLOPS | Progress: (16/20) | 13.74 s
-[Task 10/25] Current/Best: 11.78/ 20.88 GFLOPS | Progress: (20/20) | 17.26 s Done.
+[Task 10/25] Current/Best: 13.73/ 15.29 GFLOPS | Progress: (4/20) | 5.02 s
+[Task 10/25] Current/Best: 4.42/ 15.29 GFLOPS | Progress: (8/20) | 8.77 s
+[Task 10/25] Current/Best: 17.72/ 22.73 GFLOPS | Progress: (12/20) | 10.84 s
+[Task 10/25] Current/Best: 11.84/ 22.73 GFLOPS | Progress: (16/20) | 12.57 s
+[Task 10/25] Current/Best: 15.79/ 22.73 GFLOPS | Progress: (20/20) | 15.73 s Done.
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 11/25] Current/Best: 22.59/ 23.73 GFLOPS | Progress: (4/20) | 5.68 s
-[Task 11/25] Current/Best: 12.70/ 23.73 GFLOPS | Progress: (8/20) | 8.40 s
-[Task 11/25] Current/Best: 18.17/ 23.73 GFLOPS | Progress: (12/20) | 11.53 s
-[Task 11/25] Current/Best: 16.20/ 23.73 GFLOPS | Progress: (16/20) | 13.89 s
-[Task 11/25] Current/Best: 21.83/ 23.73 GFLOPS | Progress: (20/20) | 16.19 s Done.
+[Task 11/25] Current/Best: 7.79/ 20.71 GFLOPS | Progress: (4/20) | 5.53 s
+[Task 11/25] Current/Best: 18.20/ 22.65 GFLOPS | Progress: (8/20) | 7.68 s
+[Task 11/25] Current/Best: 11.50/ 22.65 GFLOPS | Progress: (12/20) | 10.79 s
+[Task 11/25] Current/Best: 9.88/ 22.65 GFLOPS | Progress: (16/20) | 14.32 s
+[Task 11/25] Current/Best: 14.43/ 22.65 GFLOPS | Progress: (20/20) | 16.61 s Done.
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 12/25] Current/Best: 15.56/ 15.56 GFLOPS | Progress: (4/20) | 5.85 s
-[Task 12/25] Current/Best: 3.14/ 15.56 GFLOPS | Progress: (8/20) | 8.81 s
-[Task 12/25] Current/Best: 14.36/ 15.56 GFLOPS | Progress: (12/20) | 11.21 s
-[Task 12/25] Current/Best: 18.27/ 18.27 GFLOPS | Progress: (16/20) | 13.80 s
-[Task 12/25] Current/Best: 8.54/ 18.27 GFLOPS | Progress: (20/20) | 16.84 s Done.
+[Task 12/25] Current/Best: 9.84/ 13.82 GFLOPS | Progress: (4/20) | 6.97 s
+[Task 12/25] Current/Best: 16.00/ 18.49 GFLOPS | Progress: (8/20) | 9.26 s
+[Task 12/25] Current/Best: 13.35/ 18.49 GFLOPS | Progress: (12/20) | 11.73 s
+[Task 12/25] Current/Best: 9.70/ 19.01 GFLOPS | Progress: (16/20) | 15.57 s
+[Task 12/25] Current/Best: 15.77/ 19.01 GFLOPS | Progress: (20/20) | 17.75 s Done.
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 13/25] Current/Best: 10.22/ 10.22 GFLOPS | Progress: (4/20) | 6.84 s
-[Task 13/25] Current/Best: 6.03/ 17.34 GFLOPS | Progress: (8/20) | 10.41 s
-[Task 13/25] Current/Best: 13.53/ 19.00 GFLOPS | Progress: (12/20) | 12.87 s
-[Task 13/25] Current/Best: 3.11/ 19.00 GFLOPS | Progress: (16/20) | 16.89 s
-[Task 13/25] Current/Best: 16.11/ 19.00 GFLOPS | Progress: (20/20) | 20.64 s Done.
+[Task 13/25] Current/Best: 17.55/ 19.46 GFLOPS | Progress: (4/20) | 5.76 s
+[Task 13/25] Current/Best: 14.97/ 19.46 GFLOPS | Progress: (8/20) | 7.95 s
+[Task 13/25] Current/Best: 10.14/ 19.46 GFLOPS | Progress: (12/20) | 10.41 s
+[Task 13/25] Current/Best: 16.42/ 19.46 GFLOPS | Progress: (16/20) | 14.32 s
+[Task 13/25] Current/Best: 3.12/ 19.46 GFLOPS | Progress: (20/20) | 17.90 s Done.
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 14/25] Current/Best: 1.59/ 7.35 GFLOPS | Progress: (4/20) | 11.53 s
-[Task 14/25] Current/Best: 8.32/ 17.03 GFLOPS | Progress: (8/20) | 15.22 s
-[Task 14/25] Current/Best: 3.41/ 20.82 GFLOPS | Progress: (12/20) | 19.55 s
-[Task 14/25] Current/Best: 13.08/ 20.82 GFLOPS | Progress: (16/20) | 22.50 s
-[Task 14/25] Current/Best: 9.52/ 20.82 GFLOPS | Progress: (20/20) | 28.08 s Done.
-
+[Task 14/25] Current/Best: 16.10/ 16.10 GFLOPS | Progress: (4/20) | 5.02 s
+[Task 14/25] Current/Best: 7.35/ 16.10 GFLOPS | Progress: (8/20) | 8.90 s
+[Task 14/25] Current/Best: 14.16/ 16.10 GFLOPS | Progress: (12/20) | 11.76 s
+[Task 14/25] Current/Best: 5.42/ 16.10 GFLOPS | Progress: (16/20) | 14.35 s
+[Task 14/25] Current/Best: 9.46/ 16.10 GFLOPS | Progress: (20/20) | 19.48 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 15/25] Current/Best: 7.15/ 20.93 GFLOPS | Progress: (4/20) | 6.34 s
-[Task 15/25] Current/Best: 8.58/ 20.93 GFLOPS | Progress: (8/20) | 13.51 s
-[Task 15/25] Current/Best: 14.29/ 20.93 GFLOPS | Progress: (12/20) | 20.92 s
-[Task 15/25] Current/Best: 11.68/ 22.91 GFLOPS | Progress: (16/20) | 22.50 s
-[Task 15/25] Current/Best: 3.13/ 22.91 GFLOPS | Progress: (20/20) | 25.76 s
+[Task 15/25] Current/Best: 11.37/ 19.11 GFLOPS | Progress: (4/20) | 4.68 s
+[Task 15/25] Current/Best: 7.13/ 21.25 GFLOPS | Progress: (8/20) | 9.19 s
+[Task 15/25] Current/Best: 17.06/ 21.25 GFLOPS | Progress: (12/20) | 12.28 s Done.
+
+[Task 15/25] Current/Best: 9.16/ 21.25 GFLOPS | Progress: (16/20) | 18.47 s
+[Task 15/25] Current/Best: 18.32/ 21.25 GFLOPS | Progress: (20/20) | 22.25 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 16/25] Current/Best: 10.49/ 20.41 GFLOPS | Progress: (4/20) | 4.92 s
-[Task 16/25] Current/Best: 19.04/ 20.85 GFLOPS | Progress: (8/20) | 6.39 s
-[Task 16/25] Current/Best: 9.47/ 20.85 GFLOPS | Progress: (12/20) | 10.64 s
-[Task 16/25] Current/Best: 13.69/ 20.85 GFLOPS | Progress: (16/20) | 13.35 s
-[Task 16/25] Current/Best: 10.03/ 20.85 GFLOPS | Progress: (20/20) | 16.72 s Done.
+[Task 16/25] Current/Best: 15.25/ 21.15 GFLOPS | Progress: (4/20) | 4.70 s
+[Task 16/25] Current/Best: 16.37/ 21.15 GFLOPS | Progress: (8/20) | 8.12 s
+[Task 16/25] Current/Best: 9.39/ 21.15 GFLOPS | Progress: (12/20) | 9.76 s
+[Task 16/25] Current/Best: 15.13/ 21.15 GFLOPS | Progress: (16/20) | 11.67 s
+[Task 16/25] Current/Best: 13.02/ 21.15 GFLOPS | Progress: (20/20) | 13.19 s Done.
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 17/25] Current/Best: 18.15/ 18.15 GFLOPS | Progress: (4/20) | 5.64 s
-[Task 17/25] Current/Best: 9.65/ 20.76 GFLOPS | Progress: (8/20) | 8.15 s
-[Task 17/25] Current/Best: 11.67/ 20.76 GFLOPS | Progress: (12/20) | 10.75 s
-[Task 17/25] Current/Best: 5.13/ 20.76 GFLOPS | Progress: (16/20) | 13.72 s
-[Task 17/25] Current/Best: 20.70/ 21.09 GFLOPS | Progress: (20/20) | 16.44 s Done.
+[Task 17/25] Current/Best: 20.24/ 20.24 GFLOPS | Progress: (4/20) | 5.76 s
+[Task 17/25] Current/Best: 9.46/ 20.24 GFLOPS | Progress: (8/20) | 8.92 s
+[Task 17/25] Current/Best: 11.87/ 20.24 GFLOPS | Progress: (12/20) | 11.96 s
+[Task 17/25] Current/Best: 20.62/ 20.62 GFLOPS | Progress: (16/20) | 15.99 s
+[Task 17/25] Current/Best: 17.95/ 21.00 GFLOPS | Progress: (20/20) | 18.82 s Done.
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 18/25] Current/Best: 12.05/ 19.12 GFLOPS | Progress: (4/20) | 6.98 s
-[Task 18/25] Current/Best: 9.50/ 19.12 GFLOPS | Progress: (8/20) | 15.46 s
-[Task 18/25] Current/Best: 13.79/ 19.12 GFLOPS | Progress: (12/20) | 19.40 s
-[Task 18/25] Current/Best: 1.57/ 19.12 GFLOPS | Progress: (16/20) | 23.62 s
-[Task 18/25] Current/Best: 4.15/ 19.12 GFLOPS | Progress: (20/20) | 28.60 s Done.
+[Task 18/25] Current/Best: 2.88/ 18.16 GFLOPS | Progress: (4/20) | 5.46 s
+[Task 18/25] Current/Best: 13.46/ 18.16 GFLOPS | Progress: (8/20) | 8.59 s
+[Task 18/25] Current/Best: 15.13/ 18.16 GFLOPS | Progress: (12/20) | 12.61 s
+[Task 18/25] Current/Best: 10.27/ 18.16 GFLOPS | Progress: (16/20) | 16.18 s
+[Task 18/25] Current/Best: 9.13/ 20.35 GFLOPS | Progress: (20/20) | 22.01 s Done.
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 19/25] Current/Best: 2.69/ 20.80 GFLOPS | Progress: (4/20) | 5.83 s
-[Task 19/25] Current/Best: 9.28/ 20.80 GFLOPS | Progress: (8/20) | 8.67 s
-[Task 19/25] Current/Best: 10.39/ 20.80 GFLOPS | Progress: (12/20) | 13.58 s
-[Task 19/25] Current/Best: 15.65/ 20.80 GFLOPS | Progress: (16/20) | 17.39 s
-[Task 19/25] Current/Best: 9.44/ 20.80 GFLOPS | Progress: (20/20) | 20.29 s Done.
+[Task 19/25] Current/Best: 10.37/ 18.57 GFLOPS | Progress: (4/20) | 5.56 s
+[Task 19/25] Current/Best: 4.44/ 18.57 GFLOPS | Progress: (8/20) | 13.41 s
+[Task 19/25] Current/Best: 13.86/ 21.45 GFLOPS | Progress: (12/20) | 16.75 s
+[Task 19/25] Current/Best: 10.81/ 21.45 GFLOPS | Progress: (16/20) | 21.96 s
+[Task 19/25] Current/Best: 5.05/ 21.45 GFLOPS | Progress: (20/20) | 25.51 s Done.
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 20/25] Current/Best: 8.12/ 15.41 GFLOPS | Progress: (4/20) | 5.33 s
-[Task 20/25] Current/Best: 5.97/ 17.38 GFLOPS | Progress: (8/20) | 7.66 s
-[Task 20/25] Current/Best: 7.18/ 18.13 GFLOPS | Progress: (12/20) | 10.37 s
-[Task 20/25] Current/Best: 9.94/ 18.13 GFLOPS | Progress: (16/20) | 14.70 s
-[Task 20/25] Current/Best: 12.07/ 18.13 GFLOPS | Progress: (20/20) | 17.05 s
+[Task 20/25] Current/Best: 10.71/ 13.76 GFLOPS | Progress: (4/20) | 7.05 s
+[Task 20/25] Current/Best: 17.40/ 17.40 GFLOPS | Progress: (8/20) | 10.36 s
+[Task 20/25] Current/Best: 17.71/ 17.71 GFLOPS | Progress: (12/20) | 12.23 s
+[Task 20/25] Current/Best: 17.14/ 17.71 GFLOPS | Progress: (16/20) | 15.73 s
+[Task 20/25] Current/Best: 10.42/ 17.71 GFLOPS | Progress: (20/20) | 19.15 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 21/25] Current/Best: 6.95/ 16.08 GFLOPS | Progress: (4/20) | 5.68 s
-[Task 21/25] Current/Best: 19.38/ 20.64 GFLOPS | Progress: (8/20) | 7.19 s Done.
- Done.
+[Task 21/25] Current/Best: 8.69/ 8.69 GFLOPS | Progress: (4/20) | 5.30 s
+[Task 21/25] Current/Best: 16.47/ 18.88 GFLOPS | Progress: (8/20) | 11.76 s
+[Task 21/25] Current/Best: 18.74/ 18.88 GFLOPS | Progress: (12/20) | 13.76 s Done.
-[Task 21/25] Current/Best: 7.62/ 20.64 GFLOPS | Progress: (12/20) | 11.45 s
-[Task 21/25] Current/Best: 17.38/ 20.64 GFLOPS | Progress: (16/20) | 14.52 s
-[Task 21/25] Current/Best: 5.41/ 20.64 GFLOPS | Progress: (20/20) | 18.27 s
+[Task 21/25] Current/Best: 10.41/ 18.88 GFLOPS | Progress: (16/20) | 17.37 s
+[Task 21/25] Current/Best: 7.43/ 18.88 GFLOPS | Progress: (20/20) | 22.27 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 22/25] Current/Best: 12.20/ 16.83 GFLOPS | Progress: (4/20) | 4.49 s
-[Task 22/25] Current/Best: 11.76/ 16.83 GFLOPS | Progress: (8/20) | 7.25 s
-[Task 22/25] Current/Best: 11.72/ 18.57 GFLOPS | Progress: (12/20) | 9.24 s
-[Task 22/25] Current/Best: 5.27/ 20.78 GFLOPS | Progress: (16/20) | 12.56 s
-[Task 22/25] Current/Best: 14.03/ 20.78 GFLOPS | Progress: (20/20) | 15.69 s Done.
+[Task 22/25] Current/Best: 8.91/ 19.30 GFLOPS | Progress: (4/20) | 5.71 s
+[Task 22/25] Current/Best: 17.81/ 21.90 GFLOPS | Progress: (8/20) | 7.49 s
+[Task 22/25] Current/Best: 5.36/ 21.90 GFLOPS | Progress: (12/20) | 9.44 s
+[Task 22/25] Current/Best: 20.70/ 21.90 GFLOPS | Progress: (16/20) | 11.45 s
+[Task 22/25] Current/Best: 7.67/ 21.90 GFLOPS | Progress: (20/20) | 13.90 s Done.
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 23/25] Current/Best: 20.10/ 20.10 GFLOPS | Progress: (4/20) | 4.97 s
-[Task 23/25] Current/Best: 23.26/ 23.26 GFLOPS | Progress: (8/20) | 7.87 s
-[Task 23/25] Current/Best: 1.55/ 23.26 GFLOPS | Progress: (12/20) | 12.65 s
-[Task 23/25] Current/Best: 22.14/ 23.26 GFLOPS | Progress: (16/20) | 15.27 s
-[Task 23/25] Current/Best: 19.29/ 23.26 GFLOPS | Progress: (20/20) | 17.50 s Done.
+[Task 23/25] Current/Best: 3.07/ 17.93 GFLOPS | Progress: (4/20) | 6.79 s
+[Task 23/25] Current/Best: 13.08/ 22.58 GFLOPS | Progress: (8/20) | 9.80 s
+[Task 23/25] Current/Best: 20.02/ 22.58 GFLOPS | Progress: (12/20) | 12.54 s
+[Task 23/25] Current/Best: 10.10/ 22.58 GFLOPS | Progress: (16/20) | 16.51 s
+[Task 23/25] Current/Best: 19.74/ 24.05 GFLOPS | Progress: (20/20) | 19.18 s Done.
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 24/25] Current/Best: 4.90/ 8.19 GFLOPS | Progress: (4/20) | 13.74 s
-[Task 24/25] Current/Best: 1.08/ 8.19 GFLOPS | Progress: (8/20) | 25.89 s
-[Task 24/25] Current/Best: 3.87/ 8.66 GFLOPS | Progress: (12/20) | 36.83 s
-[Task 24/25] Current/Best: 1.50/ 8.66 GFLOPS | Progress: (16/20) | 40.11 s
-[Task 24/25] Current/Best: 7.03/ 9.16 GFLOPS | Progress: (20/20) | 50.75 s
+[Task 24/25] Current/Best: 2.08/ 6.95 GFLOPS | Progress: (4/20) | 13.80 s
+[Task 24/25] Current/Best: 8.09/ 8.09 GFLOPS | Progress: (8/20) | 26.19 s
+[Task 24/25] Current/Best: 4.22/ 8.09 GFLOPS | Progress: (12/20) | 38.83 s
+[Task 24/25] Current/Best: 7.90/ 8.09 GFLOPS | Progress: (16/20) | 49.50 s
+[Task 24/25] Current/Best: 10.34/ 10.34 GFLOPS | Progress: (20/20) | 62.50 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+ Done.
+ Done.
-[Task 25/25] Current/Best: 6.38/ 6.38 GFLOPS | Progress: (4/20) | 15.14 s
-[Task 25/25] Current/Best: 1.55/ 9.67 GFLOPS | Progress: (8/20) | 26.10 s
-[Task 25/25] Current/Best: 1.55/ 9.67 GFLOPS | Progress: (12/20) | 28.82 s
-[Task 25/25] Current/Best: 8.46/ 9.67 GFLOPS | Progress: (16/20) | 30.27 s
-[Task 25/25] Current/Best: 9.90/ 9.90 GFLOPS | Progress: (20/20) | 41.23 s
+[Task 25/25] Current/Best: 5.12/ 7.36 GFLOPS | Progress: (4/20) | 5.59 s
+[Task 25/25] Current/Best: 5.83/ 8.88 GFLOPS | Progress: (8/20) | 10.55 s
+[Task 25/25] Current/Best: 9.24/ 9.24 GFLOPS | Progress: (12/20) | 21.24 s
+[Task 25/25] Current/Best: 3.03/ 9.24 GFLOPS | Progress: (16/20) | 32.22 s
+[Task 25/25] Current/Best: 7.77/ 9.24 GFLOPS | Progress: (20/20) | 43.18 s
</pre></div>
</div>
<p>The output from this tuning process will look something like this:</p>
@@ -948,7 +949,7 @@ model using optimized operators to speed up our computations.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>class='n02123045 tabby, tabby cat' with probability=0.621104
-class='n02123159 tiger cat' with probability=0.356379
+class='n02123159 tiger cat' with probability=0.356378
class='n02124075 Egyptian cat' with probability=0.019712
class='n02129604 tiger, Panthera tigris' with probability=0.001215
class='n04040759 radiator' with probability=0.000262
@@ -985,8 +986,8 @@ improvement in comparing the optimized model to the unoptimized model.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"unoptimized: </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">unoptimized</span></a><span class="p">))</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {'mean': 411.58318418000135, 'median': 410.62697249999474, 'std': 3.2463555895772456}
-unoptimized: {'mean': 511.0277086200017, 'median': 510.5724576499995, 'std': 1.9596086703937687}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {'mean': 426.34954656999753, 'median': 425.6728701499924, 'std': 3.6461599603240114}
+unoptimized: {'mean': 550.6390348200011, 'median': 546.4105448499993, 'std': 15.583710586461082}
</pre></div>
</div>
</div>
@@ -1000,7 +1001,7 @@ models.</p>
<p>Here we presented a simple example using ResNet-50 v2 locally. However, TVM
supports many more features including cross-compilation, remote execution and
profiling/benchmarking.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 12 minutes 34.020 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 13 minutes 15.632 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-autotvm-relay-x86-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../_downloads/57a45d9bef1af358191e7d50043e652c/autotvm_relay_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">autotvm_relay_x86.py</span></code></a></p>
diff --git a/docs/tutorial/cross_compilation_and_rpc.html b/docs/tutorial/cross_compilation_and_rpc.html
index e58bf74270..6140d16b74 100644
--- a/docs/tutorial/cross_compilation_and_rpc.html
+++ b/docs/tutorial/cross_compilation_and_rpc.html
@@ -543,7 +543,7 @@ device and returns the measured cost. Network overhead is excluded.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"</span><span class="si">%g</span><span class="s2"> secs/op"</span> <span class="o">%</span> <span class="n">cost</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>1.344e-07 secs/op
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>1.219e-07 secs/op
</pre></div>
</div>
</div>
diff --git a/docs/tutorial/intro_topi.html b/docs/tutorial/intro_topi.html
index 05fb839da8..f7167611ba 100644
--- a/docs/tutorial/intro_topi.html
+++ b/docs/tutorial/intro_topi.html
@@ -513,7 +513,7 @@ class Module:
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/ir.html#tvm.ir.Array" title="tvm.ir.Array" class="sphx-glr-backref-module-tvm-ir sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sg</span><span class="o">.</span><span class="n">stages</span></a><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[stage(a, placeholder(a, 0x250a2f60)), stage(b, placeholder(b, 0xf1445b0)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs [...]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[stage(a, placeholder(a, 0x2379ba40)), stage(b, placeholder(b, 0xd6b9d50)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs [...]
</pre></div>
</div>
<p>We can test the correctness by comparing with <code class="code docutils literal notranslate"><span class="pre">numpy</span></code> result as follows</p>
diff --git a/docs/tutorial/sg_execution_times.html b/docs/tutorial/sg_execution_times.html
index dba9beb2b2..09bb8391d8 100644
--- a/docs/tutorial/sg_execution_times.html
+++ b/docs/tutorial/sg_execution_times.html
@@ -345,7 +345,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-tutorial-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>16:19.758</strong> total execution time for <strong>tutorial</strong> files:</p>
+<p><strong>16:41.930</strong> total execution time for <strong>tutorial</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -354,35 +354,35 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="autotvm_relay_x86.html#sphx-glr-tutorial-autotvm-relay-x86-py"><span class="std std-ref">Compiling and Optimizing a Model with the Python Interface (AutoTVM)</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_relay_x86.py</span></code>)</p></td>
-<td><p>12:34.020</p></td>
+<td><p>13:15.632</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="auto_scheduler_matmul_x86.html#sphx-glr-tutorial-auto-scheduler-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Auto-scheduling</span></a> (<code class="docutils literal notranslate"><span class="pre">auto_scheduler_matmul_x86.py</span></code>)</p></td>
-<td><p>01:30.840</p></td>
+<td><p>01:22.267</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tensor_expr_get_started.html#sphx-glr-tutorial-tensor-expr-get-started-py"><span class="std std-ref">Working with Operators Using Tensor Expression</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_expr_get_started.py</span></code>)</p></td>
-<td><p>01:00.705</p></td>
+<td><p>00:58.402</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="relay_quick_start.html#sphx-glr-tutorial-relay-quick-start-py"><span class="std std-ref">Quick Start Tutorial for Compiling Deep Learning Models</span></a> (<code class="docutils literal notranslate"><span class="pre">relay_quick_start.py</span></code>)</p></td>
-<td><p>00:36.382</p></td>
+<td><p>00:37.142</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="autotvm_matmul_x86.html#sphx-glr-tutorial-autotvm-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Schedule Templates and AutoTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_matmul_x86.py</span></code>)</p></td>
-<td><p>00:35.404</p></td>
+<td><p>00:25.867</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tensor_ir_blitz_course.html#sphx-glr-tutorial-tensor-ir-blitz-course-py"><span class="std std-ref">Blitz Course to TensorIR</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_ir_blitz_course.py</span></code>)</p></td>
-<td><p>00:01.385</p></td>
+<td><p>00:01.563</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="intro_topi.html#sphx-glr-tutorial-intro-topi-py"><span class="std std-ref">Introduction to TOPI</span></a> (<code class="docutils literal notranslate"><span class="pre">intro_topi.py</span></code>)</p></td>
-<td><p>00:00.852</p></td>
+<td><p>00:00.864</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="cross_compilation_and_rpc.html#sphx-glr-tutorial-cross-compilation-and-rpc-py"><span class="std std-ref">Cross Compilation and RPC</span></a> (<code class="docutils literal notranslate"><span class="pre">cross_compilation_and_rpc.py</span></code>)</p></td>
-<td><p>00:00.170</p></td>
+<td><p>00:00.194</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="uma.html#sphx-glr-tutorial-uma-py"><span class="std std-ref">Making your Hardware Accelerator TVM-ready with UMA</span></a> (<code class="docutils literal notranslate"><span class="pre">uma.py</span></code>)</p></td>
diff --git a/docs/tutorial/tensor_expr_get_started.html b/docs/tutorial/tensor_expr_get_started.html
index 229d967ea9..1fe05740bc 100644
--- a/docs/tutorial/tensor_expr_get_started.html
+++ b/docs/tutorial/tensor_expr_get_started.html
@@ -610,7 +610,7 @@ compile and run this new schedule with the parallel operation applied:</p>
<span class="n">evaluate_addition</span><span class="p">(</span><span class="n">fadd_parallel</span><span class="p">,</span> <a href="../reference/api/python/target.html#tvm.target.Target" title="tvm.target.Target" class="sphx-glr-backref-module-tvm-target sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">tgt</span></a><span class="p">,</span> <span class="s2">"parallel"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.h [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallel: 0.000010
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallel: 0.000008
</pre></div>
</div>
</div>
@@ -686,10 +686,10 @@ class Module:
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Operator Timing Performance
- numpy 6.761279998954706e-06 1.0
- naive 6.642e-06 0.9823583701646513
-parallel 1.0234400000000001e-05 1.5136778837116995
- vector 2.45438e-05 3.630052298350974
+ numpy 7.415540001147746e-06 1.0
+ naive 6.6986e-06 0.9033192456602241
+parallel 8.3874e-06 1.1310572121115703
+ vector 2.4674099999999998e-05 3.3273504014786575
</pre></div>
</div>
<div class="admonition-code-specialization admonition">
@@ -1005,7 +1005,7 @@ matrix multiplication.</p>
<span class="n">answer</span> <span class="o">=</span> <span class="n">numpy</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">a</span><span class="o">.</span><span class="n">numpy</span><span class="p">(),</span> <span class="n">b</span><span class="o">.</span><span class="n">numpy</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: 0.018208
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018502
</pre></div>
</div>
<p>Now we write a basic matrix multiplication using TVM TE and verify that it
@@ -1046,7 +1046,7 @@ optimizations.</p>
<span class="n">evaluate_operation</span><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">s</span></a><span class="p">,</span> <span class="p">[</span><a href="../reference/api/python/te.html#tvm.te.Tensor" title="tvm.te.Tensor" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>none: 3.384123
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>none: 3.183745
</pre></div>
</div>
<p>Let’s take a look at the intermediate representation of the operator and
@@ -1110,7 +1110,7 @@ schedule.</p>
<span class="n">evaluate_operation</span><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">s</span></a><span class="p">,</span> <span class="p">[</span><a href="../reference/api/python/te.html#tvm.te.Tensor" title="tvm.te.Tensor" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>blocking: 0.305746
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>blocking: 0.300639
</pre></div>
</div>
<p>By reordering the computation to take advantage of caching, you should see a
@@ -1159,7 +1159,7 @@ already cache friendly from our previous optimizations.</p>
<span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vectorization: 0.340842
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vectorization: 0.338995
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1208,7 +1208,7 @@ more cache friendly.</p>
<span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loop permutation: 0.113210
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loop permutation: 0.120766
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1278,7 +1278,7 @@ optimized schedule.</p>
<span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>array packing: 0.107266
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>array packing: 0.108812
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1344,7 +1344,7 @@ to `C</cite> when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>block caching: 0.110337
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>block caching: 0.111143
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1401,7 +1401,7 @@ of thread-level parallelization.</p>
<span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallelization: 0.145685
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallelization: 0.146442
# from tvm.script import ir as I
# from tvm.script import tir as T
@@ -1454,13 +1454,13 @@ working, we can compare the results.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span> Operator Timing Performance
- none 3.3841233051 1.0
- blocking 0.30574600900000004 0.09034718343129797
- vectorization 0.3408423891 0.1007180762551819
-loop permutation 0.1132099054 0.03345324481214631
- array packing 0.10726575680000001 0.031696763719674906
- block caching 0.1103372088 0.0326043701285109
- parallelization 0.1456849141 0.04304952892243832
+ none 3.1837453195 1.0
+ blocking 0.3006391989 0.09442941213250523
+ vectorization 0.3389950156 0.10647680061709158
+loop permutation 0.1207660036 0.03793205532500509
+ array packing 0.10881237490000002 0.03417747463452534
+ block caching 0.1111432627 0.034909595946404026
+ parallelization 0.14644234039999998 0.045996876541305264
</pre></div>
</div>
<p>Note that the outputs on the web page reflect the running times on a
@@ -1492,7 +1492,6 @@ is</p>
you can build generic templates of the matrix multiplication and other
operations with tunable parameters that allows you to automatically optimize
the computation for specific platforms.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 0.705 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-tensor-expr-get-started-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../_downloads/40a01cffb015a67aaec0fad7e27cf80d/tensor_expr_get_started.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">tensor_expr_get_started.py</span></code></a></p>