You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2022/10/17 07:04:15 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@cbca28da003cf4791a19f0c4f9e0c930ecee522d)
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 ae55588d45 deploying docs (apache/tvm@cbca28da003cf4791a19f0c4f9e0c930ecee522d)
ae55588d45 is described below
commit ae55588d454cc013ceabc7b1fef0c819a177dccd
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Mon Oct 17 07:04:08 2022 +0000
deploying docs (apache/tvm@cbca28da003cf4791a19f0c4f9e0c930ecee522d)
---
docs/_images/sphx_glr_micro_train_001.png | Bin 327199 -> 322550 bytes
docs/_images/sphx_glr_micro_train_thumb.png | Bin 22934 -> 23209 bytes
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_keras.rst.txt | 2 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.rst.txt | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 2 +-
.../compile_models/sg_execution_times.rst.txt | 22 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 18 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 2 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 10 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 14 +-
.../tune_conv2d_layer_cuda.rst.txt | 1460 ++++++++------------
.../tune_network_cuda.rst.txt | 4 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 1055 +++++++++++++-
.../tune_with_autotvm/sg_execution_times.rst.txt | 6 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 190 +--
.../work_with_microtvm/micro_autotune.rst.txt | 16 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 18 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 10 +-
.../work_with_relay/sg_execution_times.rst.txt | 8 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 14 +-
.../how_to/work_with_schedules/tensorize.rst.txt | 2 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 4 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 4 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 58 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 24 +-
.../tutorial/tensor_expr_get_started.rst.txt | 42 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_darknet.html | 2 +-
docs/how_to/compile_models/from_keras.html | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 13 +-
docs/how_to/compile_models/from_pytorch.html | 10 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 22 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 41 +-
docs/how_to/deploy_models/deploy_prequantized.html | 8 +-
.../deploy_models/deploy_prequantized_tflite.html | 4 +-
docs/how_to/deploy_models/deploy_quantized.html | 2 +-
docs/how_to/deploy_models/deploy_ssd_gluoncv.html | 35 +-
docs/how_to/deploy_models/sg_execution_times.html | 18 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
docs/how_to/extend_tvm/sg_execution_times.html | 10 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 1460 ++++++++------------
.../tune_with_autoscheduler/tune_network_cuda.html | 4 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 1055 +++++++++++++-
.../tune_with_autotvm/sg_execution_times.html | 6 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 190 +--
docs/how_to/work_with_microtvm/micro_autotune.html | 16 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 10 +-
.../how_to/work_with_relay/sg_execution_times.html | 8 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 14 +-
docs/how_to/work_with_schedules/tensorize.html | 2 +-
docs/install/nnpack.html | 12 +-
...stvm_1_1meta__schedule_1_1Database-members.html | 8 +-
.../classtvm_1_1meta__schedule_1_1Database.html | 80 +-
..._1_1meta__schedule_1_1DatabaseNode-members.html | 12 +-
...classtvm_1_1meta__schedule_1_1DatabaseNode.html | 86 +-
...meta__schedule_1_1DatabaseNode__coll__graph.svg | 120 +-
...a__schedule_1_1DatabaseNode__inherit__graph.svg | 157 +--
..._1meta__schedule_1_1PyDatabaseNode-members.html | 81 +-
...asstvm_1_1meta__schedule_1_1PyDatabaseNode.html | 57 +-
...ta__schedule_1_1PyDatabaseNode__coll__graph.svg | 339 ++---
..._schedule_1_1PyDatabaseNode__inherit__graph.svg | 159 +--
docs/reference/api/doxygen/database_8h.html | 3 +-
docs/reference/api/doxygen/database_8h__incl.svg | 1050 +++++++-------
docs/reference/api/doxygen/database_8h_source.html | 84 +-
docs/reference/api/doxygen/functions_0x7e.html | 2 +-
docs/reference/api/doxygen/functions_d.html | 5 +-
.../reference/api/doxygen/functions_func_0x7e.html | 2 +-
docs/reference/api/doxygen/functions_func_d.html | 9 +-
docs/reference/api/doxygen/functions_func_g.html | 11 +-
docs/reference/api/doxygen/functions_func_j.html | 2 +-
docs/reference/api/doxygen/functions_func_m.html | 2 +-
docs/reference/api/doxygen/functions_func_p.html | 5 +-
docs/reference/api/doxygen/functions_func_s.html | 4 +-
docs/reference/api/doxygen/functions_func_w.html | 3 +
docs/reference/api/doxygen/functions_g.html | 5 +-
docs/reference/api/doxygen/functions_j.html | 2 +-
docs/reference/api/doxygen/functions_m.html | 4 +-
docs/reference/api/doxygen/functions_p.html | 7 +-
docs/reference/api/doxygen/functions_s.html | 6 +-
docs/reference/api/doxygen/functions_t.html | 12 +-
docs/reference/api/doxygen/functions_u.html | 2 +-
docs/reference/api/doxygen/functions_v.html | 12 +-
docs/reference/api/doxygen/functions_w.html | 3 +
.../api/doxygen/measure__callback_8h.html | 2 +-
.../api/doxygen/measure__callback_8h__incl.svg | 1244 ++++++++---------
docs/reference/api/doxygen/search/all_10.js | 2 +-
docs/reference/api/doxygen/search/all_11.js | 6 +-
docs/reference/api/doxygen/search/all_13.js | 4 +-
docs/reference/api/doxygen/search/all_14.js | 12 +-
docs/reference/api/doxygen/search/all_15.js | 6 +-
docs/reference/api/doxygen/search/all_16.js | 4 +-
docs/reference/api/doxygen/search/all_17.js | 2 +-
docs/reference/api/doxygen/search/all_18.js | 4 +-
docs/reference/api/doxygen/search/all_1c.js | 2 +-
docs/reference/api/doxygen/search/all_5.js | 2 +-
docs/reference/api/doxygen/search/all_8.js | 1 +
docs/reference/api/doxygen/search/all_b.js | 2 +-
docs/reference/api/doxygen/search/all_e.js | 8 +-
docs/reference/api/doxygen/search/functions_10.js | 3 +-
docs/reference/api/doxygen/search/functions_12.js | 2 +-
docs/reference/api/doxygen/search/functions_13.js | 8 +-
docs/reference/api/doxygen/search/functions_15.js | 2 +-
docs/reference/api/doxygen/search/functions_17.js | 1 +
docs/reference/api/doxygen/search/functions_1a.js | 2 +-
docs/reference/api/doxygen/search/functions_4.js | 1 +
docs/reference/api/doxygen/search/functions_7.js | 1 +
docs/reference/api/doxygen/search/functions_a.js | 2 +-
docs/reference/api/doxygen/search/functions_d.js | 6 +-
.../reference/api/doxygen/search__strategy_8h.html | 2 +-
.../api/doxygen/search__strategy_8h__incl.svg | 1164 ++++++++--------
...1_1meta__schedule_1_1WorkloadEqual-members.html | 3 +-
...ructtvm_1_1meta__schedule_1_1WorkloadEqual.html | 39 +-
...eta__schedule_1_1WorkloadEqual__coll__graph.svg | 21 +-
docs/reference/api/doxygen/tune__context_8h.html | 2 +-
.../api/doxygen/tune__context_8h__incl.svg | 1338 +++++++++---------
docs/reference/api/python/auto_scheduler.html | 4 +-
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
.../api/typedoc/classes/graphexecutor.html | 16 +-
docs/reference/api/typedoc/classes/instance.html | 40 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 112 +-
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 4 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 6 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 4 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 272 ++--
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 28 +-
docs/tutorial/tensor_expr_get_started.html | 42 +-
190 files changed, 7323 insertions(+), 5912 deletions(-)
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 9acba7fd3b..b961179557 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 fb0f49ab60..ab1b531c46 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 48ba8d7d22..4653d5a763 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -315,7 +315,7 @@ The process is no different from other examples.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 12.004 seconds)
+ **Total running time of the script:** ( 1 minutes 12.318 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 cdd123cd74..441a24774d 100644
--- a/docs/_sources/how_to/compile_models/from_keras.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_keras.rst.txt
@@ -228,7 +228,7 @@ Look up prediction top 1 index in 1000 class synset.
.. code-block:: none
Relay top-1 id: 285, class name: Egyptian cat
-
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 992ms/step
+
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 981ms/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 395f47a9b2..84f9cf2099 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -115,7 +115,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip3988ca5b-e686-486a-b79f-2c757f8bb17e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipadd651b9-c7d9-4b22-a086-3c17190aed2b 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 fc4464dd3b..1cd049a3ae 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -116,7 +116,7 @@ Load a pretrained OneFlow model and save model
.. code-block:: none
Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 48.7MB/s]
39%|###8 | 16.0M/41.5M [00:00<00:00, 56.1MB/s]
57%|#####7 | 23.7M/41.5M [00:00<00:00, 64.7MB/s]
73%|#######2 | 30.1M/41.5M [00:00<00:00, 56.6MB/s]
86%|########6 | 35.7M/41.5M [00:00<00:00, 54.8MB/s]
99%|#########9| 41.1M/41.5M [00:00<00:00, 52.3MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 55.0MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
19%|#9 | 7.99M/41.5M [00:00<00:00, 79.5MB/s]
38%|###7 | 15.6M/41.5M [00:00<00:00, 78.6MB/s]
56%|#####5 | 23.1M/41.5M [00:00<00:00, 65.0MB/s]
71%|#######1 | 29.5M/41.5M [00:00<00:00, 65.5MB/s]
86%|########6 | 35.9M/41.5M [00:00<00:00, 61.4MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 61.9MB/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 a885df2832..43047dd120 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -94,7 +94,7 @@ Load a pretrained PyTorch model
.. code-block:: none
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
0%| | 0.00/44.7M [00:00<?, ?B/s]
23%|##2 | 10.1M/44.7M [00:00<00:00, 101MB/s]
54%|#####3 | 24.0M/44.7M [00:00<00:00, 108MB/s]
77%|#######6 | 34.2M/44.7M [00:00<00:00, 87.5MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 100MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
18%|#7 | 7.99M/44.7M [00:00<00:00, 63.5MB/s]
36%|###5 | 16.0M/44.7M [00:00<00:00, 69.2MB/s]
54%|#####3 | 24.0M/44.7M [00:00<00:00, 72.5MB/s]
72%|#######1 | 32.0M/44.7M [00:00<00:00, 75.1MB/s]
90%|########9 | 40.0M/44.7M [00:00<00:00, 75.2MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 79.0MB/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 c05e69a25c..356e91123e 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -416,7 +416,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 15.522 seconds)
+ **Total running time of the script:** ( 1 minutes 14.190 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 dbdfb202bc..42eb8cb3d1 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
Computation times
=================
-**05:53.767** total execution time for **how_to_compile_models** files:
+**05:50.252** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:15.522 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:14.190 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:12.004 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:12.318 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:47.158 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:47.247 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:33.509 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:32.367 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:30.271 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:29.760 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:26.577 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:26.458 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:25.032 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:25.446 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:22.601 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:22.051 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:18.727 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:17.994 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.365 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.420 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index ba171ea5ae..1692229893 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -434,7 +434,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.0638 16.0509 16.1987 15.8900 0.1052
+ 16.4224 16.2221 17.1125 16.1277 0.3585
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 5c16e73504..471c324766 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
@@ -123,7 +123,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
.. code-block:: none
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 7.99M/170M [00:00<00:02, 67.1MB/s]
14%|#4 | 24.0M/170M [00:00<00:01, 101MB/s]
22%|##2 | 38.0M/170M [00:00<00:01, 119MB/s]
29%|##9 | 49.5M/170M [00:00<00:01, 95.1MB/s]
35%|###4 | 59.2M/170M [00:00<00:01, 92.8MB/s]
40%|#### | 68.3M/170M [00:00<00:01, 81.2MB/s]
47%|####7 | 80.0M/170M [00:00<00:01, 83.4MB/s]
52%|#####1 | 88.2M/170M [00:01<00:01, 73.1MB/s]
57%|#####6 | 96.0M/170M [00:01<00:01, 73.7MB/s]
62%|######1 | 105M/170M [00:01<00:00, 79.1MB/s]
66%|######6 | 113M/170M [00:01<00:00, 62.7MB/s]
75%|#######5 | 128M/170M [00:01<00:00, 77.4MB/s]
80%|######## | 136M/170M [00:01<00:00, 76.1MB/s]
86%|########5 | 145M/170M [00:01<00:00, 81.2MB/s]
90%|######### | 153M/170M [00:02<00:00, 76.3MB/s]
99%|#########8| 168M/170M [00:02<00:00, 91.7MB/s]
100%|##########| 170M/170M [00:02<00:00, 83.7MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
5%|4 | 7.99M/170M [00:00<00:02, 82.7MB/s]
9%|9 | 15.9M/170M [00:00<00:02, 78.2MB/s]
14%|#4 | 24.0M/170M [00:00<00:02, 55.7MB/s]
18%|#7 | 30.3M/170M [00:00<00:02, 57.4MB/s]
21%|##1 | 36.2M/170M [00:00<00:02, 57.9MB/s]
27%|##7 | 45.9M/170M [00:00<00:01, 70.9MB/s]
33%|###2 | 56.0M/170M [00:00<00:01, 72.3MB/s]
38%|###7 | 64.0M/170M [00:00<00:01, 69.9MB/s]
42%|####2 | 72.0M/170M [00:01<00:01, 73.2MB/s]
47%|####7 | 80.0M/170M [00:01<00:01, 68.2MB/s]
52%|#####1 | 88.0M/170M [00:01<00:01, 59.3MB/s]
57%|#####6 | 96.0M/170M [00:01<00:01, 64.8MB/s]
61%|######1 | 104M/170M [00:01<00:01, 61.2MB/s]
66%|######5 | 112M/170M [00:01<00:00, 65.6MB/s]
71%|####### | 120M/170M [00:01<00:00, 68.9MB/s]
75%|#######5 | 128M/170M [00:02<00:00, 68.9MB/s]
80%|######## | 136M/170M [00:02<00:00, 66.6MB/s]
85%|########4 | 144M/170M [00:02<00:00, 68.4MB/s]
89%|########9 | 152M/170M [00:02<00:00, 66.3MB/s]
94%|#########4| 160M/170M [00:02<00:00, 71.2MB/s]
99%|#########8| 168M/170M [00:02<00:00, 74.1MB/s]
100%|##########| 170M/170M [00:02<00:00, 68.1MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3878: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -292,7 +292,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 16.844 seconds)
+ **Total running time of the script:** ( 3 minutes 20.980 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 47e7872610..970702691a 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -232,7 +232,7 @@ training. Other models require a full post training calibration.
.. code-block:: none
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
0%| | 0.00/13.6M [00:00<?, ?B/s]
73%|#######3 | 9.91M/13.6M [00:00<00:00, 104MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 98.2MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
59%|#####8 | 7.99M/13.6M [00:00<00:00, 64.8MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 93.3MB/s]
@@ -414,7 +414,7 @@ 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)
- 90.5351 90.3460 99.2452 90.2206 1.0205
+ 90.7599 90.4873 99.4567 90.3017 1.1233
@@ -463,7 +463,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 20.750 seconds)
+ **Total running time of the script:** ( 1 minutes 21.970 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 02aa0d1945..995ea37fa0 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -432,7 +432,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 120.2857 120.2028 122.1906 119.6793 0.3909
+ 119.4545 119.4453 121.5507 118.7648 0.3889
@@ -469,7 +469,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 28.020 seconds)
+ **Total running time of the script:** ( 2 minutes 22.289 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 20e0bc4f36..cb02c96c9e 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -253,7 +253,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 21.499 seconds)
+ **Total running time of the script:** ( 1 minutes 22.216 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 c2cafa7d7a..62beaada12 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -166,7 +166,7 @@ Convert and compile model for CPU.
data: None
input_sym_arg_type = in_param.infer_type()[0]
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
0%| | 0/132723 [00:00<?, ?KB/s]
4%|3 | 4678/132723 [00:00<00:02, 46742.60KB/s]
10%|9 | 12780/132723 [00:00<00:01, 66895.52KB/s]
16%|#6 | 21353/132723 [00:00<00:01, 75488.80KB/s]
23%|##2 | 29913/132723 [00:00<00:01, 79477.56KB/s]
29%|##8 | 38389/132723 [00:00<00:01, 81379.36KB/s]
35%|###5 | 46945/132723 [00:00<00:01, 82796.70KB/s]
42%|####1 | 55566/132723 [00:00<00:00, 83910.41KB/s]
48%|####8 | 64167/132723 [00:00<00:00, 84576.67KB/s]
55%|#####4 | 72790/132723 [00:00<00:00, 85092.30KB/s]
61%|######1 | 81400/132723 [00:01<00:00, 85402.15KB/s]
68%|######7 | 90003/132723 [00:01<00:00, 85591.63KB/s]
74%|#######4 | 98697/132723 [00:01<00:00, 85998.64KB/s]
81%|######## | 107396/132723 [00:01<00:00, 86293.42KB/s]
87%|########7 | 116070/132723 [00:01<00:00, 86425.89KB/s]
94%|#########3| 124713/132723 [00:01<00:00, 86394.87KB/s]
100%|########
##| 132723/132723 [00:01<00:00, 83248.74KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
5%|4 | 6479/132723 [00:00<00:01, 64759.34KB/s]
11%|# | 14287/132723 [00:00<00:01, 72585.70KB/s]
17%|#6 | 22207/132723 [00:00<00:01, 75587.03KB/s]
23%|##2 | 30169/132723 [00:00<00:01, 77175.50KB/s]
29%|##8 | 38087/132723 [00:00<00:01, 77895.95KB/s]
35%|###4 | 46016/132723 [00:00<00:01, 78361.82KB/s]
41%|#### | 53965/132723 [00:00<00:01, 78726.99KB/s]
47%|####6 | 61932/132723 [00:00<00:00, 79024.24KB/s]
53%|#####2 | 69929/132723 [00:00<00:00, 79318.41KB/s]
59%|#####8 | 77865/132723 [00:01<00:00, 79326.37KB/s]
65%|######4 | 85798/132723 [00:01<00:00, 79315.41KB/s]
71%|####### | 93730/132723 [00:01<00:00, 79296.01KB/s]
77%|#######6 | 101713/132723 [00:01<00:00, 79456.54KB/s]
83%|########2 | 109659/132723 [00:01<00:00, 79446.40KB/s]
89%|########8 | 117613/132723 [00:01<00:00, 79470.92KB/s]
95%|########
#4| 125561/132723 [00:01<00:00, 79434.46KB/s]
100%|##########| 132723/132723 [00:01<00:00, 78487.37KB/s]
@@ -242,7 +242,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 57.793 seconds)
+ **Total running time of the script:** ( 2 minutes 59.652 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 f605eb7c53..db2025a784 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,24 +5,24 @@
Computation times
=================
-**12:51.604** total execution time for **how_to_deploy_models** files:
+**12:54.562** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:16.844 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:20.980 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:57.793 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:59.652 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:28.020 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 02:22.289 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:21.499 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:22.216 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:20.750 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:21.970 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:36.545 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:36.694 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:25.439 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:25.638 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:24.709 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:25.117 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.007 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
index efe324a383..b259f139ea 100644
--- a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
@@ -472,7 +472,7 @@ First let us define two helper functions to get the mobilenet model and a cat im
.. code-block:: none
- Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip094176ab-3666-4e35-9f1d-f186eef4307c from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipd4812d1f-f782-4b5f-a0f4-fecb03ebb59e 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 52d71c0396..9b95107683 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
=================
-**00:47.775** total execution time for **how_to_extend_tvm** files:
+**00:46.433** 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``) | 00:44.230 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:43.020 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.469 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.374 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.067 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:01.032 | 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 |
+| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.008 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index c37ed6772a..f0c50e73a6 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -216,10 +216,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 7149us [7149us] (46.17%; 46.17%)
- FoldScaleAxis: 8336us [6us] (53.83%; 53.83%)
- FoldConstant: 8330us [1735us] (53.79%; 99.93%)
- InferType: 6595us [6595us] (42.59%; 79.17%)
+ InferType: 6730us [6730us] (46.03%; 46.03%)
+ FoldScaleAxis: 7891us [5us] (53.97%; 53.97%)
+ FoldConstant: 7885us [1640us] (53.93%; 99.93%)
+ InferType: 6245us [6245us] (42.72%; 79.20%)
@@ -258,10 +258,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 6676us [6676us] (45.02%; 45.02%)
- FoldScaleAxis: 8152us [6us] (54.98%; 54.98%)
- FoldConstant: 8147us [1703us] (54.94%; 99.93%)
- InferType: 6444us [6444us] (43.46%; 79.10%)
+ InferType: 6273us [6273us] (44.56%; 44.56%)
+ FoldScaleAxis: 7805us [5us] (55.44%; 55.44%)
+ FoldConstant: 7801us [1625us] (55.41%; 99.94%)
+ InferType: 6175us [6175us] (43.86%; 79.16%)
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 97e4c98651..76e77dee23 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -340,7 +340,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 54.174976 ms
+ Convolution: 54.167552 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 c35d33f241..ba49d1669f 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
@@ -671,7 +671,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 13.356854 ms
+ conv2d with tensor core: 9.187165 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 0b44a536d4..1c4e903f1f 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -143,8 +143,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.019563
- Baseline: 3.465363
+ Numpy running time: 0.019505
+ Baseline: 3.489831
@@ -239,7 +239,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.310024
+ Opt1: 0.319293
@@ -342,7 +342,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.345835
+ Opt2: 0.351425
@@ -438,7 +438,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.120150
+ Opt3: 0.118191
@@ -563,7 +563,7 @@ flattening.
.. code-block:: none
- Opt4: 0.109662
+ Opt4: 0.109712
@@ -685,7 +685,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.112447
+ Opt5: 0.111329
@@ -810,7 +810,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.148376
+ Opt6: 0.147327
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 346210d552..e523c7d21e 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
=================
-**00:35.658** total execution time for **how_to_optimize_operators** files:
+**00:35.574** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.964 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:33.142 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.527 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.353 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.167 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.080 | 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 f7f36d2d5a..69b107411a 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,18 +5,18 @@
Computation times
=================
-**08:29.376** total execution time for **how_to_tune_with_autoscheduler** files:
+**08:23.574** 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``) | 05:03.406 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 04:57.852 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:32.824 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:32.655 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:03.095 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 01:02.447 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:27.842 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:29.023 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:11.270 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:10.962 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:10.939 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:10.635 | 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 356b12d18b..47a4b35e14 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -240,482 +240,360 @@ cooperative fetching, unrolling and operator fusion.
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 28;
- allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [28]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [1296]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [4608]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [196], [], scope="local", align=32)[0] = 0f32
+ conv2d_nchw_1[14] = 0f32
conv2d_nchw_1[1] = 0f32
+ conv2d_nchw_1[15] = 0f32
conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[16] = 0f32
conv2d_nchw_1[3] = 0f32
+ conv2d_nchw_1[17] = 0f32
conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[18] = 0f32
conv2d_nchw_1[5] = 0f32
+ conv2d_nchw_1[19] = 0f32
conv2d_nchw_1[6] = 0f32
+ conv2d_nchw_1[20] = 0f32
conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[21] = 0f32
conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[22] = 0f32
conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[23] = 0f32
conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[24] = 0f32
conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[25] = 0f32
conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[26] = 0f32
conv2d_nchw_1[13] = 0f32
- for (rc.outer.outer: int32, 0, 64) {
- for (ry.outer.outer: int32, 0, 3) {
- let cse_var_2: int32 = (rc.outer.outer*72)
- let cse_var_1: int32 = (ry.outer.outer*3)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [72], [], scope="shared")[(threadIdx.x_1*4)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod((threadIdx.x_1*4), 9))) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*4), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod((threadIdx.x_1*4), 9)) - 8)], 0f3 [...]
- }
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*4) + 1)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 1), 9))) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 1), 9)) - 8)], 0f32, dtype=float32)
- }
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*4) + 2)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 2), 9))) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 2), 9)) - 8)], 0f32, dtype=float32)
- }
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*4) + 3)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 3), 9))) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 3), 9)) - 8)], 0f32, dtype=float32)
+ conv2d_nchw_1[27] = 0f32
+ for (rc.outer.outer: int32, 0, 32) {
+ let cse_var_2: int32 = (rc.outer.outer*784)
+ let cse_var_1: int32 = (rc.outer.outer*144)
+ {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1296], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((9 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[(((cse_var_2 + (floordiv(threadIdx.x_1, 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 56)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 56), 81)) && (floormod((threadIdx.x_1 + 56), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 56), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 31), 81)) && (floormod((threadIdx.x_1 + 31), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 31), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 6), 81)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 6), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 62), 81)) && (floormod((threadIdx.x_1 + 62), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 62), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 37), 81)) && (floormod((threadIdx.x_1 + 37), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 37), 81), 9)*7)) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 3), 9)) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 336), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 12), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 68), 81)) && (floormod((threadIdx.x_1 + 68), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 68), 81), 9)*7)) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 43), 81)) && (floormod((threadIdx.x_1 + 43), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 448), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 43), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 504)] = @tir.if_then_else((((threadIdx.x_1 < 54) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 504), 81)*49)) + ((floordiv(threadIdx.x_1, 9) + 2)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 560)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 74), 81)) && (floormod((threadIdx.x_1 + 74), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 560), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 74), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 616)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 49), 81)) && (floormod((threadIdx.x_1 + 49), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 616), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 49), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 672)] = @tir.if_then_else((((threadIdx.x_1 < 48) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 672), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 24), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 728)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 80), 81)) && (floormod((threadIdx.x_1 + 80), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 728), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 80), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 55), 81)) && (floormod((threadIdx.x_1 + 55), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 784), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 55), 81), 9)*7)) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 840)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 30), 81)) && (floormod((threadIdx.x_1 + 30), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 840), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 30), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 896)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 5), 81)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 896), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 5), 81), 9)*7)) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 952)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 61), 81)) && (floormod((threadIdx.x_1 + 61), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 952), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 61), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1008)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 9) + 4), 9)) && (floormod((threadIdx.x_1 + 36), 81) < 72)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1008), 81)*49)) + (floormod((floordiv(threadIdx.x_1, 9) + 4), 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1064)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 2), 9)) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1064), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 11), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1120)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 67), 81)) && (floormod((threadIdx.x_1 + 67), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1120), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 67), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1176)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 42), 81)) && (floormod((threadIdx.x_1 + 42), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1176), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 42), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1232)] = @tir.if_then_else((((threadIdx.x_1 < 55) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1232), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 17), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_1 < 8), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 1288)] = 0f32
+ }
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1: Buffer(kernel.shared, float32, [4608], [], scope="shared")[threadIdx.x_2] = kernel[(((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[((((blockIdx.x*147456) + cse_var_1) + (floordiv((threadIdx.x_2 + 56), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 112), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 168), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 224), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 280)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 280), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 336), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 392), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 448), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 504)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 504), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 560), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 616)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 616), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 672), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 728)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 728), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 784), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 840)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 840), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 896), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 952)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 952), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1008)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 32256)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1064)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1064), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1120), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1176), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1232)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1232), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1288)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1288), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1344), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1400)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1400), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1456)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1456), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1512)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1512), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1568), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1624)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1624), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1680)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1680), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1736)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1736), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1792), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1848)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1848), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1904)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1904), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1960)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1960), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 64512)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2072)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2072), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2128)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2128), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2184)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2184), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2240), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2296)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2296), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2352)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2352), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2408)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2408), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2464)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2464), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2520)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2520), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2576)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2576), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2632)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2632), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2688), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2744)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2744), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2800)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2800), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2856)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2856), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2912)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2912), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2968)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2968), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3024)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 96768)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3080)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3080), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3136), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3192)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3192), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3248)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3248), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3304)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3304), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3360)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3360), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3416)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3416), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3472)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3472), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3528)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3528), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3584)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3584), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3640)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3640), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3696)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3696), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3752)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3752), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3808)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3808), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3864)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3864), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3920)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3920), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3976)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3976), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 129024)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4088)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4088), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4144)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4144), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4200)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4200), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4256)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4256), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4312)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4312), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4368)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4368), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4424)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4424), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4480)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4480), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4536)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4536), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_2 < 16), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 4592)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4592), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ }
+ for (rc.outer.inner: int32, 0, 2) {
+ for (rx.outer.inner: int32, 0, 3) {
+ for (rc.inner: int32, 0, 8) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
}
}
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 64), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 128), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 256), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 320), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 448), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 512), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 640), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 704), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 147456)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 832), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 896), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 184320)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1024), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1088), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 221184)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1216), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1280), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1408), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1472), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 294912)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1600), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1664), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 331776)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1792), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1856), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 368640)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1984), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2048), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 405504)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2176), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2240), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 442368)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2368), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2432), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 479232)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2560), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2624), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 516096)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2752), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2816), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 552960)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2944), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 3008), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((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[9]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 47)]))
}
}
}
for (i1.inner: int32, 0, 2) {
- for (i3.inner: int32, 0, 7) {
- compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*98)) + (i1.inner*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[((i1.inner*7) + i3.inner)] + bias[(((floordiv(blockIdx.x, 7)*128) + (threadIdx.x*2)) + i1.inner)]), 0f32)
+ for (i2.inner: int32, 0, 7) {
+ let cse_var_3: int32 = ((i1.inner*7) + i2.inner)
+ {
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[cse_var_3] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
+ compute[((((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7)) + 784)] = max((conv2d_nchw_1[(cse_var_3 + 14)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner) + 16)]), 0f32)
+ }
}
}
}
@@ -771,7 +649,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.355 ms
+ Execution time of this operator: 0.412 ms
@@ -819,21 +697,21 @@ 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_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
- conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
- conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
- conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, 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=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=2)
+ conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=7)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
- conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
- conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
+ conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+ conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
- conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
- conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=8)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+ conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
@@ -842,13 +720,13 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
- compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
- compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+ 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=2)
+ 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_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
- compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
- compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
+ compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+ compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -868,12 +746,12 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=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=4)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=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, "unroll_explicit", True)
@@ -893,429 +771,247 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+ extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[28];
+ __shared__ float pad_temp_shared[1296];
+ __shared__ float kernel_shared[4608];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[16] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[17] = 0.000000e+00f;
conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[18] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[19] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[20] = 0.000000e+00f;
conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[21] = 0.000000e+00f;
conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[22] = 0.000000e+00f;
conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[23] = 0.000000e+00f;
conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[24] = 0.000000e+00f;
conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[25] = 0.000000e+00f;
conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[26] = 0.000000e+00f;
conv2d_nchw[13] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
- for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 4) % 9))) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 1) % 9))) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 2) % 9))) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 3) % 9))) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
+ conv2d_nchw[27] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = ((((9 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((9 <= ((((int)threadIdx.x) + 56) % 81)) && (((((int)threadIdx.x) + 56) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 56) / 81) * 49)) + ((((((int)threadIdx.x) + 56) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((9 <= ((((int)threadIdx.x) + 31) % 81)) && (((((int)threadIdx.x) + 31) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 112) / 81) * 49)) + ((((((int)threadIdx.x) + 31) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((3 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 168) / 81) * 49)) + (((((int)threadIdx.x) + 6) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((9 <= ((((int)threadIdx.x) + 62) % 81)) && (((((int)threadIdx.x) + 62) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 224) / 81) * 49)) + ((((((int)threadIdx.x) + 62) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 280)] = (((((9 <= ((((int)threadIdx.x) + 37) % 81)) && (((((int)threadIdx.x) + 37) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 280) / 81) * 49)) + ((((((int)threadIdx.x) + 37) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 336)] = (((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 336) / 81) * 49)) + (((((int)threadIdx.x) + 12) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((9 <= ((((int)threadIdx.x) + 68) % 81)) && (((((int)threadIdx.x) + 68) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 392) / 81) * 49)) + ((((((int)threadIdx.x) + 68) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((9 <= ((((int)threadIdx.x) + 43) % 81)) && (((((int)threadIdx.x) + 43) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 448) / 81) * 49)) + ((((((int)threadIdx.x) + 43) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 504)] = ((((((int)threadIdx.x) < 54) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 504) / 81) * 49)) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) + 6)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((9 <= ((((int)threadIdx.x) + 74) % 81)) && (((((int)threadIdx.x) + 74) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 560) / 81) * 49)) + ((((((int)threadIdx.x) + 74) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 616)] = (((((9 <= ((((int)threadIdx.x) + 49) % 81)) && (((((int)threadIdx.x) + 49) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 616) / 81) * 49)) + ((((((int)threadIdx.x) + 49) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 672)] = ((((((int)threadIdx.x) < 48) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 672) / 81) * 49)) + (((((int)threadIdx.x) + 24) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 728)] = (((((9 <= ((((int)threadIdx.x) + 80) % 81)) && (((((int)threadIdx.x) + 80) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 728) / 81) * 49)) + ((((((int)threadIdx.x) + 80) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((9 <= ((((int)threadIdx.x) + 55) % 81)) && (((((int)threadIdx.x) + 55) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 784) / 81) * 49)) + ((((((int)threadIdx.x) + 55) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 840)] = (((((9 <= ((((int)threadIdx.x) + 30) % 81)) && (((((int)threadIdx.x) + 30) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 840) / 81) * 49)) + ((((((int)threadIdx.x) + 30) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 896)] = ((((4 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 896) / 81) * 49)) + (((((int)threadIdx.x) + 5) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 952)] = (((((9 <= ((((int)threadIdx.x) + 61) % 81)) && (((((int)threadIdx.x) + 61) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 952) / 81) * 49)) + ((((((int)threadIdx.x) + 61) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1008)] = (((((1 <= (((((int)threadIdx.x) / 9) + 4) % 9)) && (((((int)threadIdx.x) + 36) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1008) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 4) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1064)] = (((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1064) / 81) * 49)) + (((((int)threadIdx.x) + 11) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1120)] = (((((9 <= ((((int)threadIdx.x) + 67) % 81)) && (((((int)threadIdx.x) + 67) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1120) / 81) * 49)) + ((((((int)threadIdx.x) + 67) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((9 <= ((((int)threadIdx.x) + 42) % 81)) && (((((int)threadIdx.x) + 42) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1176) / 81) * 49)) + ((((((int)threadIdx.x) + 42) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1232)] = ((((((int)threadIdx.x) < 55) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1232) / 81) * 49)) + (((((int)threadIdx.x) + 17) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 8) {
+ pad_temp_shared[(((int)threadIdx.x) + 1288)] = 0.000000e+00f;
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 504) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 616) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 672) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 728) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 840) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 896) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 952) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1064) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1120) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1232) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1288) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1344) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1400) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1456) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1512) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1624) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1680) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1736) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1792) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1848) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1904) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 2072)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2072) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2128) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2184)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2184) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2240) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2296)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2296) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 2408)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2408) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2464) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2520)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2520) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 2576)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2576) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2632)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2632) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2688) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2800)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2800) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2856)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2856) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2912) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2968)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2968) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3024)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 96768)];
+ kernel_shared[(((int)threadIdx.x) + 3080)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3080) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3136) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3192)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3192) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 3248)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3248) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3304)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3304) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3360) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 3416)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3416) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3472)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3472) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3528) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3584) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3640)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3640) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3696)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3696) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3752)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3752) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3808) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3864)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3864) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3920) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3976)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3976) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 129024)];
+ kernel_shared[(((int)threadIdx.x) + 4088)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4088) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4144)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4144) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4200)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4200) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4256) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4312) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4368)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4368) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 4424)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4424) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4480) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4536)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4536) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ if (((int)threadIdx.x) < 16) {
+ kernel_shared[(((int)threadIdx.x) + 4592)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4592) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 128) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 2; ++rc_outer_inner) {
+ for (int rx_outer_inner = 0; rx_outer_inner < 3; ++rx_outer_inner) {
+ for (int rc_inner = 0; rc_inner < 8; ++rc_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ }
}
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((int)threadIdx.x) + 64)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
- kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
- kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
- kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
- kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
- kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
- kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
- kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
- kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
}
}
for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[((((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7)) + 784)] = max((conv2d_nchw[(((i1_inner * 7) + i2_inner) + 14)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
}
}
}
@@ -1378,7 +1074,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 5 minutes 3.406 seconds)
+ **Total running time of the script:** ( 4 minutes 57.852 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 37fbaebb03..814e189cc0 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -643,7 +643,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1759 8.1757 8.1766 8.1755 0.0005
+ 8.2262 8.2225 8.2344 8.2218 0.0058
@@ -671,7 +671,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 3.095 seconds)
+ **Total running time of the script:** ( 1 minutes 2.447 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 babc90f9eb..b148c37761 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -662,7 +662,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 760.6483 760.0349 764.3487 757.5613 2.8047
+ 750.4478 750.8018 751.6778 748.8640 1.1757
@@ -690,7 +690,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 32.824 seconds)
+ **Total running time of the script:** ( 1 minutes 32.655 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 682dd40289..472c10b89c 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
@@ -386,30 +386,1039 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_6: placeholder_17: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_7: placeholder_19: Buffer(placeholder_12, int32, [4916], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 16) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [4096]), storage_scope = global {
- for (i.outer.inner: int32, 0, 8) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 16) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [4096], [])[((((i.outer.inner*512) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
- }
- }
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (i.inner: int32, 0, 16) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner)
- let cse_var_2: int32 = ((((i.outer.inner*512) + (i.inner*32)) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((i.outer.inner*4096) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
- }
- }
- }
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_16: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
+ let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ allocate(compute_4: Pointer(global float32), float32, [256]), storage_scope = global {
+ compute_5: Buffer(compute_4, float32, [256], [])[0] = 0f32
+ compute_5[1] = 0f32
+ compute_5[2] = 0f32
+ compute_5[3] = 0f32
+ compute_5[4] = 0f32
+ compute_5[5] = 0f32
+ compute_5[6] = 0f32
+ compute_5[7] = 0f32
+ compute_5[8] = 0f32
+ compute_5[9] = 0f32
+ compute_5[10] = 0f32
+ compute_5[11] = 0f32
+ compute_5[12] = 0f32
+ compute_5[13] = 0f32
+ compute_5[14] = 0f32
+ compute_5[15] = 0f32
+ compute_5[16] = 0f32
+ compute_5[17] = 0f32
+ compute_5[18] = 0f32
+ compute_5[19] = 0f32
+ compute_5[20] = 0f32
+ compute_5[21] = 0f32
+ compute_5[22] = 0f32
+ compute_5[23] = 0f32
+ compute_5[24] = 0f32
+ compute_5[25] = 0f32
+ compute_5[26] = 0f32
+ compute_5[27] = 0f32
+ compute_5[28] = 0f32
+ compute_5[29] = 0f32
+ compute_5[30] = 0f32
+ compute_5[31] = 0f32
+ compute_5[32] = 0f32
+ compute_5[33] = 0f32
+ compute_5[34] = 0f32
+ compute_5[35] = 0f32
+ compute_5[36] = 0f32
+ compute_5[37] = 0f32
+ compute_5[38] = 0f32
+ compute_5[39] = 0f32
+ compute_5[40] = 0f32
+ compute_5[41] = 0f32
+ compute_5[42] = 0f32
+ compute_5[43] = 0f32
+ compute_5[44] = 0f32
+ compute_5[45] = 0f32
+ compute_5[46] = 0f32
+ compute_5[47] = 0f32
+ compute_5[48] = 0f32
+ compute_5[49] = 0f32
+ compute_5[50] = 0f32
+ compute_5[51] = 0f32
+ compute_5[52] = 0f32
+ compute_5[53] = 0f32
+ compute_5[54] = 0f32
+ compute_5[55] = 0f32
+ compute_5[56] = 0f32
+ compute_5[57] = 0f32
+ compute_5[58] = 0f32
+ compute_5[59] = 0f32
+ compute_5[60] = 0f32
+ compute_5[61] = 0f32
+ compute_5[62] = 0f32
+ compute_5[63] = 0f32
+ compute_5[64] = 0f32
+ compute_5[65] = 0f32
+ compute_5[66] = 0f32
+ compute_5[67] = 0f32
+ compute_5[68] = 0f32
+ compute_5[69] = 0f32
+ compute_5[70] = 0f32
+ compute_5[71] = 0f32
+ compute_5[72] = 0f32
+ compute_5[73] = 0f32
+ compute_5[74] = 0f32
+ compute_5[75] = 0f32
+ compute_5[76] = 0f32
+ compute_5[77] = 0f32
+ compute_5[78] = 0f32
+ compute_5[79] = 0f32
+ compute_5[80] = 0f32
+ compute_5[81] = 0f32
+ compute_5[82] = 0f32
+ compute_5[83] = 0f32
+ compute_5[84] = 0f32
+ compute_5[85] = 0f32
+ compute_5[86] = 0f32
+ compute_5[87] = 0f32
+ compute_5[88] = 0f32
+ compute_5[89] = 0f32
+ compute_5[90] = 0f32
+ compute_5[91] = 0f32
+ compute_5[92] = 0f32
+ compute_5[93] = 0f32
+ compute_5[94] = 0f32
+ compute_5[95] = 0f32
+ compute_5[96] = 0f32
+ compute_5[97] = 0f32
+ compute_5[98] = 0f32
+ compute_5[99] = 0f32
+ compute_5[100] = 0f32
+ compute_5[101] = 0f32
+ compute_5[102] = 0f32
+ compute_5[103] = 0f32
+ compute_5[104] = 0f32
+ compute_5[105] = 0f32
+ compute_5[106] = 0f32
+ compute_5[107] = 0f32
+ compute_5[108] = 0f32
+ compute_5[109] = 0f32
+ compute_5[110] = 0f32
+ compute_5[111] = 0f32
+ compute_5[112] = 0f32
+ compute_5[113] = 0f32
+ compute_5[114] = 0f32
+ compute_5[115] = 0f32
+ compute_5[116] = 0f32
+ compute_5[117] = 0f32
+ compute_5[118] = 0f32
+ compute_5[119] = 0f32
+ compute_5[120] = 0f32
+ compute_5[121] = 0f32
+ compute_5[122] = 0f32
+ compute_5[123] = 0f32
+ compute_5[124] = 0f32
+ compute_5[125] = 0f32
+ compute_5[126] = 0f32
+ compute_5[127] = 0f32
+ compute_5[128] = 0f32
+ compute_5[129] = 0f32
+ compute_5[130] = 0f32
+ compute_5[131] = 0f32
+ compute_5[132] = 0f32
+ compute_5[133] = 0f32
+ compute_5[134] = 0f32
+ compute_5[135] = 0f32
+ compute_5[136] = 0f32
+ compute_5[137] = 0f32
+ compute_5[138] = 0f32
+ compute_5[139] = 0f32
+ compute_5[140] = 0f32
+ compute_5[141] = 0f32
+ compute_5[142] = 0f32
+ compute_5[143] = 0f32
+ compute_5[144] = 0f32
+ compute_5[145] = 0f32
+ compute_5[146] = 0f32
+ compute_5[147] = 0f32
+ compute_5[148] = 0f32
+ compute_5[149] = 0f32
+ compute_5[150] = 0f32
+ compute_5[151] = 0f32
+ compute_5[152] = 0f32
+ compute_5[153] = 0f32
+ compute_5[154] = 0f32
+ compute_5[155] = 0f32
+ compute_5[156] = 0f32
+ compute_5[157] = 0f32
+ compute_5[158] = 0f32
+ compute_5[159] = 0f32
+ compute_5[160] = 0f32
+ compute_5[161] = 0f32
+ compute_5[162] = 0f32
+ compute_5[163] = 0f32
+ compute_5[164] = 0f32
+ compute_5[165] = 0f32
+ compute_5[166] = 0f32
+ compute_5[167] = 0f32
+ compute_5[168] = 0f32
+ compute_5[169] = 0f32
+ compute_5[170] = 0f32
+ compute_5[171] = 0f32
+ compute_5[172] = 0f32
+ compute_5[173] = 0f32
+ compute_5[174] = 0f32
+ compute_5[175] = 0f32
+ compute_5[176] = 0f32
+ compute_5[177] = 0f32
+ compute_5[178] = 0f32
+ compute_5[179] = 0f32
+ compute_5[180] = 0f32
+ compute_5[181] = 0f32
+ compute_5[182] = 0f32
+ compute_5[183] = 0f32
+ compute_5[184] = 0f32
+ compute_5[185] = 0f32
+ compute_5[186] = 0f32
+ compute_5[187] = 0f32
+ compute_5[188] = 0f32
+ compute_5[189] = 0f32
+ compute_5[190] = 0f32
+ compute_5[191] = 0f32
+ compute_5[192] = 0f32
+ compute_5[193] = 0f32
+ compute_5[194] = 0f32
+ compute_5[195] = 0f32
+ compute_5[196] = 0f32
+ compute_5[197] = 0f32
+ compute_5[198] = 0f32
+ compute_5[199] = 0f32
+ compute_5[200] = 0f32
+ compute_5[201] = 0f32
+ compute_5[202] = 0f32
+ compute_5[203] = 0f32
+ compute_5[204] = 0f32
+ compute_5[205] = 0f32
+ compute_5[206] = 0f32
+ compute_5[207] = 0f32
+ compute_5[208] = 0f32
+ compute_5[209] = 0f32
+ compute_5[210] = 0f32
+ compute_5[211] = 0f32
+ compute_5[212] = 0f32
+ compute_5[213] = 0f32
+ compute_5[214] = 0f32
+ compute_5[215] = 0f32
+ compute_5[216] = 0f32
+ compute_5[217] = 0f32
+ compute_5[218] = 0f32
+ compute_5[219] = 0f32
+ compute_5[220] = 0f32
+ compute_5[221] = 0f32
+ compute_5[222] = 0f32
+ compute_5[223] = 0f32
+ compute_5[224] = 0f32
+ compute_5[225] = 0f32
+ compute_5[226] = 0f32
+ compute_5[227] = 0f32
+ compute_5[228] = 0f32
+ compute_5[229] = 0f32
+ compute_5[230] = 0f32
+ compute_5[231] = 0f32
+ compute_5[232] = 0f32
+ compute_5[233] = 0f32
+ compute_5[234] = 0f32
+ compute_5[235] = 0f32
+ compute_5[236] = 0f32
+ compute_5[237] = 0f32
+ compute_5[238] = 0f32
+ compute_5[239] = 0f32
+ compute_5[240] = 0f32
+ compute_5[241] = 0f32
+ compute_5[242] = 0f32
+ compute_5[243] = 0f32
+ compute_5[244] = 0f32
+ compute_5[245] = 0f32
+ compute_5[246] = 0f32
+ compute_5[247] = 0f32
+ compute_5[248] = 0f32
+ compute_5[249] = 0f32
+ compute_5[250] = 0f32
+ compute_5[251] = 0f32
+ compute_5[252] = 0f32
+ compute_5[253] = 0f32
+ compute_5[254] = 0f32
+ compute_5[255] = 0f32
+ for (elem_idx: int32, 0, (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[0] = (compute_5[0] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[1] = (compute_5[1] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[2] = (compute_5[2] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[3] = (compute_5[3] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[4] = (compute_5[4] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[5] = (compute_5[5] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[6] = (compute_5[6] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[7] = (compute_5[7] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[8] = (compute_5[8] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[9] = (compute_5[9] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[10] = (compute_5[10] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[11] = (compute_5[11] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[12] = (compute_5[12] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[13] = (compute_5[13] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[14] = (compute_5[14] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[15] = (compute_5[15] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[16] = (compute_5[16] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[17] = (compute_5[17] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[18] = (compute_5[18] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[19] = (compute_5[19] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[20] = (compute_5[20] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[21] = (compute_5[21] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[22] = (compute_5[22] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[23] = (compute_5[23] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[24] = (compute_5[24] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[25] = (compute_5[25] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[26] = (compute_5[26] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[27] = (compute_5[27] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[28] = (compute_5[28] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[29] = (compute_5[29] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[30] = (compute_5[30] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[31] = (compute_5[31] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[32] = (compute_5[32] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[33] = (compute_5[33] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[34] = (compute_5[34] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[35] = (compute_5[35] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[36] = (compute_5[36] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[37] = (compute_5[37] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[38] = (compute_5[38] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[39] = (compute_5[39] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[40] = (compute_5[40] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[41] = (compute_5[41] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[42] = (compute_5[42] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[43] = (compute_5[43] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[44] = (compute_5[44] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[45] = (compute_5[45] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[46] = (compute_5[46] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[47] = (compute_5[47] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[48] = (compute_5[48] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[49] = (compute_5[49] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[50] = (compute_5[50] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[51] = (compute_5[51] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[52] = (compute_5[52] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[53] = (compute_5[53] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[54] = (compute_5[54] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[55] = (compute_5[55] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[56] = (compute_5[56] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[57] = (compute_5[57] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[58] = (compute_5[58] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[59] = (compute_5[59] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[60] = (compute_5[60] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[61] = (compute_5[61] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[62] = (compute_5[62] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[63] = (compute_5[63] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[64] = (compute_5[64] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[65] = (compute_5[65] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[66] = (compute_5[66] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[67] = (compute_5[67] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[68] = (compute_5[68] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[69] = (compute_5[69] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[70] = (compute_5[70] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[71] = (compute_5[71] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[72] = (compute_5[72] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[73] = (compute_5[73] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[74] = (compute_5[74] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[75] = (compute_5[75] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[76] = (compute_5[76] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[77] = (compute_5[77] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[78] = (compute_5[78] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[79] = (compute_5[79] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[80] = (compute_5[80] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[81] = (compute_5[81] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[82] = (compute_5[82] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[83] = (compute_5[83] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[84] = (compute_5[84] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[85] = (compute_5[85] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[86] = (compute_5[86] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[87] = (compute_5[87] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[88] = (compute_5[88] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[89] = (compute_5[89] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[90] = (compute_5[90] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[91] = (compute_5[91] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[92] = (compute_5[92] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[93] = (compute_5[93] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[94] = (compute_5[94] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[95] = (compute_5[95] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[96] = (compute_5[96] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[97] = (compute_5[97] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[98] = (compute_5[98] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[99] = (compute_5[99] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[100] = (compute_5[100] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[101] = (compute_5[101] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[102] = (compute_5[102] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[103] = (compute_5[103] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[104] = (compute_5[104] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[105] = (compute_5[105] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[106] = (compute_5[106] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[107] = (compute_5[107] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[108] = (compute_5[108] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[109] = (compute_5[109] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[110] = (compute_5[110] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[111] = (compute_5[111] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[112] = (compute_5[112] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[113] = (compute_5[113] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[114] = (compute_5[114] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[115] = (compute_5[115] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[116] = (compute_5[116] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[117] = (compute_5[117] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[118] = (compute_5[118] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[119] = (compute_5[119] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[120] = (compute_5[120] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[121] = (compute_5[121] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[122] = (compute_5[122] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[123] = (compute_5[123] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[124] = (compute_5[124] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[125] = (compute_5[125] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[126] = (compute_5[126] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[127] = (compute_5[127] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[128] = (compute_5[128] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[129] = (compute_5[129] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[130] = (compute_5[130] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[131] = (compute_5[131] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[132] = (compute_5[132] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[133] = (compute_5[133] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[134] = (compute_5[134] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[135] = (compute_5[135] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[136] = (compute_5[136] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[137] = (compute_5[137] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[138] = (compute_5[138] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[139] = (compute_5[139] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[140] = (compute_5[140] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[141] = (compute_5[141] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[142] = (compute_5[142] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[143] = (compute_5[143] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[144] = (compute_5[144] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[145] = (compute_5[145] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[146] = (compute_5[146] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[147] = (compute_5[147] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[148] = (compute_5[148] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[149] = (compute_5[149] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[150] = (compute_5[150] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[151] = (compute_5[151] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[152] = (compute_5[152] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[153] = (compute_5[153] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[154] = (compute_5[154] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[155] = (compute_5[155] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[156] = (compute_5[156] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[157] = (compute_5[157] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[158] = (compute_5[158] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[159] = (compute_5[159] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[160] = (compute_5[160] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[161] = (compute_5[161] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[162] = (compute_5[162] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[163] = (compute_5[163] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[164] = (compute_5[164] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[165] = (compute_5[165] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[166] = (compute_5[166] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[167] = (compute_5[167] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[168] = (compute_5[168] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[169] = (compute_5[169] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[170] = (compute_5[170] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[171] = (compute_5[171] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[172] = (compute_5[172] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[173] = (compute_5[173] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[174] = (compute_5[174] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[175] = (compute_5[175] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[176] = (compute_5[176] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[177] = (compute_5[177] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[178] = (compute_5[178] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[179] = (compute_5[179] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[180] = (compute_5[180] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[181] = (compute_5[181] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[182] = (compute_5[182] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[183] = (compute_5[183] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[184] = (compute_5[184] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[185] = (compute_5[185] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[186] = (compute_5[186] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[187] = (compute_5[187] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[188] = (compute_5[188] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[189] = (compute_5[189] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[190] = (compute_5[190] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[191] = (compute_5[191] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[192] = (compute_5[192] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[193] = (compute_5[193] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[194] = (compute_5[194] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[195] = (compute_5[195] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[196] = (compute_5[196] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[197] = (compute_5[197] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[198] = (compute_5[198] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[199] = (compute_5[199] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[200] = (compute_5[200] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[201] = (compute_5[201] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[202] = (compute_5[202] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[203] = (compute_5[203] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[204] = (compute_5[204] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[205] = (compute_5[205] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[206] = (compute_5[206] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[207] = (compute_5[207] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[208] = (compute_5[208] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[209] = (compute_5[209] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[210] = (compute_5[210] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[211] = (compute_5[211] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[212] = (compute_5[212] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[213] = (compute_5[213] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[214] = (compute_5[214] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[215] = (compute_5[215] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[216] = (compute_5[216] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[217] = (compute_5[217] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[218] = (compute_5[218] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[219] = (compute_5[219] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[220] = (compute_5[220] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[221] = (compute_5[221] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[222] = (compute_5[222] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[223] = (compute_5[223] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[224] = (compute_5[224] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[225] = (compute_5[225] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[226] = (compute_5[226] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[227] = (compute_5[227] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[228] = (compute_5[228] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[229] = (compute_5[229] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[230] = (compute_5[230] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[231] = (compute_5[231] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[232] = (compute_5[232] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[233] = (compute_5[233] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[234] = (compute_5[234] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[235] = (compute_5[235] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[236] = (compute_5[236] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[237] = (compute_5[237] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[238] = (compute_5[238] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[239] = (compute_5[239] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[240] = (compute_5[240] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[241] = (compute_5[241] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[242] = (compute_5[242] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[243] = (compute_5[243] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[244] = (compute_5[244] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[245] = (compute_5[245] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[246] = (compute_5[246] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[247] = (compute_5[247] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[248] = (compute_5[248] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[249] = (compute_5[249] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[250] = (compute_5[250] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[251] = (compute_5[251] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[252] = (compute_5[252] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[253] = (compute_5[253] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[254] = (compute_5[254] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[255] = (compute_5[255] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
}
}
- for (i0.inner: int32, 0, 128) {
- let cse_var_4: int32 = ((i0.inner*512) + (i0.outer.i1.outer.fused*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 16) {
+ let cse_var_2: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i0.inner*512)) + (cse_var_1*16))
+ compute[ramp(cse_var_2, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_2, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -465,7 +1474,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.498 ms
+ Execution time of this operator: 3.546 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 70e5a14f7f..31bbb6bc11 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
Computation times
=================
-**00:34.299** total execution time for **how_to_tune_with_autotvm** files:
+**00:29.623** total execution time for **how_to_tune_with_autotvm** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:34.263 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:29.587 | 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``) | 00:00.021 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``) | 00:00.005 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index edf3b04f17..fdae2a8b95 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
@@ -399,8 +399,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 16, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5634509
- No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5575908
+ No: 2 GFLOPS: 9.07/9.07 result: MeasureResult(costs=(0.02553097525,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.472261667251587, timestamp=1665986039.31081) [('tile_f', [-1, 8, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3497937
+ No: 3 GFLOPS: 0.00/9.07 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -522,9 +523,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6063340
- No: 3 GFLOPS: 1.30/1.30 result: MeasureResult(costs=(0.17854515075,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.044389009475708, timestamp=1665892986.3266659) [('tile_f', [-1, 2, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1907181
- No: 4 GFLOPS: 0.00/1.30 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8691660
+ No: 4 GFLOPS: 0.00/9.07 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -646,9 +646,10 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9813116
- No: 5 GFLOPS: 65.06/65.06 result: MeasureResult(costs=(0.003558434909090909,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3140735626220703, timestamp=1665892988.8143306) [('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8760073
- No: 6 GFLOPS: 0.00/65.06 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2807403
+ No: 5 GFLOPS: 37.86/37.86 result: MeasureResult(costs=(0.006114725411764706,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5770478248596191, timestamp=1665986043.0153155) [('tile_f', [-1, 4, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,98572
+ No: 6 GFLOPS: 100.64/100.64 result: MeasureResult(costs=(0.002300276318181818,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4073760509490967, timestamp=1665986043.7016363) [('tile_f', [-1, 4, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4649736
+ No: 7 GFLOPS: 0.00/100.64 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -770,8 +771,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('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,7580739
- No: 7 GFLOPS: 0.00/65.06 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2303991
+ No: 8 GFLOPS: 338.83/338.83 result: MeasureResult(costs=(0.0006832309115646258,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.099130630493164, timestamp=1665986045.2803166) [('tile_f', [-1, 8, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2336027
+ No: 9 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -893,8 +895,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 128, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4563777
- No: 8 GFLOPS: 0.00/65.06 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 64, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7379334
+ No: 10 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1016,11 +1018,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 8, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,433212
- No: 9 GFLOPS: 787.40/787.40 result: MeasureResult(costs=(0.00029400837535014003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2635645866394043, timestamp=1665892993.1627276) [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9221146
- No: 10 GFLOPS: 257.16/787.40 result: MeasureResult(costs=(0.0009002157062146893,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7292754650115967, timestamp=1665892994.0822725) [('tile_f', [-1, 2, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3945162
- No: 11 GFLOPS: 98.73/787.40 result: MeasureResult(costs=(0.002344907162790698,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1468474864959717, timestamp=1665892994.7787464) [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5859414
- No: 12 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2435702
+ No: 11 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1142,8 +1141,9 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7516258
- No: 13 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 512]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1389739
+ No: 12 GFLOPS: 6.66/338.83 result: MeasureResult(costs=(0.034783584500000006,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4205341339111328, timestamp=1665986046.9337785) [('tile_f', [-1, 4, 4, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,7877
+ No: 13 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1265,8 +1265,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9387431
- No: 14 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,843834
+ No: 14 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1388,8 +1388,10 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 512]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10438999
- No: 15 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6378833
+ No: 15 GFLOPS: 22.50/338.83 result: MeasureResult(costs=(0.010287662818181819,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6355750560760498, timestamp=1665986051.6126497) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2252599
+ No: 16 GFLOPS: 5.41/338.83 result: MeasureResult(costs=(0.04276027275,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.494520425796509, timestamp=1665986052.5202076) [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7074676
+ No: 17 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1511,8 +1513,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7666966
- No: 16 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10224784
+ No: 18 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1634,8 +1636,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3772509
- No: 17 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 4]), ('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, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2735610
+ No: 19 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1757,9 +1759,8 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8003956
- No: 18 GFLOPS: 10.68/787.40 result: MeasureResult(costs=(0.021684544833333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.315142631530762, timestamp=1665893000.459618) [('tile_f', [-1, 32, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1873315
- No: 19 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 8]), ('tile_y', [-1, 1, 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', 1)],None,7440558
+ No: 20 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1881,130 +1882,7 @@ for this template
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4865648
- No: 20 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
- func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
- func = build(s, args, target_host=task.target_host, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
- File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
- File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
- File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
- tvm._ffi.base.TVMError: Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1731
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1671
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1631
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1646
- 13: operator()
- at ../src/driver/driver_api.cc:380
- 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:366
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:261
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:453
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1750
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1694
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1618
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
-
- Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1731
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1671
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1631
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1646
- 13: operator()
- at ../src/driver/driver_api.cc:380
- 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:366
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:261
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:453
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1750
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1694
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1618
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
- tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1897725
+ tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6521196
@@ -2059,9 +1937,9 @@ and measure running time.
Finish loading 20 records
Best config:
- [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9221146
+ [('tile_f', [-1, 8, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2336027
Finish loading 20 records
- Time cost of this operator: 0.000657
+ Time cost of this operator: 0.001080
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 adb1a2437d..f8da4bccac 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -327,10 +327,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 313.3 98.735 (1, 2, 10, 10, 3) 2 1 [313.3]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.039 0.958 (1, 6, 10, 10) 1 1 [3.039]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.976 0.307 (1, 1, 10, 10, 3) 1 1 [0.976]
- Total_time - 317.315 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 311.7 98.727 (1, 2, 10, 10, 3) 2 1 [311.7]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.048 0.966 (1, 6, 10, 10) 1 1 [3.048]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 0.307 (1, 1, 10, 10, 3) 1 1 [0.969]
+ Total_time - 315.718 - - - - -
@@ -394,10 +394,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.8 97.338 (1, 6, 10, 10, 1) 2 1 [100.8]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.788 1.726 (1, 6, 10, 10) 1 1 [1.788]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 0.936 (1, 1, 10, 10, 3) 1 1 [0.969]
- Total_time - 103.557 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 102.6 97.486 (1, 6, 10, 10, 1) 2 1 [102.6]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.805 1.715 (1, 6, 10, 10) 1 1 [1.805]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.841 0.799 (1, 3, 10, 10, 1) 1 1 [0.841]
+ Total_time - 105.246 - - - - -
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 b280d6176b..0301481e74 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -225,7 +225,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmpgbq49kqt/images/random'
+ '/tmp/tmpzq21sad0/images/random'
@@ -316,7 +316,7 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. image-sg:: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
- :alt: [0.0, 1.0], [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], [0.0, 1.0], [0.0, 1.0]
+ :alt: [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0]
:srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
:class: sphx-glr-single-img
@@ -325,8 +325,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmpgbq49kqt/images/target contains 8144 images
- /tmp/tmpgbq49kqt/images/random contains 5000 images
+ /tmp/tmpzq21sad0/images/target contains 8144 images
+ /tmp/tmpzq21sad0/images/random contains 5000 images
@@ -501,13 +501,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 47s - loss: 0.2382 - accuracy: 0.9208 - val_loss: 0.1159 - val_accuracy: 0.9573 - 47s/epoch - 144ms/step
+ 328/328 - 47s - loss: 0.2032 - accuracy: 0.9269 - val_loss: 0.1450 - val_accuracy: 0.9520 - 47s/epoch - 142ms/step
Epoch 2/3
- 328/328 - 43s - loss: 0.1063 - accuracy: 0.9589 - val_loss: 0.1400 - val_accuracy: 0.9505 - 43s/epoch - 133ms/step
+ 328/328 - 43s - loss: 0.0946 - accuracy: 0.9623 - val_loss: 0.1637 - val_accuracy: 0.9475 - 43s/epoch - 131ms/step
Epoch 3/3
- 328/328 - 43s - loss: 0.0643 - accuracy: 0.9746 - val_loss: 0.0948 - val_accuracy: 0.9690 - 43s/epoch - 132ms/step
+ 328/328 - 43s - loss: 0.0555 - accuracy: 0.9786 - val_loss: 0.1109 - val_accuracy: 0.9690 - 43s/epoch - 131ms/step
- <keras.callbacks.History object at 0x7fb3e0199750>
+ <keras.callbacks.History object at 0x7f90cbd78cd0>
@@ -864,7 +864,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 27.390 seconds)
+ **Total running time of the script:** ( 4 minutes 23.229 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 e42b60d187..6d07540472 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,16 +5,16 @@
Computation times
=================
-**05:31.738** total execution time for **how_to_work_with_microtvm** files:
+**05:25.689** total execution time for **how_to_work_with_microtvm** files:
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:27.390 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:23.229 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:50.894 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:49.363 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:09.597 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:09.335 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.855 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.761 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``) | 00:00.001 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index 802a878fd1..da290132ab 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
=================
-**00:45.047** total execution time for **how_to_work_with_relay** files:
+**00:42.422** total execution time for **how_to_work_with_relay** files:
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.795 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.124 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:10.440 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:08.415 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.806 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.876 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.007 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
index be32ed94b9..1355346019 100644
--- a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
@@ -261,7 +261,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7fb3cff18e60>
+ <function my_cuda_math_rule at 0x7f91245228c0>
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 97c6c137c1..bb96ae3420 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
@@ -5,20 +5,20 @@
Computation times
=================
-**00:08.092** total execution time for **how_to_work_with_schedules** files:
+**00:07.432** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:05.613 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:05.066 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.117 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.034 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.584 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.569 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.567 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.554 | 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_extern_op.py` (``extern_op.py``) | 00:00.115 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.047 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.048 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.029 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
index aa7865b8ee..621528ea5b 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -347,7 +347,7 @@ The importing needs to happen before the tensorized GEMV being executed.
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp3f0a0xyg/input0.cc'\nsource_filename = \"/tmp/tmp3f0a0xyg/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmphly3l076/input0.cc'\nsource_filename = \"/tmp/tmphly3l076/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index bf024247d1..60f74359e4 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
=================
-**00:26.919** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:25.979** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:26.912 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:25.973 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.006 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
index f1e3cc35dc..7d29bf7aa3 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -289,7 +289,7 @@ The compilation steps are:
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
- resnet18_v1 inference graph built in 28.83s!
+ resnet18_v1 inference graph built in 27.71s!
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
index 07a8869a7d..0bbf74ca25 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -333,7 +333,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 19.85s!
+ yolov3-tiny inference graph built in 19.21s!
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 dc223fadde..11f7d48acf 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
=================
-**01:41.140** total execution time for **topic_vta_tutorials_frontend** files:
+**01:38.820** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:52.104 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:51.192 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:49.037 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:47.628 | 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 385b3dad5c..2e6419d017 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
=================
-**00:03.170** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.059** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.704 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.622 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.465 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.437 | 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 0952991381..a361694e8a 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
=================
-**00:00.827** total execution time for **topic_vta_tutorials** files:
+**00:00.780** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.447 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.417 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.381 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.363 | 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 1d34c269e0..34b0569c7b 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -326,7 +326,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 95.033 ms
+ Execution time of this operator: 93.543 ms
@@ -444,7 +444,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 18.373 seconds)
+ **Total running time of the script:** ( 1 minutes 13.803 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 f45778fee3..4ddf3426dd 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -462,16 +462,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 3.82/3.82 result: MeasureResult(costs=(0.070208407,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.296924352645874, timestamp=1665891607.1500974) [('tile_y', [-1, 4]), ('tile_x', [-1, 16])],None,42
- No: 2 GFLOPS: 3.91/3.91 result: MeasureResult(costs=(0.06870180179999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3101081848144531, timestamp=1665891608.4555542) [('tile_y', [-1, 32]), ('tile_x', [-1, 16])],None,45
- No: 3 GFLOPS: 12.99/12.99 result: MeasureResult(costs=(0.020665182400000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4857490062713623, timestamp=1665891609.710765) [('tile_y', [-1, 4]), ('tile_x', [-1, 512])],None,92
- No: 4 GFLOPS: 9.16/12.99 result: MeasureResult(costs=(0.029303448,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.9238376617431641, timestamp=1665891610.3498635) [('tile_y', [-1, 16]), ('tile_x', [-1, 128])],None,74
- No: 5 GFLOPS: 0.89/12.99 result: MeasureResult(costs=(0.3010986344,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.966350078582764, timestamp=1665891615.4445813) [('tile_y', [-1, 128]), ('tile_x', [-1, 2])],None,17
- No: 6 GFLOPS: 3.52/12.99 result: MeasureResult(costs=(0.0762139024,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3772945404052734, timestamp=1665891617.58973) [('tile_y', [-1, 16]), ('tile_x', [-1, 8])],None,34
- No: 7 GFLOPS: 12.28/12.99 result: MeasureResult(costs=(0.0218537594,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5233402252197266, timestamp=1665891618.8733473) [('tile_y', [-1, 64]), ('tile_x', [-1, 256])],None,86
- No: 8 GFLOPS: 3.27/12.99 result: MeasureResult(costs=(0.0819798514,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5147910118103027, timestamp=1665891620.3954968) [('tile_y', [-1, 32]), ('tile_x', [-1, 8])],None,35
- No: 9 GFLOPS: 2.77/12.99 result: MeasureResult(costs=(0.09708193039999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6804118156433105, timestamp=1665891622.1922264) [('tile_y', [-1, 16]), ('tile_x', [-1, 4])],None,24
- No: 10 GFLOPS: 1.09/12.99 result: MeasureResult(costs=(0.24729373139999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.9578802585601807, timestamp=1665891626.1980004) [('tile_y', [-1, 16]), ('tile_x', [-1, 1])],None,4
+ No: 1 GFLOPS: 2.77/2.77 result: MeasureResult(costs=(0.0967549174,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7043232917785645, timestamp=1665984679.5478013) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
+ No: 2 GFLOPS: 11.56/11.56 result: MeasureResult(costs=(0.0232139088,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.595611572265625, timestamp=1665984680.1249952) [('tile_y', [-1, 32]), ('tile_x', [-1, 32])],None,55
+ No: 3 GFLOPS: 13.21/13.21 result: MeasureResult(costs=(0.020318877800000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.523080587387085, timestamp=1665984681.3709538) [('tile_y', [-1, 16]), ('tile_x', [-1, 512])],None,94
+ No: 4 GFLOPS: 10.27/13.21 result: MeasureResult(costs=(0.0261388228,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6143655776977539, timestamp=1665984682.7211378) [('tile_y', [-1, 512]), ('tile_x', [-1, 128])],None,79
+ No: 5 GFLOPS: 2.40/13.21 result: MeasureResult(costs=(0.11170911559999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9347271919250488, timestamp=1665984684.802368) [('tile_y', [-1, 2]), ('tile_x', [-1, 4])],None,21
+ No: 6 GFLOPS: 13.64/13.64 result: MeasureResult(costs=(0.0196755144,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4885129928588867, timestamp=1665984686.0404818) [('tile_y', [-1, 128]), ('tile_x', [-1, 64])],None,67
+ No: 7 GFLOPS: 1.90/13.64 result: MeasureResult(costs=(0.14158192939999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.402756690979004, timestamp=1665984688.469896) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
+ No: 8 GFLOPS: 12.89/13.64 result: MeasureResult(costs=(0.0208255758,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5428996086120605, timestamp=1665984689.0181139) [('tile_y', [-1, 64]), ('tile_x', [-1, 512])],None,96
+ No: 9 GFLOPS: 2.56/13.64 result: MeasureResult(costs=(0.10471223059999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.789703130722046, timestamp=1665984690.9210284) [('tile_y', [-1, 512]), ('tile_x', [-1, 8])],None,39
+ No: 10 GFLOPS: 10.11/13.64 result: MeasureResult(costs=(0.0265516592,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5466616153717041, timestamp=1665984691.5209844) [('tile_y', [-1, 512]), ('tile_x', [-1, 256])],None,89
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 76dd66f68d..47c23a0a41 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -320,7 +320,7 @@ standard deviation.
.. code-block:: none
- {'mean': 518.3671750400208, 'median': 517.828456100051, 'std': 2.9975491441980844}
+ {'mean': 511.8934048999881, 'median': 511.7396063499882, 'std': 1.0176497639789417}
@@ -554,29 +554,31 @@ 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: 9.07/ 17.57 GFLOPS | Progress: (4/20) | 8.04 s
[Task 1/25] Current/Best: 12.86/ 23.47 GFLOPS | Progress: (8/20) | 11.83 s
[Task 1/25] Current/Best: 23.78/ 23.78 GFLOPS | Progress: (12/20) | 13.63 s
[Task 1/25] Current/Best: 15.05/ 23.78 GFLOPS | Progress: (16/20) | 17.06 s
[Task 1/25] Current/Best: 6.41/ 23.78 GFLOPS | Progress: (20/20) | 20.19 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 15.84/ 18.10 GFLOPS | Progress: (4/20) | 2.60 s
[Task 2/25] Current/Best: 8.61/ 18.10 GFLOPS | Progress: (8/20) | 3.90 s
[Task 2/25] Current/Best: 17.14/ 19.17 GFLOPS | Progress: (12/20) | 5.15 s
[Task 2/25] Current/Best: 11.86/ 19.17 GFLOPS | Progress: (16/20) | 6.46 s
[Task 2/25] Current/Best: 17.24/ 19.17 GFLOPS | Progress: (20/20) | 8.25 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 16.99/ 16.99 GFLOPS | Progress: (4/20) | 3.66 s
[Task 3/25] Current/Best: 5.66/ 16.99 GFLOPS | Progress: (8/20) | 6.30 s
[Task 3/25] Current/Best: 23.27/ 23.27 GFLOPS | Progress: (12/20) | 7.85 s
[Task 3/25] Current/Best: 14.03/ 23.27 GFLOPS | Progress: (16/20) | 10.10 s
[Task 3/25] Current/Best: 18.69/ 23.27 GFLOPS | Progress: (20/20) | 12.43 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 11.49/ 14.58 GFLOPS | Progress: (4/20) | 3.55 s
[Task 4/25] Current/Best: 11.49/ 14.58 GFLOPS | Progress: (8/20) | 6.49 s
[Task 4/25] Current/Best: 7.27/ 18.33 GFLOPS | Progress: (12/20) | 8.02 s
[Task 4/25] Current/Best: 10.34/ 18.33 GFLOPS | Progress: (16/20) | 12.41 s
[Task 4/25] Current/Best: 16.60/ 21.40 GFLOPS | Progress: (20/20) | 13.92 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 13.76/ 18.08 GFLOPS | Progress: (4/20) | 3.67 s
[Task 5/25] Current/Best: 19.43/ 20.97 GFLOPS | Progress: (8/20) | 5.28 s
[Task 5/25] Current/Best: 13.26/ 20.97 GFLOPS | Progress: (12/20) | 6.80 s
[Task 5/25] Current/Best: 4.10/ 20.97 GFLOPS | Progress: (16/20) | 8.32 s
[Task 5/25] Current/Best: 14.75/ 20.97 GFLOPS | Progress: (20/20) | 10.44 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 20.72/ 20.72 GFLOPS | Progress: (4/20) | 4.06 s
[Task 6/25] Current/Best: 4.28/ 20.72 GFLOPS | Progress: (8/20) | 7.85 s
[Task 6/25] Current/Best: 15.16/ 20.72 GFLOPS | Progress: (12/20) | 10.09 s
[Task 6/25] Current/Best: 5.14/ 20.72 GFLOPS | Progress: (16/20) | 12.22 s
[Task 6/25] Current/Best: 9.76/ 20.72 GFLOPS | Progress: (20/20) | 14.33 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 15.58/ 17.69 GFLOPS | Progress: (4/20) | 3.81 s
[Task 7/25] Current/Best: 11.45/ 17.69 GFLOPS | Progress: (8/20) | 5.75 s
[Task 7/25] Current/Best: 18.91/ 18.91 GFLOPS | Progress: (12/20) | 7.99 s
[Task 7/25] Current/Best: 14.13/ 18.91 GFLOPS | Progress: (16/20) | 10.91 s
[Task 7/25] Current/Best: 9.41/ 18.91 GFLOPS | Progress: (20/20) | 13.72 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 2.86/ 12.06 GFLOPS | Progress: (4/20) | 4.23 s
[Task 8/25] Current/Best: 12.25/ 16.97 GFLOPS | Progress: (8/20) | 7.26 s
[Task 8/25] Current/Best: 11.16/ 16.97 GFLOPS | Progress: (12/20) | 9.73 s
[Task 8/25] Current/Best: 10.36/ 18.02 GFLOPS | Progress: (16/20) | 21.28 s
[Task 8/25] Current/Best: 8.01/ 18.02 GFLOPS | Progress: (20/20) | 24.69 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 20.39/ 20.39 GFLOPS | Progress: (4/20) | 3.89 s
[Task 9/25] Current/Best: 14.50/ 20.39 GFLOPS | Progress: (8/20) | 5.70 s
[Task 9/25] Current/Best: 23.01/ 23.01 GFLOPS | Progress: (12/20) | 7.09 s
[Task 9/25] Current/Best: 8.22/ 23.01 GFLOPS | Progress: (16/20) | 12.15 s
[Task 9/25] Current/Best: 12.37/ 23.01 GFLOPS | Progress: (20/20) | 23.02 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 14.63/ 19.33 GFLOPS | Progress: (4/20) | 3.79 s
[Task 10/25] Current/Best: 11.28/ 19.33 GFLOPS | Progress: (8/20) | 6.07 s
[Task 10/25] Current/Best: 8.40/ 19.33 GFLOPS | Progress: (12/20) | 7.73 s
[Task 10/25] Current/Best: 13.95/ 19.33 GFLOPS | Progress: (16/20) | 9.93 s
[Task 10/25] Current/Best: 10.27/ 19.33 GFLOPS | Progress: (20/20) |
11.53 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 17.86/ 21.87 GFLOPS | Progress: (4/20) | 3.08 s
[Task 11/25] Current/Best: 7.19/ 21.87 GFLOPS | Progress: (8/20) | 5.74 s
[Task 11/25] Current/Best: 15.49/ 21.87 GFLOPS | Progress: (12/20) | 7.65 s
[Task 11/25] Current/Best: 10.54/ 21.87 GFLOPS | Progress: (16/20) | 11.07 s
[Task 11/25] Current/Best: 6.20/ 21.87 GFLOPS | Progress: (20/20) | 13.61 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 9.71/ 21.17 GFLOPS | Progress: (4/20) | 3.26 s
[Task 12/25] Current/Best: 9.38/ 21.17 GFLOPS | Progress: (8/20) | 11.65 s
[Task 12/25] Current/Best: 13.96/ 21.17 GFLOPS | Progress: (12/20) | 16.27 s
[Task 12/25] Current/Best: 15.31/ 21.17 GFLOPS | Progress: (16/20) | 21.34 s
[Task 12/25] Current/Best: 16.24/ 21.17 GFLOPS | Progress: (20/20) | 23.88 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 10.15/ 16.15 GFLOPS | Progress: (4/20) | 4.56 s
[Task 13/25] Current/Best: 9.86/ 16.43 GFLOPS | Progress: (8/20) | 7.07 s
[Task 13/25] Current/Best: 13.31/ 16.43 GFLOPS | Progress: (12/20) | 9.33 s
[Task 13/25] Current/Best: 6.04/ 20.09 GFLOPS | Progress: (16/20) | 11.54 s
[Task 13/25] Current/Best: 13.17/ 20.09 GFLOPS | Progress: (20/20) | 14.91 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 16.77/ 18.10 GFLOPS | Progress: (4/20) | 6.67 s
[Task 14/25] Current/Best: 15.83/ 18.10 GFLOPS | Progress: (8/20) | 9.19 s
[Task 14/25] Current/Best: 17.46/ 20.57 GFLOPS | Progress: (12/20) | 10.70 s
[Task 14/25] Current/Best: 19.13/ 20.57 GFLOPS | Progress: (16/20) | 12.39 s
[Task 14/25] Current/Best: 9.28/ 20.57 GFLOPS | Progress: (20/20) | 15.27 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 14.10/ 17.41 GFLOPS | Progress: (4/20) | 7.27 s
[Task 1/25] Current/Best: 7.69/ 22.43 GFLOPS | Progress: (8/20) | 10.55 s
[Task 1/25] Current/Best: 14.96/ 22.43 GFLOPS | Progress: (12/20) | 12.63 s
[Task 1/25] Current/Best: 15.64/ 23.31 GFLOPS | Progress: (16/20) | 14.59 s
[Task 1/25] Current/Best: 7.98/ 23.31 GFLOPS | Progress: (20/20) | 17.53 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 3.68/ 13.88 GFLOPS | Progress: (4/20) | 3.44 s
[Task 2/25] Current/Best: 7.83/ 15.05 GFLOPS | Progress: (8/20) | 4.50 s
[Task 2/25] Current/Best: 15.18/ 17.01 GFLOPS | Progress: (12/20) | 5.82 s
[Task 2/25] Current/Best: 6.51/ 18.11 GFLOPS | Progress: (16/20) | 7.39 s
[Task 2/25] Current/Best: 6.02/ 18.21 GFLOPS | Progress: (20/20) | 8.68 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 15.60/ 15.60 GFLOPS | Progress: (4/20) | 3.65 s
[Task 3/25] Current/Best: 18.90/ 18.90 GFLOPS | Progress: (8/20) | 5.36 s
[Task 3/25] Current/Best: 6.98/ 20.11 GFLOPS | Progress: (12/20) | 7.40 s
[Task 3/25] Current/Best: 17.24/ 20.11 GFLOPS | Progress: (16/20) | 9.46 s
[Task 3/25] Current/Best: 14.42/ 20.11 GFLOPS | Progress: (20/20) | 12.14 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 12.30/ 17.93 GFLOPS | Progress: (4/20) | 3.40 s
[Task 4/25] Current/Best: 17.22/ 17.93 GFLOPS | Progress: (8/20) | 5.70 s
[Task 4/25] Current/Best: 18.57/ 18.57 GFLOPS | Progress: (12/20) | 7.29 s
[Task 4/25] Current/Best: 9.98/ 20.08 GFLOPS | Progress: (16/20) | 11.18 s
[Task 4/25] Current/Best: 5.61/ 23.13 GFLOPS | Progress: (20/20) | 15.92 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 15.91/ 20.47 GFLOPS | Progress: (4/20) | 3.00 s
[Task 5/25] Current/Best: 11.96/ 20.47 GFLOPS | Progress: (8/20) | 4.86 s
[Task 5/25] Current/Best: 16.50/ 20.89 GFLOPS | Progress: (12/20) | 6.96 s
[Task 5/25] Current/Best: 5.30/ 20.89 GFLOPS | Progress: (16/20) | 8.93 s
[Task 5/25] Current/Best: 9.05/ 20.89 GFLOPS | Progress: (20/20) | 10.73 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 11.83/ 13.48 GFLOPS | Progress: (4/20) | 5.35 s
[Task 6/25] Current/Best: 15.00/ 18.33 GFLOPS | Progress: (8/20) | 8.67 s
[Task 6/25] Current/Best: 15.97/ 18.33 GFLOPS | Progress: (12/20) | 10.44 s
[Task 6/25] Current/Best: 11.45/ 18.49 GFLOPS | Progress: (16/20) | 12.97 s
[Task 6/25] Current/Best: 18.02/ 19.59 GFLOPS | Progress: (20/20) | 14.63 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 21.95/ 21.95 GFLOPS | Progress: (4/20) | 4.17 s
[Task 7/25] Current/Best: 18.45/ 21.95 GFLOPS | Progress: (8/20) | 6.16 s
[Task 7/25] Current/Best: 11.45/ 21.95 GFLOPS | Progress: (12/20) | 8.43 s
[Task 7/25] Current/Best: 18.76/ 21.95 GFLOPS | Progress: (16/20) | 10.76 s
[Task 7/25] Current/Best: 5.33/ 21.95 GFLOPS | Progress: (20/20) | 12.79 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 14.20/ 14.20 GFLOPS | Progress: (4/20) | 8.05 s
[Task 8/25] Current/Best: 11.79/ 17.81 GFLOPS | Progress: (8/20) | 10.36 s
[Task 8/25] Current/Best: 9.91/ 17.81 GFLOPS | Progress: (12/20) | 15.23 s
[Task 8/25] Current/Best: 16.16/ 17.81 GFLOPS | Progress: (16/20) | 17.24 s
[Task 8/25] Current/Best: 9.88/ 17.81 GFLOPS | Progress: (20/20) | 22.88 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.50/ 14.50 GFLOPS | Progress: (4/20) | 2.97 s
[Task 9/25] Current/Best: 9.24/ 20.01 GFLOPS | Progress: (8/20) | 6.24 s
[Task 9/25] Current/Best: 4.80/ 20.01 GFLOPS | Progress: (12/20) | 8.88 s
[Task 9/25] Current/Best: 20.88/ 20.88 GFLOPS | Progress: (16/20) | 14.58 s
[Task 9/25] Current/Best: 11.43/ 20.88 GFLOPS | Progress: (20/20) | 18.42 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 17.94/ 18.57 GFLOPS | Progress: (4/20) | 3.15 s
[Task 10/25] Current/Best: 13.08/ 18.57 GFLOPS | Progress: (8/20) | 5.44 s
[Task 10/25] Current/Best: 10.45/ 18.57 GFLOPS | Progress: (12/20) | 7.32 s
[Task 10/25] Current/Best: 12.38/ 18.57 GFLOPS | Progress: (16/20) | 8.96 s
[Task 10/25] Current/Best: 13.14/ 18.64 GFLOPS | Progress: (20/20) | 10.58 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 6.26/ 18.32 GFLOPS | Progress: (4/20) | 3.61 s
[Task 11/25] Current/Best: 12.81/ 21.14 GFLOPS | Progress: (8/20) | 5.70 s
[Task 11/25] Current/Best: 16.05/ 21.14 GFLOPS | Progress: (12/20) | 7.73 s
[Task 11/25] Current/Best: 10.68/ 21.14 GFLOPS | Progress: (16/20) | 10.65 s
[Task 11/25] Current/Best: 17.32/ 21.14 GFLOPS | Progress: (20/20) | 12.92 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 1.58/ 9.21 GFLOPS | Progress: (4/20) | 5.40 s
[Task 12/25] Current/Best: 13.88/ 14.52 GFLOPS | Progress: (8/20) | 11.87 s
[Task 12/25] Current/Best: 14.33/ 19.42 GFLOPS | Progress: (12/20) | 13.91 s
[Task 12/25] Current/Best: 11.35/ 19.42 GFLOPS | Progress: (16/20) | 18.36 s
[Task 12/25] Current/Best: 6.22/ 19.42 GFLOPS | Progress: (20/20) | 20.51 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 10.58/ 20.67 GFLOPS | Progress: (4/20) | 3.82 s
[Task 13/25] Current/Best: 18.54/ 21.49 GFLOPS | Progress: (8/20) | 5.79 s
[Task 13/25] Current/Best: 17.39/ 21.49 GFLOPS | Progress: (12/20) | 8.66 s
[Task 13/25] Current/Best: 13.83/ 21.49 GFLOPS | Progress: (16/20) | 11.49 s
[Task 13/25] Current/Best: 17.33/ 21.49 GFLOPS | Progress: (20/20) | 14.82 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 17.81/ 17.81 GFLOPS | Progress: (4/20) | 3.54 s
[Task 14/25] Current/Best: 13.06/ 17.81 GFLOPS | Progress: (8/20) | 5.76 s
[Task 14/25] Current/Best: 14.57/ 17.81 GFLOPS | Progress: (12/20) | 8.17 s
[Task 14/25] Current/Best: 1.59/ 17.81 GFLOPS | Progress: (16/20) | 12.66 s
[Task 14/25] Current/Best: 15.06/ 17.81 GFLOPS | Progress: (20/20) | 14.28 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 14.04/ 19.50 GFLOPS | Progress: (4/20) | 3.46 s
[Task 15/25] Current/Best: 9.59/ 19.50 GFLOPS | Progress: (8/20) | 5.87 s Done.
+
[Task 15/25] Current/Best: 16.73/ 19.50 GFLOPS | Progress: (12/20) | 7.46 s
[Task 15/25] Current/Best: 17.43/ 19.50 GFLOPS | Progress: (16/20) | 8.93 s
[Task 15/25] Current/Best: 11.77/ 19.50 GFLOPS | Progress: (20/20) | 10.78 s Done.
+
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 12.68/ 17.53 GFLOPS | Progress: (4/20) | 4.53 s
[Task 16/25] Current/Best: 7.15/ 17.58 GFLOPS | Progress: (8/20) | 6.35 s
[Task 16/25] Current/Best: 14.33/ 17.58 GFLOPS | Progress: (12/20) | 9.17 s
[Task 16/25] Current/Best: 3.09/ 18.09 GFLOPS | Progress: (16/20) | 12.04 s
[Task 16/25] Current/Best: 17.52/ 18.09 GFLOPS | Progress: (20/20) | 14.03 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 16.65/ 18.66 GFLOPS | Progress: (4/20) | 3.92 s
[Task 17/25] Current/Best: 16.56/ 20.95 GFLOPS | Progress: (8/20) | 5.70 s
[Task 17/25] Current/Best: 11.63/ 20.95 GFLOPS | Progress: (12/20) | 8.75 s
[Task 17/25] Current/Best: 6.20/ 21.52 GFLOPS | Progress: (16/20) | 11.04 s
[Task 17/25] Current/Best: 6.15/ 21.52 GFLOPS | Progress: (20/20) | 13.43 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 9.22/ 20.76 GFLOPS | Progress: (4/20) | 7.61 s
[Task 18/25] Current/Best: 11.62/ 20.76 GFLOPS | Progress: (8/20) | 10.48 s
[Task 18/25] Current/Best: 12.30/ 20.76 GFLOPS | Progress: (12/20) | 13.03 s
[Task 18/25] Current/Best: 14.28/ 21.51 GFLOPS | Progress: (16/20) | 16.07 s
[Task 18/25] Current/Best: 9.23/ 21.51 GFLOPS | Progress: (20/20) | 18.84 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 21.35/ 21.35 GFLOPS | Progress: (4/20) | 4.72 s
[Task 19/25] Current/Best: 6.16/ 21.35 GFLOPS | Progress: (8/20) | 8.07 s
[Task 19/25] Current/Best: 19.52/ 21.41 GFLOPS | Progress: (12/20) | 10.55 s
[Task 19/25] Current/Best: 10.41/ 21.72 GFLOPS | Progress: (16/20) | 13.71 s
[Task 19/25] Current/Best: 10.40/ 21.72 GFLOPS | Progress: (20/20) | 17.14 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 4.88/ 12.82 GFLOPS | Progress: (4/20) | 3.66 s
[Task 20/25] Current/Best: 10.83/ 18.68 GFLOPS | Progress: (8/20) | 8.03 s
[Task 20/25] Current/Best: 6.18/ 18.68 GFLOPS | Progress: (12/20) | 10.72 s
[Task 20/25] Current/Best: 7.56/ 18.68 GFLOPS | Progress: (16/20) | 12.88 s
[Task 20/25] Current/Best: 12.80/ 18.68 GFLOPS | Progress: (20/20) | 16.70 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 14.86/ 14.86 GFLOPS | Progress: (4/20) | 2.96 s
[Task 21/25] Current/Best: 6.81/ 14.86 GFLOPS | Progress: (8/20) | 6.07 s
[Task 21/25] Current/Best: 10.86/ 14.86 GFLOPS | Progress: (12/20) | 8.69 s
[Task 21/25] Current/Best: 16.26/ 20.58 GFLOPS | Progress: (16/20) | 10.64 s
[Task 21/25] Current/Best: 8.81/ 20.58 GFLOPS | Progress: (20/20)
| 12.47 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
Done.
-
[Task 15/25] Current/Best: 12.60/ 17.41 GFLOPS | Progress: (4/20) | 3.89 s
[Task 15/25] Current/Best: 5.25/ 18.43 GFLOPS | Progress: (8/20) | 5.99 s
[Task 15/25] Current/Best: 16.36/ 18.43 GFLOPS | Progress: (12/20) | 7.35 s
[Task 15/25] Current/Best: 13.59/ 18.43 GFLOPS | Progress: (16/20) | 9.65 s
[Task 15/25] Current/Best: 7.55/ 18.43 GFLOPS | Progress: (20/20) | 13.12 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 19.06/ 19.06 GFLOPS | Progress: (4/20) | 4.11 s
[Task 16/25] Current/Best: 8.42/ 20.44 GFLOPS | Progress: (8/20) | 5.67 s
[Task 16/25] Current/Best: 13.55/ 20.44 GFLOPS | Progress: (12/20) | 7.25 s
[Task 16/25] Current/Best: 10.07/ 20.44 GFLOPS | Progress: (16/20) | 8.76 s
[Task 16/25] Current/Best: 13.19/ 20.44 GFLOPS | Progress: (20/20) | 10.81 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 16.53/ 22.10 GFLOPS | Progress: (4/20) | 3.24 s
[Task 17/25] Current/Best: 12.54/ 22.10 GFLOPS | Progress: (8/20) | 5.23 s
[Task 17/25] Current/Best: 18.24/ 22.10 GFLOPS | Progress: (12/20) | 7.24 s
[Task 17/25] Current/Best: 8.00/ 22.10 GFLOPS | Progress: (16/20) | 10.46 s
[Task 17/25] Current/Best: 1.55/ 22.10 GFLOPS | Progress: (20/20) | 13.69 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 15.57/ 17.94 GFLOPS | Progress: (4/20) | 3.50 s
[Task 18/25] Current/Best: 14.80/ 17.94 GFLOPS | Progress: (8/20) | 7.38 s
[Task 18/25] Current/Best: 8.21/ 17.94 GFLOPS | Progress: (12/20) | 10.15 s
[Task 18/25] Current/Best: 18.31/ 20.57 GFLOPS | Progress: (16/20) | 11.96 s
[Task 18/25] Current/Best: 17.94/ 20.57 GFLOPS | Progress: (20/20) | 13.52 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 3.09/ 21.69 GFLOPS | Progress: (4/20) | 4.00 s
[Task 19/25] Current/Best: 21.75/ 21.75 GFLOPS | Progress: (8/20) | 9.02 s
[Task 19/25] Current/Best: 19.25/ 21.75 GFLOPS | Progress: (12/20) | 11.34 s
[Task 19/25] Current/Best: 10.59/ 21.75 GFLOPS | Progress: (16/20) | 16.70 s
[Task 19/25] Current/Best: 17.64/ 21.75 GFLOPS | Progress: (20/20) | 18.76 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 21.43/ 21.43 GFLOPS | Progress: (4/20) | 3.22 s
[Task 20/25] Current/Best: 9.86/ 21.43 GFLOPS | Progress: (8/20) | 7.81 s
[Task 20/25] Current/Best: 11.38/ 21.43 GFLOPS | Progress: (12/20) | 9.52 s
[Task 20/25] Current/Best: 10.27/ 21.43 GFLOPS | Progress: (16/20) | 12.48 s
[Task 20/25] Current/Best: 5.11/ 21.43 GFLOPS | Progress: (20/20) | 14.95 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 20.91/ 20.91 GFLOPS | Progress: (4/20) | 3.43 s
[Task 21/25] Current/Best: 9.75/ 20.91 GFLOPS | Progress: (8/20) | 6.20 s
[Task 21/25] Current/Best: 2.71/ 20.91 GFLOPS | Progress: (12/20) | 9.33 s
[Task 21/25] Current/Best: 7.21/ 20.91 GFLOPS | Progress: (16/20) | 10.86 s
[Task 21/25] Current/Best: 18.05/ 20.91 GFLOPS | Progress: (20/20)
| 13.17 s Done.
-
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 16.45/ 16.45 GFLOPS | Progress: (4/20) | 4.05 s
[Task 22/25] Current/Best: 18.43/ 18.43 GFLOPS | Progress: (8/20) | 5.63 s
[Task 22/25] Current/Best: 6.79/ 18.43 GFLOPS | Progress: (12/20) | 8.33 s
[Task 22/25] Current/Best: 9.46/ 18.43 GFLOPS | Progress: (16/20) | 10.76 s
[Task 22/25] Current/Best: 16.88/ 18.43 GFLOPS | Progress: (20/20) | 13.27 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 10.75/ 11.02 GFLOPS | Progress: (4/20) | 5.23 s
[Task 23/25] Current/Best: 12.30/ 20.03 GFLOPS | Progress: (8/20) | 7.68 s
[Task 23/25] Current/Best: 9.03/ 21.60 GFLOPS | Progress: (12/20) | 10.58 s
[Task 23/25] Current/Best: 18.08/ 21.60 GFLOPS | Progress: (16/20) | 16.07 s
[Task 23/25] Current/Best: 7.88/ 21.60 GFLOPS | Progress: (20/20) | 19.53 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 7.40/ 8.72 GFLOPS | Progress: (4/20) | 12.05 s
[Task 24/25] Current/Best: 5.34/ 8.72 GFLOPS | Progress: (8/20) | 23.47 s
[Task 24/25] Current/Best: 2.99/ 8.72 GFLOPS | Progress: (12/20) | 31.09 s
[Task 24/25] Current/Best: 5.50/ 8.72 GFLOPS | Progress: (16/20) | 42.37 s
[Task 24/25] Current/Best: 6.63/ 8.72 GFLOPS | Progress: (20/20) | 53.09 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
[Task 25/25] Current/Best: 8.35/ 8.62 GFLOPS | Progress: (4/20) | 13.09 s
[Task 25/25] Current/Best: 5.89/ 8.64 GFLOPS | Progress: (8/20) | 23.82 s
[Task 25/25] Current/Best: 9.28/ 9.44 GFLOPS | Progress: (12/20) | 28.58 s
[Task 25/25] Current/Best: 5.65/ 9.44 GFLOPS | Progress: (16/20) | 39.33 s
[Task 25/25] Current/Best: 5.85/ 9.44 GFLOPS | Progress: (20/20) | 50.65 s
+
[Task 22/25] Current/Best: 12.58/ 17.49 GFLOPS | Progress: (4/20) | 4.09 s
[Task 22/25] Current/Best: 2.75/ 17.49 GFLOPS | Progress: (8/20) | 8.29 s
[Task 22/25] Current/Best: 11.05/ 17.49 GFLOPS | Progress: (12/20) | 10.39 s
[Task 22/25] Current/Best: 16.97/ 18.52 GFLOPS | Progress: (16/20) | 12.24 s
[Task 22/25] Current/Best: 5.16/ 18.52 GFLOPS | Progress: (20/20) | 13.99 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 19.62/ 19.62 GFLOPS | Progress: (4/20) | 4.29 s
[Task 23/25] Current/Best: 20.49/ 20.49 GFLOPS | Progress: (8/20) | 6.93 s
[Task 23/25] Current/Best: 12.00/ 20.49 GFLOPS | Progress: (12/20) | 9.31 s
[Task 23/25] Current/Best: 11.68/ 20.49 GFLOPS | Progress: (16/20) | 12.07 s
[Task 23/25] Current/Best: 10.46/ 20.49 GFLOPS | Progress: (20/20) | 14.61 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 3.65/ 8.32 GFLOPS | Progress: (4/20) | 11.97 s
[Task 24/25] Current/Best: 0.77/ 8.32 GFLOPS | Progress: (8/20) | 23.47 s
[Task 24/25] Current/Best: 4.62/ 8.92 GFLOPS | Progress: (12/20) | 35.37 s
[Task 24/25] Current/Best: 1.98/ 8.92 GFLOPS | Progress: (16/20) | 46.10 s
[Task 24/25] Current/Best: 1.76/ 8.92 GFLOPS | Progress: (20/20) | 57.83 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
[Task 25/25] Current/Best: 8.97/ 8.97 GFLOPS | Progress: (4/20) | 13.16 s
[Task 25/25] Current/Best: 9.04/ 9.04 GFLOPS | Progress: (8/20) | 15.13 s
[Task 25/25] Current/Best: 8.32/ 9.04 GFLOPS | Progress: (12/20) | 25.87 s
[Task 25/25] Current/Best: 1.54/ 9.04 GFLOPS | Progress: (16/20) | 36.61 s
[Task 25/25] Current/Best: 7.53/ 9.04 GFLOPS | Progress: (20/20) | 46.15 s
@@ -672,8 +674,8 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
- class='n02123045 tabby, tabby cat' with probability=0.621102
- class='n02123159 tiger cat' with probability=0.356380
+ class='n02123045 tabby, tabby cat' with probability=0.621105
+ class='n02123159 tiger cat' with probability=0.356378
class='n02124075 Egyptian cat' with probability=0.019712
class='n02129604 tiger, Panthera tigris' with probability=0.001215
class='n04040759 radiator' with probability=0.000262
@@ -730,8 +732,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 407.7192681000088, 'median': 407.82887029999983, 'std': 0.7552311734284762}
- unoptimized: {'mean': 518.3671750400208, 'median': 517.828456100051, 'std': 2.9975491441980844}
+ optimized: {'mean': 430.4673170099977, 'median': 430.19669249997605, 'std': 1.846446171792923}
+ unoptimized: {'mean': 511.8934048999881, 'median': 511.7396063499882, 'std': 1.0176497639789417}
@@ -754,7 +756,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 11 minutes 4.455 seconds)
+ **Total running time of the script:** ( 10 minutes 52.949 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 00f1926763..0f81a1e365 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -282,7 +282,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.244e-07 secs/op
+ 1.279e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index c746ad6792..fe5f2c9815 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -263,7 +263,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0xde45c80)), stage(b, placeholder(b, 0x5d55f00)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min= [...]
+ [stage(a, placeholder(a, 0x22d938f0)), stage(b, placeholder(b, 0x18059b70)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(mi [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index 3dd9c343cc..03a4edddb2 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,32 +5,32 @@
Computation times
=================
-**14:26.874** total execution time for **tutorial** files:
+**14:03.867** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 11:04.455 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 10:52.949 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:18.373 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:13.803 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:01.872 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:02.306 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:35.901 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:35.776 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:24.501 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:17.603 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.853 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.733 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.728 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.527 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.182 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.162 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.006 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.005 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``) | 00:00.001 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_install.py` (``install.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_install.py` (``install.py``) | 00:00.001 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index bf9c1169e5..48d14e4458 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -294,7 +294,7 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
- Numpy running time: 0.000008
+ Numpy running time: 0.000007
naive: 0.000007
@@ -501,10 +501,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 7.515449997299584e-06 1.0
- naive 6.6426000000000006e-06 0.8838592502626973
- parallel 6.91e-06 0.9194392887295996
- vector 2.4587900000000002e-05 3.2716470748704083
+ numpy 7.395829998131376e-06 1.0
+ naive 6.6719e-06 0.9021164631536577
+ parallel 6.9341e-06 0.9375688735073635
+ vector 2.4630399999999997e-05 3.3303091074596214
@@ -925,7 +925,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.019000
+ Numpy running time: 0.018789
@@ -983,7 +983,7 @@ optimizations.
.. code-block:: none
- none: 3.465411
+ none: 3.488838
@@ -1086,7 +1086,7 @@ schedule.
.. code-block:: none
- blocking: 0.304412
+ blocking: 0.317260
@@ -1182,7 +1182,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.342519
+ vectorization: 0.342259
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1256,7 +1256,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.116870
+ loop permutation: 0.117839
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1355,7 +1355,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.108036
+ array packing: 0.108581
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1448,7 +1448,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.110349
+ block caching: 0.110871
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1534,7 +1534,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.146668
+ parallelization: 0.146338
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1615,13 +1615,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.4654111111 1.0
- blocking 0.3044119027 0.08784294069033928
- vectorization 0.3425189822 0.0988393501431571
- loop permutation 0.11686987280000001 0.03372467769427301
- array packing 0.108036249 0.031175593756813127
- block caching 0.11034887450000001 0.03184293896517599
- parallelization 0.1466683301 0.04232350084819926
+ none 3.4888384268 1.0
+ blocking 0.3172602255 0.09093577480198602
+ vectorization 0.34225855139999994 0.09810100369535403
+ loop permutation 0.11783892359999999 0.033775976180153204
+ array packing 0.1085813028 0.03112247960980868
+ block caching 0.11087110470000001 0.03177880174912318
+ parallelization 0.1463375169 0.041944480941246205
@@ -1663,7 +1663,7 @@ the computation for specific platforms.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 1.872 seconds)
+ **Total running time of the script:** ( 1 minutes 2.306 seconds)
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 8a272627b0..337073797d 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-9299a29fd339b6b728487b32c0526f02a0181e3b
+cbca28da003cf4791a19f0c4f9e0c930ecee522d
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index 1a435dd3c0..7c8e5f508a 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -577,7 +577,7 @@ class:['truck 0.9266'] left:471 top:83 right:689 bottom:169
class:['bicycle 0.9984'] left:111 top:113 right:577 bottom:447
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 12.004 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 12.318 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 f2cf6d22b2..9492abe70c 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -498,7 +498,7 @@ pip install -U tensorflow --user
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Relay top-1 id: 285, class name: Egyptian cat
1/1 [==============================] - ETA: 0s
-1/1 [==============================] - 1s 992ms/step
+1/1 [==============================] - 1s 981ms/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 5757f05aa8..08060b5c26 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -432,7 +432,7 @@ to download the full example code</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip3988ca5b-e686-486a-b79f-2c757f8bb17e 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.zipadd651b9-c7d9-4b22-a086-3c17190aed2b 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 daa9075222..1efecc7ce0 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -440,13 +440,12 @@ 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, 48.7MB/s]
- 39%|###8 | 16.0M/41.5M [00:00<00:00, 56.1MB/s]
- 57%|#####7 | 23.7M/41.5M [00:00<00:00, 64.7MB/s]
- 73%|#######2 | 30.1M/41.5M [00:00<00:00, 56.6MB/s]
- 86%|########6 | 35.7M/41.5M [00:00<00:00, 54.8MB/s]
- 99%|#########9| 41.1M/41.5M [00:00<00:00, 52.3MB/s]
-100%|##########| 41.5M/41.5M [00:00<00:00, 55.0MB/s]
+ 19%|#9 | 7.99M/41.5M [00:00<00:00, 79.5MB/s]
+ 38%|###7 | 15.6M/41.5M [00:00<00:00, 78.6MB/s]
+ 56%|#####5 | 23.1M/41.5M [00:00<00:00, 65.0MB/s]
+ 71%|#######1 | 29.5M/41.5M [00:00<00:00, 65.5MB/s]
+ 86%|########6 | 35.9M/41.5M [00:00<00:00, 61.4MB/s]
+100%|##########| 41.5M/41.5M [00:00<00:00, 61.9MB/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 77d34328cd..f07121896d 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -419,10 +419,12 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
- 23%|##2 | 10.1M/44.7M [00:00<00:00, 101MB/s]
- 54%|#####3 | 24.0M/44.7M [00:00<00:00, 108MB/s]
- 77%|#######6 | 34.2M/44.7M [00:00<00:00, 87.5MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 100MB/s]
+ 18%|#7 | 7.99M/44.7M [00:00<00:00, 63.5MB/s]
+ 36%|###5 | 16.0M/44.7M [00:00<00:00, 69.2MB/s]
+ 54%|#####3 | 24.0M/44.7M [00:00<00:00, 72.5MB/s]
+ 72%|#######1 | 32.0M/44.7M [00:00<00:00, 75.1MB/s]
+ 90%|########9 | 40.0M/44.7M [00:00<00:00, 75.2MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 79.0MB/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 02ee8ebeed..63fcd45c98 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -637,7 +637,7 @@ 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 15.522 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 14.190 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 805b10c9fd..25d70157d1 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -332,7 +332,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:53.767</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:50.252</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -341,43 +341,43 @@
</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>01:15.522</p></td>
+<td><p>01:14.190</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>01:12.004</p></td>
+<td><p>01:12.318</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>00:47.158</p></td>
+<td><p>00:47.247</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>00:33.509</p></td>
+<td><p>00:32.367</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></td>
-<td><p>00:30.271</p></td>
+<td><p>00:29.760</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></td>
-<td><p>00:26.577</p></td>
+<td><p>00:26.458</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></td>
-<td><p>00:25.032</p></td>
+<td><p>00:25.446</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></td>
-<td><p>00:22.601</p></td>
+<td><p>00:22.051</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>00:18.727</p></td>
+<td><p>00:17.994</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>00:02.365</p></td>
+<td><p>00:02.420</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index d57acc55e4..3a192bf7b0 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -654,7 +654,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.0638 16.0509 16.1987 15.8900 0.1052
+ 16.4224 16.2221 17.1125 16.1277 0.3585
</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 174d0b9495..58cd74cea3 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -441,23 +441,28 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 5%|4 | 7.99M/170M [00:00<00:02, 67.1MB/s]
- 14%|#4 | 24.0M/170M [00:00<00:01, 101MB/s]
- 22%|##2 | 38.0M/170M [00:00<00:01, 119MB/s]
- 29%|##9 | 49.5M/170M [00:00<00:01, 95.1MB/s]
- 35%|###4 | 59.2M/170M [00:00<00:01, 92.8MB/s]
- 40%|#### | 68.3M/170M [00:00<00:01, 81.2MB/s]
- 47%|####7 | 80.0M/170M [00:00<00:01, 83.4MB/s]
- 52%|#####1 | 88.2M/170M [00:01<00:01, 73.1MB/s]
- 57%|#####6 | 96.0M/170M [00:01<00:01, 73.7MB/s]
- 62%|######1 | 105M/170M [00:01<00:00, 79.1MB/s]
- 66%|######6 | 113M/170M [00:01<00:00, 62.7MB/s]
- 75%|#######5 | 128M/170M [00:01<00:00, 77.4MB/s]
- 80%|######## | 136M/170M [00:01<00:00, 76.1MB/s]
- 86%|########5 | 145M/170M [00:01<00:00, 81.2MB/s]
- 90%|######### | 153M/170M [00:02<00:00, 76.3MB/s]
- 99%|#########8| 168M/170M [00:02<00:00, 91.7MB/s]
-100%|##########| 170M/170M [00:02<00:00, 83.7MB/s]
+ 5%|4 | 7.99M/170M [00:00<00:02, 82.7MB/s]
+ 9%|9 | 15.9M/170M [00:00<00:02, 78.2MB/s]
+ 14%|#4 | 24.0M/170M [00:00<00:02, 55.7MB/s]
+ 18%|#7 | 30.3M/170M [00:00<00:02, 57.4MB/s]
+ 21%|##1 | 36.2M/170M [00:00<00:02, 57.9MB/s]
+ 27%|##7 | 45.9M/170M [00:00<00:01, 70.9MB/s]
+ 33%|###2 | 56.0M/170M [00:00<00:01, 72.3MB/s]
+ 38%|###7 | 64.0M/170M [00:00<00:01, 69.9MB/s]
+ 42%|####2 | 72.0M/170M [00:01<00:01, 73.2MB/s]
+ 47%|####7 | 80.0M/170M [00:01<00:01, 68.2MB/s]
+ 52%|#####1 | 88.0M/170M [00:01<00:01, 59.3MB/s]
+ 57%|#####6 | 96.0M/170M [00:01<00:01, 64.8MB/s]
+ 61%|######1 | 104M/170M [00:01<00:01, 61.2MB/s]
+ 66%|######5 | 112M/170M [00:01<00:00, 65.6MB/s]
+ 71%|####### | 120M/170M [00:01<00:00, 68.9MB/s]
+ 75%|#######5 | 128M/170M [00:02<00:00, 68.9MB/s]
+ 80%|######## | 136M/170M [00:02<00:00, 66.6MB/s]
+ 85%|########4 | 144M/170M [00:02<00:00, 68.4MB/s]
+ 89%|########9 | 152M/170M [00:02<00:00, 66.3MB/s]
+ 94%|#########4| 160M/170M [00:02<00:00, 71.2MB/s]
+ 99%|#########8| 168M/170M [00:02<00:00, 74.1MB/s]
+100%|##########| 170M/170M [00:02<00:00, 68.1MB/s]
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3878: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode=& [...]
@@ -555,7 +560,7 @@ 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 16.844 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 20.980 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 fcea962b98..772cab97e1 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -485,8 +485,8 @@ training. Other models require a full post training calibration.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
0%| | 0.00/13.6M [00:00<?, ?B/s]
- 73%|#######3 | 9.91M/13.6M [00:00<00:00, 104MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 98.2MB/s]
+ 59%|#####8 | 7.99M/13.6M [00:00<00:00, 64.8MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 93.3MB/s]
</pre></div>
</div>
</div>
@@ -577,7 +577,7 @@ 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)
- 90.5351 90.3460 99.2452 90.2206 1.0205
+ 90.7599 90.4873 99.4567 90.3017 1.1233
</pre></div>
</div>
<div class="admonition note">
@@ -616,7 +616,7 @@ This includes support for the VNNI 8 bit dot product instruction (CascadeLake or
<div class="section" id="deploy-a-quantized-tflite-model">
<h2>Deploy a quantized TFLite Model<a class="headerlink" href="#deploy-a-quantized-tflite-model" title="Permalink to this headline">¶</a></h2>
<p>TODO</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.750 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 21.970 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 cabc899f3d..70f68156c1 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -574,7 +574,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 120.2857 120.2028 122.1906 119.6793 0.3909
+ 119.4545 119.4453 121.5507 118.7648 0.3889
</pre></div>
</div>
<div class="admonition note">
@@ -602,7 +602,7 @@ 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 28.020 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 22.289 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 529586c263..4992f55e6f 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -512,7 +512,7 @@ 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 21.499 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 22.216 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 83396c11a1..5df2a1b411 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -454,22 +454,23 @@ 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]
- 4%|3 | 4678/132723 [00:00<00:02, 46742.60KB/s]
- 10%|9 | 12780/132723 [00:00<00:01, 66895.52KB/s]
- 16%|#6 | 21353/132723 [00:00<00:01, 75488.80KB/s]
- 23%|##2 | 29913/132723 [00:00<00:01, 79477.56KB/s]
- 29%|##8 | 38389/132723 [00:00<00:01, 81379.36KB/s]
- 35%|###5 | 46945/132723 [00:00<00:01, 82796.70KB/s]
- 42%|####1 | 55566/132723 [00:00<00:00, 83910.41KB/s]
- 48%|####8 | 64167/132723 [00:00<00:00, 84576.67KB/s]
- 55%|#####4 | 72790/132723 [00:00<00:00, 85092.30KB/s]
- 61%|######1 | 81400/132723 [00:01<00:00, 85402.15KB/s]
- 68%|######7 | 90003/132723 [00:01<00:00, 85591.63KB/s]
- 74%|#######4 | 98697/132723 [00:01<00:00, 85998.64KB/s]
- 81%|######## | 107396/132723 [00:01<00:00, 86293.42KB/s]
- 87%|########7 | 116070/132723 [00:01<00:00, 86425.89KB/s]
- 94%|#########3| 124713/132723 [00:01<00:00, 86394.87KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 83248.74KB/s]
+ 5%|4 | 6479/132723 [00:00<00:01, 64759.34KB/s]
+ 11%|# | 14287/132723 [00:00<00:01, 72585.70KB/s]
+ 17%|#6 | 22207/132723 [00:00<00:01, 75587.03KB/s]
+ 23%|##2 | 30169/132723 [00:00<00:01, 77175.50KB/s]
+ 29%|##8 | 38087/132723 [00:00<00:01, 77895.95KB/s]
+ 35%|###4 | 46016/132723 [00:00<00:01, 78361.82KB/s]
+ 41%|#### | 53965/132723 [00:00<00:01, 78726.99KB/s]
+ 47%|####6 | 61932/132723 [00:00<00:00, 79024.24KB/s]
+ 53%|#####2 | 69929/132723 [00:00<00:00, 79318.41KB/s]
+ 59%|#####8 | 77865/132723 [00:01<00:00, 79326.37KB/s]
+ 65%|######4 | 85798/132723 [00:01<00:00, 79315.41KB/s]
+ 71%|####### | 93730/132723 [00:01<00:00, 79296.01KB/s]
+ 77%|#######6 | 101713/132723 [00:01<00:00, 79456.54KB/s]
+ 83%|########2 | 109659/132723 [00:01<00:00, 79446.40KB/s]
+ 89%|########8 | 117613/132723 [00:01<00:00, 79470.92KB/s]
+ 95%|#########4| 125561/132723 [00:01<00:00, 79434.46KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 78487.37KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -508,7 +509,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 57.793 seconds)</p>
+<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 59.652 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 926915b770..dac57315d8 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -332,7 +332,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>12:51.604</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>12:54.562</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -341,35 +341,35 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>03:16.844</p></td>
+<td><p>03:20.980</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>02:57.793</p></td>
+<td><p>02:59.652</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>02:28.020</p></td>
+<td><p>02:22.289</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>01:21.499</p></td>
+<td><p>01:22.216</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized.html#sphx-glr-how-to-deploy-models-deploy-prequantized-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized.py</span></code>)</p></td>
-<td><p>01:20.750</p></td>
+<td><p>01:21.970</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></td>
-<td><p>00:36.545</p></td>
+<td><p>00:36.694</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:25.439</p></td>
+<td><p>00:25.638</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:24.709</p></td>
+<td><p>00:25.117</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></td>
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 36bd41146d..dbd0d08c3d 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -613,7 +613,7 @@ 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.zip094176ab-3666-4e35-9f1d-f186eef4307c 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.zipd4812d1f-f782-4b5f-a0f4-fecb03ebb59e 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 e3eabb4045..089bba9322 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -332,7 +332,7 @@
<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>00:47.775</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:46.433</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -341,19 +341,19 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:44.230</p></td>
+<td><p>00:43.020</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>00:02.469</p></td>
+<td><p>00:02.374</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>00:01.067</p></td>
+<td><p>00:01.032</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
-<td><p>00:00.009</p></td>
+<td><p>00:00.008</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index c6662526a0..4e271d14a4 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -517,10 +517,10 @@ 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: 7149us [7149us] (46.17%; 46.17%)
-FoldScaleAxis: 8336us [6us] (53.83%; 53.83%)
- FoldConstant: 8330us [1735us] (53.79%; 99.93%)
- InferType: 6595us [6595us] (42.59%; 79.17%)
+InferType: 6730us [6730us] (46.03%; 46.03%)
+FoldScaleAxis: 7891us [5us] (53.97%; 53.97%)
+ FoldConstant: 7885us [1640us] (53.93%; 99.93%)
+ InferType: 6245us [6245us] (42.72%; 79.20%)
</pre></div>
</div>
</div>
@@ -542,10 +542,10 @@ 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: 6676us [6676us] (45.02%; 45.02%)
-FoldScaleAxis: 8152us [6us] (54.98%; 54.98%)
- FoldConstant: 8147us [1703us] (54.94%; 99.93%)
- InferType: 6444us [6444us] (43.46%; 79.10%)
+InferType: 6273us [6273us] (44.56%; 44.56%)
+FoldScaleAxis: 7805us [5us] (55.44%; 55.44%)
+ FoldConstant: 7801us [1625us] (55.41%; 99.94%)
+ InferType: 6175us [6175us] (43.86%; 79.16%)
</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 f83b685ff5..a87ec02b4a 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -569,7 +569,7 @@ latency of convolution.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Convolution: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 54.174976 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 54.167552 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 c57c7ef18e..b420581e4b 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -911,7 +911,7 @@ 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: 13.356854 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 9.187165 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 a3328ecd63..2a84f24325 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -466,8 +466,8 @@ 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: 0.019563
-Baseline: 3.465363
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019505
+Baseline: 3.489831
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -527,7 +527,7 @@ 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: 0.310024
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.319293
</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: 0.345835
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.351425
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -655,7 +655,7 @@ 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: 0.120150
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.118191
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -738,7 +738,7 @@ 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: 0.109662
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109712
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -824,7 +824,7 @@ 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: 0.112447
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111329
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -914,7 +914,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.148376
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147327
</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 4555c056a0..ee1c74b0f7 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -332,7 +332,7 @@
<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>00:35.658</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.574</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -341,15 +341,15 @@
</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>00:32.964</p></td>
+<td><p>00:33.142</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>00:01.527</p></td>
+<td><p>00:01.353</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>00:01.167</p></td>
+<td><p>00:01.080</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 4b1387aaba..ab84319ca2 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -332,7 +332,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>08:29.376</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>08:23.574</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -341,27 +341,27 @@
</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>05:03.406</p></td>
+<td><p>04:57.852</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>01:32.824</p></td>
+<td><p>01:32.655</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>01:03.095</p></td>
+<td><p>01:02.447</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></td>
-<td><p>00:27.842</p></td>
+<td><p>00:29.023</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></td>
-<td><p>00:11.270</p></td>
+<td><p>00:10.962</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></td>
-<td><p>00:10.939</p></td>
+<td><p>00:10.635</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 3278aba230..79d19a5cb1 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
@@ -496,482 +496,360 @@ cooperative fetching, unrolling and operator fusion.</p>
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 28;
- allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [28]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [1296]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [4608]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [196], [], scope="local", align=32)[0] = 0f32
+ conv2d_nchw_1[14] = 0f32
conv2d_nchw_1[1] = 0f32
+ conv2d_nchw_1[15] = 0f32
conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[16] = 0f32
conv2d_nchw_1[3] = 0f32
+ conv2d_nchw_1[17] = 0f32
conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[18] = 0f32
conv2d_nchw_1[5] = 0f32
+ conv2d_nchw_1[19] = 0f32
conv2d_nchw_1[6] = 0f32
+ conv2d_nchw_1[20] = 0f32
conv2d_nchw_1[7] = 0f32
+ conv2d_nchw_1[21] = 0f32
conv2d_nchw_1[8] = 0f32
+ conv2d_nchw_1[22] = 0f32
conv2d_nchw_1[9] = 0f32
+ conv2d_nchw_1[23] = 0f32
conv2d_nchw_1[10] = 0f32
+ conv2d_nchw_1[24] = 0f32
conv2d_nchw_1[11] = 0f32
+ conv2d_nchw_1[25] = 0f32
conv2d_nchw_1[12] = 0f32
+ conv2d_nchw_1[26] = 0f32
conv2d_nchw_1[13] = 0f32
- for (rc.outer.outer: int32, 0, 64) {
- for (ry.outer.outer: int32, 0, 3) {
- let cse_var_2: int32 = (rc.outer.outer*72)
- let cse_var_1: int32 = (ry.outer.outer*3)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [72], [], scope="shared")[(threadIdx.x_1*4)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod((threadIdx.x_1*4), 9))) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*4), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + [...]
- }
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*4) + 1)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 1), 9))) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 1), 9)) - 8)], 0 [...]
- }
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*4) + 2)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 2), 9))) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 2), 9)) - 8)], 0 [...]
- }
- if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
- pad_temp.shared_1[((threadIdx.x_1*4) + 3)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 3), 9))) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 3), 9)) - 8)], 0 [...]
+ conv2d_nchw_1[27] = 0f32
+ for (rc.outer.outer: int32, 0, 32) {
+ let cse_var_2: int32 = (rc.outer.outer*784)
+ let cse_var_1: int32 = (rc.outer.outer*144)
+ {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1296], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else((((9 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[(((cse_var_2 + (floordiv(threadIdx.x_1, 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 56)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 56), 81)) && (floormod((threadIdx.x_1 + 56), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 56), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 31), 81)) && (floormod((threadIdx.x_1 + 31), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 31), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 6), 81)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 6), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 62), 81)) && (floormod((threadIdx.x_1 + 62), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 62), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 37), 81)) && (floormod((threadIdx.x_1 + 37), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 37), 81), 9)*7)) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 3), 9)) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 336), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 12), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 68), 81)) && (floormod((threadIdx.x_1 + 68), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 68), 81), 9)*7)) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 43), 81)) && (floormod((threadIdx.x_1 + 43), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 448), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 43), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 504)] = @tir.if_then_else((((threadIdx.x_1 < 54) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 504), 81)*49)) + ((floordiv(threadIdx.x_1, 9) + 2)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 560)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 74), 81)) && (floormod((threadIdx.x_1 + 74), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 560), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 74), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 616)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 49), 81)) && (floormod((threadIdx.x_1 + 49), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 616), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 49), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 672)] = @tir.if_then_else((((threadIdx.x_1 < 48) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 672), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 24), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 728)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 80), 81)) && (floormod((threadIdx.x_1 + 80), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 728), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 80), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 55), 81)) && (floormod((threadIdx.x_1 + 55), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 784), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 55), 81), 9)*7)) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 840)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 30), 81)) && (floormod((threadIdx.x_1 + 30), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 840), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 30), 81), 9)*7)) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 896)] = @tir.if_then_else((((9 <= floormod((threadIdx.x_1 + 5), 81)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 896), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 5), 81), 9)*7)) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 952)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 61), 81)) && (floormod((threadIdx.x_1 + 61), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 952), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 61), 81), 9)*7)) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1008)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 9) + 4), 9)) && (floormod((threadIdx.x_1 + 36), 81) < 72)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1008), 81)*49)) + (floormod((floordiv(threadIdx.x_1, 9) + 4), 9)*7)) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1064)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1 + 2), 9)) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1064), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 11), 81), 9)*7)) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1120)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 67), 81)) && (floormod((threadIdx.x_1 + 67), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1120), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 67), 81), 9)*7)) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1176)] = @tir.if_then_else(((((9 <= floormod((threadIdx.x_1 + 42), 81)) && (floormod((threadIdx.x_1 + 42), 81) < 72)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1176), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 42), 81), 9)*7)) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ pad_temp.shared_1[(threadIdx.x_1 + 1232)] = @tir.if_then_else((((threadIdx.x_1 < 55) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 1232), 81)*49)) + (floordiv(floormod((threadIdx.x_1 + 17), 81), 9)*7)) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_1 < 8), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 1288)] = 0f32
+ }
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1: Buffer(kernel.shared, float32, [4608], [], scope="shared")[threadIdx.x_2] = kernel[(((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[((((blockIdx.x*147456) + cse_var_1) + (floordiv((threadIdx.x_2 + 56), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 112), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 168), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 224), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 280)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 280), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 336), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 392), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 448), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 504)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 504), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 560), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 616)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 616), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 672), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 728)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 728), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 784), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 840)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 840), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 896), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 952)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 952), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1008)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 32256)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1064)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1064), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1120), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1176), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1232)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1232), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1288)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1288), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1344), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1400)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1400), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1456)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1456), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1512)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1512), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1568), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1624)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1624), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1680)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1680), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1736)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1736), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1792), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1848)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1848), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1904)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1904), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 1960)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1960), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 64512)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2072)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2072), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2128)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2128), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2184)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2184), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2240), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2296)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2296), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2352)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2352), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2408)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2408), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2464)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2464), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2520)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2520), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2576)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2576), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2632)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2632), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2688), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2744)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2744), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2800)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2800), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2856)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2856), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2912)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2912), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 2968)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2968), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3024)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 96768)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3080)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3080), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3136), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3192)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3192), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3248)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3248), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3304)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3304), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3360)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3360), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3416)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3416), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3472)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3472), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3528)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3528), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3584)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3584), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3640)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3640), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3696)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3696), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3752)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3752), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3808)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3808), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3864)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3864), 144)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 48)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3920)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3920), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 3976)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3976), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 88), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel[((((blockIdx.x*147456) + cse_var_1) + threadIdx.x_2) + 129024)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4088)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4088), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4144)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4144), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 112), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4200)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4200), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 8)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4256)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4256), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 80), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4312)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4312), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 136), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4368)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4368), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 16)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4424)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4424), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 104), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4480)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4480), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 144), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ kernel.shared_1[(threadIdx.x_2 + 4536)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4536), 144)*4608)) + cse_var_1) + ((floordiv(threadIdx.x_2, 3) + 24)*3)) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+ if @tir.likely((threadIdx.x_2 < 16), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 4592)] = kernel[(((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4592), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 128), 144), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+ }
+ for (rc.outer.inner: int32, 0, 2) {
+ for (rx.outer.inner: int32, 0, 3) {
+ for (rc.inner: int32, 0, 8) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2304)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 144)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2448)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 3)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2307)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 147)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2451)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 6)]))
+ conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2310)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 27)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 36)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 45)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 54)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 150)]))
+ conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((((rc.outer.inner*648) + (rc.inner*81)) + rx.outer.inner) + floormod(threadIdx.x, 7)) + 72)]*kernel.shared_1[(((((floordiv(threadIdx.x, 7)*288) + (rc.outer.inner*72)) + (rc.inner*9)) + rx.outer.inner) + 2454)]))
}
}
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 64), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 128), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 256), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 320), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 448), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 512), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 640), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 704), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 147456)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 832), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 896), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 184320)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1024), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1088), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 221184)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1216), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1280), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1408), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1472), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 294912)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1600), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1664), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 331776)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1792), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1856), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 368640)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1984), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2048), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 405504)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2176), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2240), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 442368)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2368), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2432), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 479232)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2560), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2624), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 516096)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2752), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2816), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 552960)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2944), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 3008), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((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[9]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 47)]))
}
}
}
for (i1.inner: int32, 0, 2) {
- for (i3.inner: int32, 0, 7) {
- compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*98)) + (i1.inner*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[((i1.inner*7) + i3.inner)] + bias[(((floordiv(blockIdx.x, 7)*128) + (threadIdx.x*2)) + i1.inner)]), 0f32)
+ for (i2.inner: int32, 0, 7) {
+ let cse_var_3: int32 = ((i1.inner*7) + i2.inner)
+ {
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[cse_var_3] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
+ compute[((((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7)) + 784)] = max((conv2d_nchw_1[(cse_var_3 + 14)] + bias[((((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner) + 16)]), 0f32)
+ }
}
}
}
@@ -1009,7 +887,7 @@ cooperative fetching, unrolling and operator fusion.</p>
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.355 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.412 ms
</pre></div>
</div>
</div>
@@ -1038,21 +916,21 @@ 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_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
-conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
-conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, 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=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=2)
+conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=7)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
-conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
-conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
+conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
-conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=8)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
@@ -1061,13 +939,13 @@ compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
-compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
-compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
+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=2)
+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_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
-compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
-compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
+compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -1087,12 +965,12 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=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=4)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=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, "unroll_explicit", True)
@@ -1112,429 +990,247 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[28];
+ __shared__ float pad_temp_shared[1296];
+ __shared__ float kernel_shared[4608];
conv2d_nchw[0] = 0.000000e+00f;
+ conv2d_nchw[14] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
+ conv2d_nchw[15] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[16] = 0.000000e+00f;
conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[17] = 0.000000e+00f;
conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[18] = 0.000000e+00f;
conv2d_nchw[5] = 0.000000e+00f;
+ conv2d_nchw[19] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
+ conv2d_nchw[20] = 0.000000e+00f;
conv2d_nchw[7] = 0.000000e+00f;
+ conv2d_nchw[21] = 0.000000e+00f;
conv2d_nchw[8] = 0.000000e+00f;
+ conv2d_nchw[22] = 0.000000e+00f;
conv2d_nchw[9] = 0.000000e+00f;
+ conv2d_nchw[23] = 0.000000e+00f;
conv2d_nchw[10] = 0.000000e+00f;
+ conv2d_nchw[24] = 0.000000e+00f;
conv2d_nchw[11] = 0.000000e+00f;
+ conv2d_nchw[25] = 0.000000e+00f;
conv2d_nchw[12] = 0.000000e+00f;
+ conv2d_nchw[26] = 0.000000e+00f;
conv2d_nchw[13] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
- for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
- __syncthreads();
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 4) % 9))) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 1) % 9))) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 2) % 9))) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
- }
- if (((int)threadIdx.x) < 18) {
- pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 3) % 9))) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
+ conv2d_nchw[27] = 0.000000e+00f;
+ for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
+ __syncthreads();
+ pad_temp_shared[((int)threadIdx.x)] = ((((9 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((9 <= ((((int)threadIdx.x) + 56) % 81)) && (((((int)threadIdx.x) + 56) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 56) / 81) * 49)) + ((((((int)threadIdx.x) + 56) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((9 <= ((((int)threadIdx.x) + 31) % 81)) && (((((int)threadIdx.x) + 31) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 112) / 81) * 49)) + ((((((int)threadIdx.x) + 31) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((3 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 168) / 81) * 49)) + (((((int)threadIdx.x) + 6) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((9 <= ((((int)threadIdx.x) + 62) % 81)) && (((((int)threadIdx.x) + 62) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 224) / 81) * 49)) + ((((((int)threadIdx.x) + 62) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 280)] = (((((9 <= ((((int)threadIdx.x) + 37) % 81)) && (((((int)threadIdx.x) + 37) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 280) / 81) * 49)) + ((((((int)threadIdx.x) + 37) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 336)] = (((1 <= ((((int)threadIdx.x) + 3) % 9)) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 336) / 81) * 49)) + (((((int)threadIdx.x) + 12) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((9 <= ((((int)threadIdx.x) + 68) % 81)) && (((((int)threadIdx.x) + 68) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 392) / 81) * 49)) + ((((((int)threadIdx.x) + 68) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((9 <= ((((int)threadIdx.x) + 43) % 81)) && (((((int)threadIdx.x) + 43) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 448) / 81) * 49)) + ((((((int)threadIdx.x) + 43) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 504)] = ((((((int)threadIdx.x) < 54) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 504) / 81) * 49)) + ((((int)threadIdx.x) / 9) * 7)) + (((int)threadIdx.x) % 9)) + 6)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((9 <= ((((int)threadIdx.x) + 74) % 81)) && (((((int)threadIdx.x) + 74) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 560) / 81) * 49)) + ((((((int)threadIdx.x) + 74) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 616)] = (((((9 <= ((((int)threadIdx.x) + 49) % 81)) && (((((int)threadIdx.x) + 49) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 616) / 81) * 49)) + ((((((int)threadIdx.x) + 49) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 672)] = ((((((int)threadIdx.x) < 48) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 672) / 81) * 49)) + (((((int)threadIdx.x) + 24) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 728)] = (((((9 <= ((((int)threadIdx.x) + 80) % 81)) && (((((int)threadIdx.x) + 80) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 728) / 81) * 49)) + ((((((int)threadIdx.x) + 80) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((9 <= ((((int)threadIdx.x) + 55) % 81)) && (((((int)threadIdx.x) + 55) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 784) / 81) * 49)) + ((((((int)threadIdx.x) + 55) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 840)] = (((((9 <= ((((int)threadIdx.x) + 30) % 81)) && (((((int)threadIdx.x) + 30) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 840) / 81) * 49)) + ((((((int)threadIdx.x) + 30) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 896)] = ((((4 <= ((int)threadIdx.x)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 896) / 81) * 49)) + (((((int)threadIdx.x) + 5) / 9) * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 952)] = (((((9 <= ((((int)threadIdx.x) + 61) % 81)) && (((((int)threadIdx.x) + 61) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 952) / 81) * 49)) + ((((((int)threadIdx.x) + 61) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1008)] = (((((1 <= (((((int)threadIdx.x) / 9) + 4) % 9)) && (((((int)threadIdx.x) + 36) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1008) / 81) * 49)) + ((((((int)threadIdx.x) / 9) + 4) % 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1064)] = (((1 <= ((((int)threadIdx.x) + 2) % 9)) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1064) / 81) * 49)) + (((((int)threadIdx.x) + 11) / 9) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1120)] = (((((9 <= ((((int)threadIdx.x) + 67) % 81)) && (((((int)threadIdx.x) + 67) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1120) / 81) * 49)) + ((((((int)threadIdx.x) + 67) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((9 <= ((((int)threadIdx.x) + 42) % 81)) && (((((int)threadIdx.x) + 42) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1176) / 81) * 49)) + ((((((int)threadIdx.x) + 42) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1232)] = ((((((int)threadIdx.x) < 55) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 784) + (((((int)threadIdx.x) + 1232) / 81) * 49)) + (((((int)threadIdx.x) + 17) / 9) * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 8) {
+ pad_temp_shared[(((int)threadIdx.x) + 1288)] = 0.000000e+00f;
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x))];
+ kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 168) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 280) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 336) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 504)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 504) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 616)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 616) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 672) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 728)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 728) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 840) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 896) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 952) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 32256)];
+ kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1064) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1120) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1232) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1288) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1344) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1400) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1456) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1512) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1624) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1680) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1736) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1792) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1848) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1904) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 2072)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2072) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2128) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2184)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2184) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2240) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2296)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2296) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 2408)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2408) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2464) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2520)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2520) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 2576)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2576) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2632)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2632) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2688) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2800)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2800) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2856)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2856) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2912) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2968)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2968) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3024)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 96768)];
+ kernel_shared[(((int)threadIdx.x) + 3080)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3080) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3136) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3192)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3192) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 3248)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3248) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3304)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3304) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3360) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 3416)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3416) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3472)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3472) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3528) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3584) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 128) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3640)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3640) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 40) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3696)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3696) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 32) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3752)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3752) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 8) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3808) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 64) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3864)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3864) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) / 3) + 40) % 48) * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3920) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 32) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3976)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3976) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 88) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((int)blockIdx.x) * 147456) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 129024)];
+ kernel_shared[(((int)threadIdx.x) + 4088)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4088) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 56) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4144)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4144) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 112) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4200)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4200) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 24)];
+ kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4256) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 80) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4312) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 136) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4368)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4368) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 48)];
+ kernel_shared[(((int)threadIdx.x) + 4424)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4424) / 144) * 4608)) + (rc_outer_outer * 144)) + ((((((int)threadIdx.x) + 104) % 144) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4480) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 16) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4536)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4536) / 144) * 4608)) + (rc_outer_outer * 144)) + ((int)threadIdx.x)) + 72)];
+ if (((int)threadIdx.x) < 16) {
+ kernel_shared[(((int)threadIdx.x) + 4592)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4592) / 144) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 128) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ }
+ __syncthreads();
+ for (int rc_outer_inner = 0; rc_outer_inner < 2; ++rc_outer_inner) {
+ for (int rx_outer_inner = 0; rx_outer_inner < 3; ++rx_outer_inner) {
+ for (int rc_inner = 0; rc_inner < 8; ++rc_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2304)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 144)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2448)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 3)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2307)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 147)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2451)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 6)]));
+ conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2310)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 27)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 36)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 45)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 54)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 150)]));
+ conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((((rc_outer_inner * 648) + (rc_inner * 81)) + rx_outer_inner) + (((int)threadIdx.x) % 7)) + 72)] * kernel_shared[((((((((int)threadIdx.x) / 7) * 288) + (rc_outer_inner * 72)) + (rc_inner * 9)) + rx_outer_inner) + 2454)]));
+ }
}
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
- kernel_shared[(((int)threadIdx.x) + 64)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
- kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
- kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
- kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
- kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
- kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
- kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
- kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
- kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
- kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
- kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
- kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
- kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
- kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
- kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
- kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
}
}
for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
+ for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ compute[((((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7)) + 784)] = max((conv2d_nchw[(((i1_inner * 7) + i2_inner) + 14)] + bias[((((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
}
}
}
@@ -1572,7 +1268,7 @@ 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> ( 5 minutes 3.406 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 57.852 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 c99931ddab..12dd72ad4f 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -907,7 +907,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1759 8.1757 8.1766 8.1755 0.0005
+ 8.2262 8.2225 8.2344 8.2218 0.0058
</pre></div>
</div>
</div>
@@ -929,7 +929,7 @@ 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 3.095 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 2.447 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 739859de08..a1c4a0d946 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -926,7 +926,7 @@ 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)
- 760.6483 760.0349 764.3487 757.5613 2.8047
+ 750.4478 750.8018 751.6778 748.8640 1.1757
</pre></div>
</div>
</div>
@@ -948,7 +948,7 @@ 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 32.824 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 32.655 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 1da909f4c8..21b3a07eb5 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -624,30 +624,1039 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_9: placeholder_15: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_6: placeholder_17: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_7: placeholder_19: Buffer(placeholder_12, int32, [4916], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 16) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [4096]), storage_scope = global {
- for (i.outer.inner: int32, 0, 8) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 16) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [4096], [])[((((i.outer.inner*512) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
- }
- }
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (i.inner: int32, 0, 16) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner)
- let cse_var_2: int32 = ((((i.outer.inner*512) + (i.inner*32)) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((i.outer.inner*4096) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
- }
- }
- }
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_16: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
+ let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ allocate(compute_4: Pointer(global float32), float32, [256]), storage_scope = global {
+ compute_5: Buffer(compute_4, float32, [256], [])[0] = 0f32
+ compute_5[1] = 0f32
+ compute_5[2] = 0f32
+ compute_5[3] = 0f32
+ compute_5[4] = 0f32
+ compute_5[5] = 0f32
+ compute_5[6] = 0f32
+ compute_5[7] = 0f32
+ compute_5[8] = 0f32
+ compute_5[9] = 0f32
+ compute_5[10] = 0f32
+ compute_5[11] = 0f32
+ compute_5[12] = 0f32
+ compute_5[13] = 0f32
+ compute_5[14] = 0f32
+ compute_5[15] = 0f32
+ compute_5[16] = 0f32
+ compute_5[17] = 0f32
+ compute_5[18] = 0f32
+ compute_5[19] = 0f32
+ compute_5[20] = 0f32
+ compute_5[21] = 0f32
+ compute_5[22] = 0f32
+ compute_5[23] = 0f32
+ compute_5[24] = 0f32
+ compute_5[25] = 0f32
+ compute_5[26] = 0f32
+ compute_5[27] = 0f32
+ compute_5[28] = 0f32
+ compute_5[29] = 0f32
+ compute_5[30] = 0f32
+ compute_5[31] = 0f32
+ compute_5[32] = 0f32
+ compute_5[33] = 0f32
+ compute_5[34] = 0f32
+ compute_5[35] = 0f32
+ compute_5[36] = 0f32
+ compute_5[37] = 0f32
+ compute_5[38] = 0f32
+ compute_5[39] = 0f32
+ compute_5[40] = 0f32
+ compute_5[41] = 0f32
+ compute_5[42] = 0f32
+ compute_5[43] = 0f32
+ compute_5[44] = 0f32
+ compute_5[45] = 0f32
+ compute_5[46] = 0f32
+ compute_5[47] = 0f32
+ compute_5[48] = 0f32
+ compute_5[49] = 0f32
+ compute_5[50] = 0f32
+ compute_5[51] = 0f32
+ compute_5[52] = 0f32
+ compute_5[53] = 0f32
+ compute_5[54] = 0f32
+ compute_5[55] = 0f32
+ compute_5[56] = 0f32
+ compute_5[57] = 0f32
+ compute_5[58] = 0f32
+ compute_5[59] = 0f32
+ compute_5[60] = 0f32
+ compute_5[61] = 0f32
+ compute_5[62] = 0f32
+ compute_5[63] = 0f32
+ compute_5[64] = 0f32
+ compute_5[65] = 0f32
+ compute_5[66] = 0f32
+ compute_5[67] = 0f32
+ compute_5[68] = 0f32
+ compute_5[69] = 0f32
+ compute_5[70] = 0f32
+ compute_5[71] = 0f32
+ compute_5[72] = 0f32
+ compute_5[73] = 0f32
+ compute_5[74] = 0f32
+ compute_5[75] = 0f32
+ compute_5[76] = 0f32
+ compute_5[77] = 0f32
+ compute_5[78] = 0f32
+ compute_5[79] = 0f32
+ compute_5[80] = 0f32
+ compute_5[81] = 0f32
+ compute_5[82] = 0f32
+ compute_5[83] = 0f32
+ compute_5[84] = 0f32
+ compute_5[85] = 0f32
+ compute_5[86] = 0f32
+ compute_5[87] = 0f32
+ compute_5[88] = 0f32
+ compute_5[89] = 0f32
+ compute_5[90] = 0f32
+ compute_5[91] = 0f32
+ compute_5[92] = 0f32
+ compute_5[93] = 0f32
+ compute_5[94] = 0f32
+ compute_5[95] = 0f32
+ compute_5[96] = 0f32
+ compute_5[97] = 0f32
+ compute_5[98] = 0f32
+ compute_5[99] = 0f32
+ compute_5[100] = 0f32
+ compute_5[101] = 0f32
+ compute_5[102] = 0f32
+ compute_5[103] = 0f32
+ compute_5[104] = 0f32
+ compute_5[105] = 0f32
+ compute_5[106] = 0f32
+ compute_5[107] = 0f32
+ compute_5[108] = 0f32
+ compute_5[109] = 0f32
+ compute_5[110] = 0f32
+ compute_5[111] = 0f32
+ compute_5[112] = 0f32
+ compute_5[113] = 0f32
+ compute_5[114] = 0f32
+ compute_5[115] = 0f32
+ compute_5[116] = 0f32
+ compute_5[117] = 0f32
+ compute_5[118] = 0f32
+ compute_5[119] = 0f32
+ compute_5[120] = 0f32
+ compute_5[121] = 0f32
+ compute_5[122] = 0f32
+ compute_5[123] = 0f32
+ compute_5[124] = 0f32
+ compute_5[125] = 0f32
+ compute_5[126] = 0f32
+ compute_5[127] = 0f32
+ compute_5[128] = 0f32
+ compute_5[129] = 0f32
+ compute_5[130] = 0f32
+ compute_5[131] = 0f32
+ compute_5[132] = 0f32
+ compute_5[133] = 0f32
+ compute_5[134] = 0f32
+ compute_5[135] = 0f32
+ compute_5[136] = 0f32
+ compute_5[137] = 0f32
+ compute_5[138] = 0f32
+ compute_5[139] = 0f32
+ compute_5[140] = 0f32
+ compute_5[141] = 0f32
+ compute_5[142] = 0f32
+ compute_5[143] = 0f32
+ compute_5[144] = 0f32
+ compute_5[145] = 0f32
+ compute_5[146] = 0f32
+ compute_5[147] = 0f32
+ compute_5[148] = 0f32
+ compute_5[149] = 0f32
+ compute_5[150] = 0f32
+ compute_5[151] = 0f32
+ compute_5[152] = 0f32
+ compute_5[153] = 0f32
+ compute_5[154] = 0f32
+ compute_5[155] = 0f32
+ compute_5[156] = 0f32
+ compute_5[157] = 0f32
+ compute_5[158] = 0f32
+ compute_5[159] = 0f32
+ compute_5[160] = 0f32
+ compute_5[161] = 0f32
+ compute_5[162] = 0f32
+ compute_5[163] = 0f32
+ compute_5[164] = 0f32
+ compute_5[165] = 0f32
+ compute_5[166] = 0f32
+ compute_5[167] = 0f32
+ compute_5[168] = 0f32
+ compute_5[169] = 0f32
+ compute_5[170] = 0f32
+ compute_5[171] = 0f32
+ compute_5[172] = 0f32
+ compute_5[173] = 0f32
+ compute_5[174] = 0f32
+ compute_5[175] = 0f32
+ compute_5[176] = 0f32
+ compute_5[177] = 0f32
+ compute_5[178] = 0f32
+ compute_5[179] = 0f32
+ compute_5[180] = 0f32
+ compute_5[181] = 0f32
+ compute_5[182] = 0f32
+ compute_5[183] = 0f32
+ compute_5[184] = 0f32
+ compute_5[185] = 0f32
+ compute_5[186] = 0f32
+ compute_5[187] = 0f32
+ compute_5[188] = 0f32
+ compute_5[189] = 0f32
+ compute_5[190] = 0f32
+ compute_5[191] = 0f32
+ compute_5[192] = 0f32
+ compute_5[193] = 0f32
+ compute_5[194] = 0f32
+ compute_5[195] = 0f32
+ compute_5[196] = 0f32
+ compute_5[197] = 0f32
+ compute_5[198] = 0f32
+ compute_5[199] = 0f32
+ compute_5[200] = 0f32
+ compute_5[201] = 0f32
+ compute_5[202] = 0f32
+ compute_5[203] = 0f32
+ compute_5[204] = 0f32
+ compute_5[205] = 0f32
+ compute_5[206] = 0f32
+ compute_5[207] = 0f32
+ compute_5[208] = 0f32
+ compute_5[209] = 0f32
+ compute_5[210] = 0f32
+ compute_5[211] = 0f32
+ compute_5[212] = 0f32
+ compute_5[213] = 0f32
+ compute_5[214] = 0f32
+ compute_5[215] = 0f32
+ compute_5[216] = 0f32
+ compute_5[217] = 0f32
+ compute_5[218] = 0f32
+ compute_5[219] = 0f32
+ compute_5[220] = 0f32
+ compute_5[221] = 0f32
+ compute_5[222] = 0f32
+ compute_5[223] = 0f32
+ compute_5[224] = 0f32
+ compute_5[225] = 0f32
+ compute_5[226] = 0f32
+ compute_5[227] = 0f32
+ compute_5[228] = 0f32
+ compute_5[229] = 0f32
+ compute_5[230] = 0f32
+ compute_5[231] = 0f32
+ compute_5[232] = 0f32
+ compute_5[233] = 0f32
+ compute_5[234] = 0f32
+ compute_5[235] = 0f32
+ compute_5[236] = 0f32
+ compute_5[237] = 0f32
+ compute_5[238] = 0f32
+ compute_5[239] = 0f32
+ compute_5[240] = 0f32
+ compute_5[241] = 0f32
+ compute_5[242] = 0f32
+ compute_5[243] = 0f32
+ compute_5[244] = 0f32
+ compute_5[245] = 0f32
+ compute_5[246] = 0f32
+ compute_5[247] = 0f32
+ compute_5[248] = 0f32
+ compute_5[249] = 0f32
+ compute_5[250] = 0f32
+ compute_5[251] = 0f32
+ compute_5[252] = 0f32
+ compute_5[253] = 0f32
+ compute_5[254] = 0f32
+ compute_5[255] = 0f32
+ for (elem_idx: int32, 0, (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[0] = (compute_5[0] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[1] = (compute_5[1] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[2] = (compute_5[2] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[3] = (compute_5[3] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[4] = (compute_5[4] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[5] = (compute_5[5] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[6] = (compute_5[6] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[7] = (compute_5[7] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[8] = (compute_5[8] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[9] = (compute_5[9] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[10] = (compute_5[10] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[11] = (compute_5[11] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[12] = (compute_5[12] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[13] = (compute_5[13] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[14] = (compute_5[14] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[15] = (compute_5[15] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[16] = (compute_5[16] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[17] = (compute_5[17] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[18] = (compute_5[18] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[19] = (compute_5[19] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[20] = (compute_5[20] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[21] = (compute_5[21] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[22] = (compute_5[22] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[23] = (compute_5[23] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[24] = (compute_5[24] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[25] = (compute_5[25] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[26] = (compute_5[26] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[27] = (compute_5[27] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[28] = (compute_5[28] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[29] = (compute_5[29] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[30] = (compute_5[30] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[31] = (compute_5[31] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[32] = (compute_5[32] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[33] = (compute_5[33] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[34] = (compute_5[34] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[35] = (compute_5[35] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[36] = (compute_5[36] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[37] = (compute_5[37] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[38] = (compute_5[38] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[39] = (compute_5[39] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[40] = (compute_5[40] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[41] = (compute_5[41] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[42] = (compute_5[42] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[43] = (compute_5[43] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[44] = (compute_5[44] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[45] = (compute_5[45] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[46] = (compute_5[46] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[47] = (compute_5[47] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[48] = (compute_5[48] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[49] = (compute_5[49] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[50] = (compute_5[50] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[51] = (compute_5[51] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[52] = (compute_5[52] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[53] = (compute_5[53] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[54] = (compute_5[54] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[55] = (compute_5[55] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[56] = (compute_5[56] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[57] = (compute_5[57] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[58] = (compute_5[58] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[59] = (compute_5[59] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[60] = (compute_5[60] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[61] = (compute_5[61] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[62] = (compute_5[62] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[63] = (compute_5[63] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[64] = (compute_5[64] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[65] = (compute_5[65] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[66] = (compute_5[66] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[67] = (compute_5[67] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[68] = (compute_5[68] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[69] = (compute_5[69] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[70] = (compute_5[70] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[71] = (compute_5[71] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[72] = (compute_5[72] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[73] = (compute_5[73] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[74] = (compute_5[74] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[75] = (compute_5[75] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[76] = (compute_5[76] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[77] = (compute_5[77] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[78] = (compute_5[78] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[79] = (compute_5[79] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1024)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[80] = (compute_5[80] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[81] = (compute_5[81] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[82] = (compute_5[82] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[83] = (compute_5[83] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[84] = (compute_5[84] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[85] = (compute_5[85] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[86] = (compute_5[86] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[87] = (compute_5[87] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[88] = (compute_5[88] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[89] = (compute_5[89] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[90] = (compute_5[90] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[91] = (compute_5[91] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[92] = (compute_5[92] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[93] = (compute_5[93] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[94] = (compute_5[94] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[95] = (compute_5[95] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1280)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[96] = (compute_5[96] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[97] = (compute_5[97] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[98] = (compute_5[98] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[99] = (compute_5[99] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[100] = (compute_5[100] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[101] = (compute_5[101] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[102] = (compute_5[102] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[103] = (compute_5[103] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[104] = (compute_5[104] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[105] = (compute_5[105] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[106] = (compute_5[106] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[107] = (compute_5[107] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[108] = (compute_5[108] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[109] = (compute_5[109] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[110] = (compute_5[110] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[111] = (compute_5[111] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1536)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[112] = (compute_5[112] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[113] = (compute_5[113] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[114] = (compute_5[114] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[115] = (compute_5[115] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[116] = (compute_5[116] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[117] = (compute_5[117] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[118] = (compute_5[118] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[119] = (compute_5[119] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[120] = (compute_5[120] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[121] = (compute_5[121] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[122] = (compute_5[122] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[123] = (compute_5[123] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[124] = (compute_5[124] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[125] = (compute_5[125] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[126] = (compute_5[126] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[127] = (compute_5[127] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 1792)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[128] = (compute_5[128] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[129] = (compute_5[129] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[130] = (compute_5[130] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[131] = (compute_5[131] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[132] = (compute_5[132] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[133] = (compute_5[133] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[134] = (compute_5[134] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[135] = (compute_5[135] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[136] = (compute_5[136] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[137] = (compute_5[137] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[138] = (compute_5[138] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[139] = (compute_5[139] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[140] = (compute_5[140] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[141] = (compute_5[141] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[142] = (compute_5[142] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[143] = (compute_5[143] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2048)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[144] = (compute_5[144] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[145] = (compute_5[145] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[146] = (compute_5[146] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[147] = (compute_5[147] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[148] = (compute_5[148] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[149] = (compute_5[149] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[150] = (compute_5[150] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[151] = (compute_5[151] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[152] = (compute_5[152] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[153] = (compute_5[153] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[154] = (compute_5[154] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[155] = (compute_5[155] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[156] = (compute_5[156] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[157] = (compute_5[157] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[158] = (compute_5[158] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[159] = (compute_5[159] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2304)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[160] = (compute_5[160] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[161] = (compute_5[161] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[162] = (compute_5[162] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[163] = (compute_5[163] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[164] = (compute_5[164] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[165] = (compute_5[165] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[166] = (compute_5[166] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[167] = (compute_5[167] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[168] = (compute_5[168] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[169] = (compute_5[169] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[170] = (compute_5[170] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[171] = (compute_5[171] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[172] = (compute_5[172] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[173] = (compute_5[173] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[174] = (compute_5[174] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[175] = (compute_5[175] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2560)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[176] = (compute_5[176] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[177] = (compute_5[177] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[178] = (compute_5[178] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[179] = (compute_5[179] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[180] = (compute_5[180] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[181] = (compute_5[181] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[182] = (compute_5[182] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[183] = (compute_5[183] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[184] = (compute_5[184] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[185] = (compute_5[185] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[186] = (compute_5[186] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[187] = (compute_5[187] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[188] = (compute_5[188] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[189] = (compute_5[189] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[190] = (compute_5[190] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[191] = (compute_5[191] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 2816)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[192] = (compute_5[192] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[193] = (compute_5[193] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[194] = (compute_5[194] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[195] = (compute_5[195] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[196] = (compute_5[196] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[197] = (compute_5[197] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[198] = (compute_5[198] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[199] = (compute_5[199] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[200] = (compute_5[200] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[201] = (compute_5[201] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[202] = (compute_5[202] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[203] = (compute_5[203] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[204] = (compute_5[204] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[205] = (compute_5[205] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[206] = (compute_5[206] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[207] = (compute_5[207] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3072)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[208] = (compute_5[208] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[209] = (compute_5[209] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[210] = (compute_5[210] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[211] = (compute_5[211] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[212] = (compute_5[212] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[213] = (compute_5[213] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[214] = (compute_5[214] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[215] = (compute_5[215] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[216] = (compute_5[216] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[217] = (compute_5[217] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[218] = (compute_5[218] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[219] = (compute_5[219] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[220] = (compute_5[220] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[221] = (compute_5[221] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[222] = (compute_5[222] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[223] = (compute_5[223] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3328)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[224] = (compute_5[224] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[225] = (compute_5[225] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[226] = (compute_5[226] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[227] = (compute_5[227] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[228] = (compute_5[228] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[229] = (compute_5[229] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[230] = (compute_5[230] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[231] = (compute_5[231] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[232] = (compute_5[232] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[233] = (compute_5[233] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[234] = (compute_5[234] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[235] = (compute_5[235] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[236] = (compute_5[236] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[237] = (compute_5[237] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[238] = (compute_5[238] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[239] = (compute_5[239] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3584)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[240] = (compute_5[240] + (placeholder_1[((placeholder_3[cse_var_1]*16) + (elem_idx*16))]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[241] = (compute_5[241] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[242] = (compute_5[242] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[243] = (compute_5[243] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[244] = (compute_5[244] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[245] = (compute_5[245] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[246] = (compute_5[246] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[247] = (compute_5[247] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[248] = (compute_5[248] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[249] = (compute_5[249] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[250] = (compute_5[250] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[251] = (compute_5[251] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[252] = (compute_5[252] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[253] = (compute_5[253] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[254] = (compute_5[254] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
+ }
+ if @tir.likely((elem_idx < (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])), dtype=bool) {
+ compute_5[255] = (compute_5[255] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*4096) + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 3840)], 0f32)))
}
}
- for (i0.inner: int32, 0, 128) {
- let cse_var_4: int32 = ((i0.inner*512) + (i0.outer.i1.outer.fused*32))
- compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
+ for (i0.inner: int32, 0, 16) {
+ let cse_var_2: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i0.inner*512)) + (cse_var_1*16))
+ compute[ramp(cse_var_2, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_2, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -685,7 +1694,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.498 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 3.546 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 7cea7d3b73..fea7f1eae6 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -332,7 +332,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:34.299</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:29.623</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -341,11 +341,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></td>
-<td><p>00:34.263</p></td>
+<td><p>00:29.587</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>00:00.021</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></td>
diff --git a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
index b6aefd7bb8..f78e187ac2 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -684,8 +684,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 16, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5634509
-No: 2 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 256, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5575908
+No: 2 GFLOPS: 9.07/9.07 result: MeasureResult(costs=(0.02553097525,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.472261667251587, timestamp=1665986039.31081) [('tile_f', [-1, 8, 8, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3497937
+No: 3 GFLOPS: 0.00/9.07 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -807,9 +808,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6063340
-No: 3 GFLOPS: 1.30/1.30 result: MeasureResult(costs=(0.17854515075,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.044389009475708, timestamp=1665892986.3266659) [('tile_f', [-1, 2, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1907181
-No: 4 GFLOPS: 0.00/1.30 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8691660
+No: 4 GFLOPS: 0.00/9.07 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -931,9 +931,10 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9813116
-No: 5 GFLOPS: 65.06/65.06 result: MeasureResult(costs=(0.003558434909090909,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3140735626220703, timestamp=1665892988.8143306) [('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8760073
-No: 6 GFLOPS: 0.00/65.06 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 1, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2807403
+No: 5 GFLOPS: 37.86/37.86 result: MeasureResult(costs=(0.006114725411764706,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5770478248596191, timestamp=1665986043.0153155) [('tile_f', [-1, 4, 2, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,98572
+No: 6 GFLOPS: 100.64/100.64 result: MeasureResult(costs=(0.002300276318181818,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4073760509490967, timestamp=1665986043.7016363) [('tile_f', [-1, 4, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4649736
+No: 7 GFLOPS: 0.00/100.64 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1055,8 +1056,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('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,7580739
-No: 7 GFLOPS: 0.00/65.06 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2303991
+No: 8 GFLOPS: 338.83/338.83 result: MeasureResult(costs=(0.0006832309115646258,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.099130630493164, timestamp=1665986045.2803166) [('tile_f', [-1, 8, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2336027
+No: 9 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1178,8 +1180,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 128, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4563777
-No: 8 GFLOPS: 0.00/65.06 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 64, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7379334
+No: 10 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1301,11 +1303,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 8, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,433212
-No: 9 GFLOPS: 787.40/787.40 result: MeasureResult(costs=(0.00029400837535014003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2635645866394043, timestamp=1665892993.1627276) [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9221146
-No: 10 GFLOPS: 257.16/787.40 result: MeasureResult(costs=(0.0009002157062146893,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7292754650115967, timestamp=1665892994.0822725) [('tile_f', [-1, 2, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3945162
-No: 11 GFLOPS: 98.73/787.40 result: MeasureResult(costs=(0.002344907162790698,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1468474864959717, timestamp=1665892994.7787464) [('tile_f', [-1, 1, 8, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5859414
-No: 12 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2435702
+No: 11 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1427,8 +1426,9 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7516258
-No: 13 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 512]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1389739
+No: 12 GFLOPS: 6.66/338.83 result: MeasureResult(costs=(0.034783584500000006,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4205341339111328, timestamp=1665986046.9337785) [('tile_f', [-1, 4, 4, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,7877
+No: 13 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1550,8 +1550,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 8, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9387431
-No: 14 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 64, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,843834
+No: 14 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1673,8 +1673,10 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 512]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10438999
-No: 15 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6378833
+No: 15 GFLOPS: 22.50/338.83 result: MeasureResult(costs=(0.010287662818181819,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6355750560760498, timestamp=1665986051.6126497) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2252599
+No: 16 GFLOPS: 5.41/338.83 result: MeasureResult(costs=(0.04276027275,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.494520425796509, timestamp=1665986052.5202076) [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7074676
+No: 17 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1796,8 +1798,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7666966
-No: 16 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10224784
+No: 18 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1919,8 +1921,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3772509
-No: 17 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 4]), ('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, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2735610
+No: 19 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2042,9 +2044,8 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 1, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8003956
-No: 18 GFLOPS: 10.68/787.40 result: MeasureResult(costs=(0.021684544833333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.315142631530762, timestamp=1665893000.459618) [('tile_f', [-1, 32, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1873315
-No: 19 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 8]), ('tile_y', [-1, 1, 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', 1)],None,7440558
+No: 20 GFLOPS: 0.00/338.83 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2166,130 +2167,7 @@ Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4865648
-No: 20 GFLOPS: 0.00/787.40 result: Traceback (most recent call last):
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
- func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
- func = build(s, args, target_host=task.target_host, runtime=runtime)
- File "/workspace/python/tvm/driver/build_module.py", line 227, in build
- input_mod = lower(inputs, args, name=name, binds=binds)
- File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
- return ffi.lower_schedule(inp, args, name, binds, simple_mode)
- File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
- File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
- File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
-tvm._ffi.base.TVMError: Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1731
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1671
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1631
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1646
- 13: operator()
- at ../src/driver/driver_api.cc:380
- 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:366
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:261
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:453
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1750
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1694
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1618
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
-
-Traceback (most recent call last):
- 24: TVMFuncCall
- at ../src/runtime/c_runtime_api.cc:477
- 23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 22: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 21: operator()
- at ../include/tvm/runtime/packed_func.h:1731
- 20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
- at ../include/tvm/runtime/packed_func.h:1671
- 19: run<>
- at ../include/tvm/runtime/packed_func.h:1631
- 18: run<tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1631
- 14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
- at ../include/tvm/runtime/packed_func.h:1646
- 13: operator()
- at ../src/driver/driver_api.cc:380
- 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:366
- 11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
- at ../src/driver/driver_api.cc:261
- 10: tvm::transform::Pass::operator()(tvm::IRModule) const
- at ../src/ir/transform.cc:258
- 9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:453
- 7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/ir/transform.cc:274
- 6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
- at ../src/tir/ir/transform.cc:100
- 5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
- at ../include/tvm/runtime/packed_func.h:1750
- 4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
- at ../include/tvm/runtime/packed_func.h:1694
- 3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
- at ../include/tvm/runtime/packed_func.h:1618
- 2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
- at ../include/tvm/runtime/packed_func.h:1217
- 1: Call
- at ../include/tvm/runtime/packed_func.h:1213
- 0: operator()
- at ../src/runtime/c_runtime_api.cc:534
- File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
- File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
- raise InstantiationError("Skipped because of invalid gpu kernel")
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1897725
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6521196
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2328,9 +2206,9 @@ 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, 4, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9221146
+[('tile_f', [-1, 8, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2336027
Finish loading 20 records
-Time cost of this operator: 0.000657
+Time cost of this operator: 0.001080
</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 9788e15b5b..ed9c14d07a 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -587,10 +587,10 @@ 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 313.3 98.735 (1, 2, 10, 10, 3) 2 1 [313.3]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.039 0.958 (1, 6, 10, 10) 1 1 [3.039]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.976 0.307 (1, 1, 10, 10, 3) 1 1 [0.976]
-Total_time - 317.315 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 311.7 98.727 (1, 2, 10, 10, 3) 2 1 [311.7]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.048 0.966 (1, 6, 10, 10) 1 1 [3.048]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 0.307 (1, 1, 10, 10, 3) 1 1 [0.969]
+Total_time - 315.718 - - - - -
</pre></div>
</div>
</div>
@@ -641,10 +641,10 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 100.8 97.338 (1, 6, 10, 10, 1) 2 1 [100.8]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.788 1.726 (1, 6, 10, 10) 1 1 [1.788]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.969 0.936 (1, 1, 10, 10, 3) 1 1 [0.969]
-Total_time - 103.557 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 102.6 97.486 (1, 6, 10, 10, 1) 2 1 [102.6]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.805 1.715 (1, 6, 10, 10) 1 1 [1.805]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.841 0.799 (1, 3, 10, 10, 1) 1 1 [0.841]
+Total_time - 105.246 - - - - -
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index cc6f414a77..5dcc7538c2 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -521,7 +521,7 @@ 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/tmpgbq49kqt/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmpzq21sad0/images/random'
</pre></div>
</div>
</div>
@@ -581,8 +581,8 @@ 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], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpgbq49kqt/images/target contains 8144 images
-/tmp/tmpgbq49kqt/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpzq21sad0/images/target contains 8144 images
+/tmp/tmpzq21sad0/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -694,13 +694,13 @@ 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 - 47s - loss: 0.2382 - accuracy: 0.9208 - val_loss: 0.1159 - val_accuracy: 0.9573 - 47s/epoch - 144ms/step
+328/328 - 47s - loss: 0.2032 - accuracy: 0.9269 - val_loss: 0.1450 - val_accuracy: 0.9520 - 47s/epoch - 142ms/step
Epoch 2/3
-328/328 - 43s - loss: 0.1063 - accuracy: 0.9589 - val_loss: 0.1400 - val_accuracy: 0.9505 - 43s/epoch - 133ms/step
+328/328 - 43s - loss: 0.0946 - accuracy: 0.9623 - val_loss: 0.1637 - val_accuracy: 0.9475 - 43s/epoch - 131ms/step
Epoch 3/3
-328/328 - 43s - loss: 0.0643 - accuracy: 0.9746 - val_loss: 0.0948 - val_accuracy: 0.9690 - 43s/epoch - 132ms/step
+328/328 - 43s - loss: 0.0555 - accuracy: 0.9786 - val_loss: 0.1109 - val_accuracy: 0.9690 - 43s/epoch - 131ms/step
-<keras.callbacks.History object at 0x7fb3e0199750>
+<keras.callbacks.History object at 0x7f90cbd78cd0>
</pre></div>
</div>
</div>
@@ -962,7 +962,7 @@ 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 27.390 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 23.229 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 3f4c6c852d..04cf3facea 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -332,7 +332,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:31.738</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>05:25.689</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -341,19 +341,19 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>04:27.390</p></td>
+<td><p>04:23.229</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>00:50.894</p></td>
+<td><p>00:49.363</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_aot.html#sphx-glr-how-to-work-with-microtvm-micro-aot-py"><span class="std std-ref">microTVM Host-Driven AoT</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:09.597</p></td>
+<td><p>00:09.335</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:03.855</p></td>
+<td><p>00:03.761</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index f69ef8284d..fd9ba445af 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -332,7 +332,7 @@
<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>00:45.047</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:42.422</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -341,15 +341,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:32.795</p></td>
+<td><p>00:32.124</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>00:10.440</p></td>
+<td><p>00:08.415</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>00:01.806</p></td>
+<td><p>00:01.876</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 64dcfedbdd..808c8246e3 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -527,7 +527,7 @@ 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 0x7fb3cff18e60>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f91245228c0>
</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 d6b023beb3..013dfc120c 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -332,7 +332,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:08.092</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:07.432</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -341,27 +341,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></td>
-<td><p>00:05.613</p></td>
+<td><p>00:05.066</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>00:01.117</p></td>
+<td><p>00:01.034</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>00:00.584</p></td>
+<td><p>00:00.569</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>00:00.567</p></td>
+<td><p>00:00.554</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>00:00.117</p></td>
+<td><p>00:00.115</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.047</p></td>
+<td><p>00:00.048</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index 038d897194..ee621605ef 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -582,7 +582,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp3f0a0xyg/input0.cc'\nsource_filename = \"/tmp/tmp3f0a0xyg/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmphly3l076/input0.cc'\nsource_filename = \"/tmp/tmphly3l076/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index c44b6af556..37c500fb5c 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -229,17 +229,7 @@
<p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal" href="index.html">Installing TVM</a><ul class="current">
-<li class="toctree-l2 current"><a class="reference internal" href="from_source.html">Install from Source</a><ul class="current">
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#developers-get-source-from-github">Developers: Get Source from Github</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#build-the-shared-library">Build the Shared Library</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#python-package-installation">Python Package Installation</a></li>
-<li class="toctree-l3 current"><a class="reference internal" href="from_source.html#install-contrib-libraries">Install Contrib Libraries</a><ul class="current">
-<li class="toctree-l4 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a></li>
-</ul>
-</li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#enable-c-tests">Enable C++ Tests</a></li>
-</ul>
-</li>
+<li class="toctree-l2"><a class="reference internal" href="from_source.html">Install from Source</a></li>
<li class="toctree-l2"><a class="reference internal" href="docker.html">Docker Images</a></li>
<li class="toctree-l2 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#conditions">Conditions</a></li>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database-members.html b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database-members.html
index 8bfabf89ba..c7a058852b 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database-members.html
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database-members.html
@@ -82,8 +82,8 @@ $(function() {
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#aadbc0886ffa80162ff31eefd0431ba09">get</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#ae423057ecf93c18714d17f53cd1d318f">get_mutable</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">protected</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#aed593996e4076632450de8fde776707c">GetDataPtr</a>(const ObjectRef &ref)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">protected</span><span class="mlabel">static</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#a9eba4e789ead8b3ffd53280390660d1e">JSONDatabase</a>(String path_workload, String path_tuning_record, bool allow_missing)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#ad1c38592f69cc782314f4e1450230ace">MemoryDatabase</a>()</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#abc3b2cde0d0ef28d1fe23fbc97b22d63">JSONDatabase</a>(String path_workload, String path_tuning_record, bool allow_missing, String mod_eq_name="structural")</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#af410a88a002698bc01161aae39abf84c">MemoryDatabase</a>(String mod_eq_name="structural")</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#aa07c1f6d66a438ea950637d13ed09471">ObjectRef</a>()=default</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#a6a7dd7404edf1c26f8dbd9bd92d03a02">ObjectRef</a>(ObjectPtr< Object > data)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">explicit</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#aa1bd13a7185cb4b2b6bdde49416e8aa4">operator!=</a>(const ObjectRef &other) const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
@@ -91,9 +91,9 @@ $(function() {
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#a4744bf4a1b48f202d41b51dc5e08e6ee">operator<</a>(const ObjectRef &other) const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#affdf1b8cdb36e140de7b3ad7064e4617">operator==</a>(const ObjectRef &other) const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#ae4f6e0e951be446d2ab836eb8a9bcc83">OrderedUnionDatabase</a>(Array< Database, void > databases)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#a093004b24893fba8c6855aacd8cc46e2">PyDatabase</a>(PyDatabaseNode::FHasWorkload f_has_workload, PyDatabaseNode::FCommitWorkload f_commit_workload, PyDatabaseNode::FCommitTuningRecord f_commit_tuning_record, PyDatabaseNode::FGetTopK f_get_top_k, PyDatabaseNode::FGetAllTuningRecords f_get_all_tuning_records, PyDatabaseNode::FQueryTuningRecord f_query_tuning_record, PyDatabaseNode::FQuerySche [...]
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#aaaa9bf6e0367ae09fa8b3471a25dd50e">PyDatabase</a>(PyDatabaseNode::FHasWorkload f_has_workload, PyDatabaseNode::FCommitWorkload f_commit_workload, PyDatabaseNode::FCommitTuningRecord f_commit_tuning_record, PyDatabaseNode::FGetTopK f_get_top_k, PyDatabaseNode::FGetAllTuningRecords f_get_all_tuning_records, PyDatabaseNode::FQueryTuningRecord f_query_tuning_record, PyDatabaseNode::FQuerySche [...]
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#ae31a5b9f40781d60a2901994ead700e8">same_as</a>(const ObjectRef &other) const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#afd3cddb62e6fad7974e457b708c895a4">ScheduleFnDatabase</a>(runtime::TypedPackedFunc< bool(tir::Schedule)> schedule_fn)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#a75f998ad3493c37b4ad2a1d93f454443">ScheduleFnDatabase</a>(runtime::TypedPackedFunc< bool(tir::Schedule)> schedule_fn, String mod_eq_name="structural")</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#afb40a32e35f299ee0c6cd6f99f1ed44a">TVM_DEFINE_MUTABLE_NOTNULLABLE_OBJECT_REF_METHODS</a>(Database, runtime::ObjectRef, DatabaseNode)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#affcd4ddc4911a83f281de558366a6d63">UnionDatabase</a>(Array< Database, void > databases)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">tvm::meta_schedule::Database</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html#a4e7cdb1574b93a59e784d70aa47b8da7">unique</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1ObjectRef.html">tvm::runtime::ObjectRef</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database.html b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database.html
index 636d45ccdc..841583a495 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database.html
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1Database.html
@@ -134,23 +134,24 @@ Public Member Functions</h2></td></tr>
</table><table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-static-methods"></a>
Static Public Member Functions</h2></td></tr>
-<tr class="memitem:ad1c38592f69cc782314f4e1450230ace"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#ad1c38592f69cc782314f4e1450230ace">MemoryDatabase</a> ()</td></tr>
-<tr class="separator:ad1c38592f69cc782314f4e1450230ace"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:afd3cddb62e6fad7974e457b708c895a4"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#afd3cddb62e6fad7974e457b708c895a4">ScheduleFnDatabase</a> (<a class="el" href="classtvm_1_1runtime_1_1TypedPackedFunc.html">runtime::TypedPackedFunc</a>< bool(<a class="el" href="class [...]
-<tr class="memdesc:afd3cddb62e6fad7974e457b708c895a4"><td class="mdescLeft"> </td><td class="mdescRight">A database for injecting handcrafted schedule functions. <a href="#afd3cddb62e6fad7974e457b708c895a4">More...</a><br /></td></tr>
-<tr class="separator:afd3cddb62e6fad7974e457b708c895a4"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a9eba4e789ead8b3ffd53280390660d1e"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#a9eba4e789ead8b3ffd53280390660d1e">JSONDatabase</a> (<a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> path_workload, <a class="el" href="classtvm_1_1runtime_1_1String.ht [...]
-<tr class="memdesc:a9eba4e789ead8b3ffd53280390660d1e"><td class="mdescLeft"> </td><td class="mdescRight">Create a default database that uses JSON file for tuning records. <a href="#a9eba4e789ead8b3ffd53280390660d1e">More...</a><br /></td></tr>
-<tr class="separator:a9eba4e789ead8b3ffd53280390660d1e"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:af410a88a002698bc01161aae39abf84c"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#af410a88a002698bc01161aae39abf84c">MemoryDatabase</a> (<a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> mod_eq_name="structural")</td></tr>
+<tr class="memdesc:af410a88a002698bc01161aae39abf84c"><td class="mdescLeft"> </td><td class="mdescRight">An in-memory database. <a href="#af410a88a002698bc01161aae39abf84c">More...</a><br /></td></tr>
+<tr class="separator:af410a88a002698bc01161aae39abf84c"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a75f998ad3493c37b4ad2a1d93f454443"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#a75f998ad3493c37b4ad2a1d93f454443">ScheduleFnDatabase</a> (<a class="el" href="classtvm_1_1runtime_1_1TypedPackedFunc.html">runtime::TypedPackedFunc</a>< bool(<a class="el" href="class [...]
+<tr class="memdesc:a75f998ad3493c37b4ad2a1d93f454443"><td class="mdescLeft"> </td><td class="mdescRight">A database for injecting handcrafted schedule functions. <a href="#a75f998ad3493c37b4ad2a1d93f454443">More...</a><br /></td></tr>
+<tr class="separator:a75f998ad3493c37b4ad2a1d93f454443"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:abc3b2cde0d0ef28d1fe23fbc97b22d63"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#abc3b2cde0d0ef28d1fe23fbc97b22d63">JSONDatabase</a> (<a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> path_workload, <a class="el" href="classtvm_1_1runtime_1_1String.ht [...]
+<tr class="memdesc:abc3b2cde0d0ef28d1fe23fbc97b22d63"><td class="mdescLeft"> </td><td class="mdescRight">Create a default database that uses JSON file for tuning records. <a href="#abc3b2cde0d0ef28d1fe23fbc97b22d63">More...</a><br /></td></tr>
+<tr class="separator:abc3b2cde0d0ef28d1fe23fbc97b22d63"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:affcd4ddc4911a83f281de558366a6d63"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#affcd4ddc4911a83f281de558366a6d63">UnionDatabase</a> (<a class="el" href="classtvm_1_1runtime_1_1Array.html">Array</a>< <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html" [...]
<tr class="memdesc:affcd4ddc4911a83f281de558366a6d63"><td class="mdescLeft"> </td><td class="mdescRight">A database composed of multiple databases, allowing users to guide IR rewriting using combined knowledge of those databases. To each query, it returns the best record among all the databases given. <a href="#affcd4ddc4911a83f281de558366a6d63">More...</a><br /></td></tr>
<tr class="separator:affcd4ddc4911a83f281de558366a6d63"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:ae4f6e0e951be446d2ab836eb8a9bcc83"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#ae4f6e0e951be446d2ab836eb8a9bcc83">OrderedUnionDatabase</a> (<a class="el" href="classtvm_1_1runtime_1_1Array.html">Array</a>< <a class="el" href="classtvm_1_1meta__schedule_1_1Databas [...]
<tr class="memdesc:ae4f6e0e951be446d2ab836eb8a9bcc83"><td class="mdescLeft"> </td><td class="mdescRight">A database composed of multiple databases, allowing users to guide IR rewriting using combined knowledge of those databases. To each query, it returns the record from the first database that responds to the query. <a href="#ae4f6e0e951be446d2ab836eb8a9bcc83">More...</a><br /></td></tr>
<tr class="separator:ae4f6e0e951be446d2ab836eb8a9bcc83"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a093004b24893fba8c6855aacd8cc46e2"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#a093004b24893fba8c6855aacd8cc46e2">PyDatabase</a> (<a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#afb177ebca29227e94c3b3036b4908bad">PyDatabaseNode::FHasWorkload</a [...]
-<tr class="memdesc:a093004b24893fba8c6855aacd8cc46e2"><td class="mdescLeft"> </td><td class="mdescRight">Create a database with customized methods on the python-side. <a href="#a093004b24893fba8c6855aacd8cc46e2">More...</a><br /></td></tr>
-<tr class="separator:a093004b24893fba8c6855aacd8cc46e2"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:aaaa9bf6e0367ae09fa8b3471a25dd50e"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#aaaa9bf6e0367ae09fa8b3471a25dd50e">PyDatabase</a> (<a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#afb177ebca29227e94c3b3036b4908bad">PyDatabaseNode::FHasWorkload</a [...]
+<tr class="memdesc:aaaa9bf6e0367ae09fa8b3471a25dd50e"><td class="mdescLeft"> </td><td class="mdescRight">Create a database with customized methods on the python-side. <a href="#aaaa9bf6e0367ae09fa8b3471a25dd50e">More...</a><br /></td></tr>
+<tr class="separator:aaaa9bf6e0367ae09fa8b3471a25dd50e"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a4b338c39afa925bc556b067b333e27a0"><td class="memItemLeft" align="right" valign="top">static <a class="el" href="classtvm_1_1runtime_1_1Optional.html">Optional</a>< <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> > </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1Database.html#a4b338c39afa925bc556b067b333e27a0">Current</a> ()</td></tr>
<tr class="separator:a4b338c39afa925bc556b067b333e27a0"><td class="memSeparator" colspan="2"> </td></tr>
</table><table class="memberdecls">
@@ -251,8 +252,8 @@ Additional Inherited Members</h2></td></tr>
</div>
</div>
-<a id="a9eba4e789ead8b3ffd53280390660d1e"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a9eba4e789ead8b3ffd53280390660d1e">◆ </a></span>JSONDatabase()</h2>
+<a id="abc3b2cde0d0ef28d1fe23fbc97b22d63"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#abc3b2cde0d0ef28d1fe23fbc97b22d63">◆ </a></span>JSONDatabase()</h2>
<div class="memitem">
<div class="memproto">
@@ -276,7 +277,13 @@ Additional Inherited Members</h2></td></tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype">bool </td>
- <td class="paramname"><em>allow_missing</em> </td>
+ <td class="paramname"><em>allow_missing</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype"><a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> </td>
+ <td class="paramname"><em>mod_eq_name</em> = <code>"structural"</code> </td>
</tr>
<tr>
<td></td>
@@ -297,14 +304,15 @@ Additional Inherited Members</h2></td></tr>
<tr><td class="paramname">path_workload</td><td>The path to the workload table. </td></tr>
<tr><td class="paramname">path_tuning_record</td><td>The path to the database table. </td></tr>
<tr><td class="paramname">allow_missing</td><td>Whether to create new file when the given path is not found. </td></tr>
+ <tr><td class="paramname">mod_eq_name</td><td>A string to specify the module equality testing and hashing method. </td></tr>
</table>
</dd>
</dl>
</div>
</div>
-<a id="ad1c38592f69cc782314f4e1450230ace"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#ad1c38592f69cc782314f4e1450230ace">◆ </a></span>MemoryDatabase()</h2>
+<a id="af410a88a002698bc01161aae39abf84c"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#af410a88a002698bc01161aae39abf84c">◆ </a></span>MemoryDatabase()</h2>
<div class="memitem">
<div class="memproto">
@@ -315,7 +323,8 @@ Additional Inherited Members</h2></td></tr>
<tr>
<td class="memname">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> tvm::meta_schedule::Database::MemoryDatabase </td>
<td>(</td>
- <td class="paramname"></td><td>)</td>
+ <td class="paramtype"><a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> </td>
+ <td class="paramname"><em>mod_eq_name</em> = <code>"structural"</code></td><td>)</td>
<td></td>
</tr>
</table>
@@ -325,7 +334,14 @@ Additional Inherited Members</h2></td></tr>
</tr>
</table>
</div><div class="memdoc">
+
<p>An in-memory database. </p>
+<dl class="params"><dt>Parameters</dt><dd>
+ <table class="params">
+ <tr><td class="paramname">mod_eq_name</td><td>A string to specify the module equality testing and hashing method. </td></tr>
+ </table>
+ </dd>
+</dl>
</div>
</div>
@@ -364,8 +380,8 @@ Additional Inherited Members</h2></td></tr>
</div>
</div>
-<a id="a093004b24893fba8c6855aacd8cc46e2"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a093004b24893fba8c6855aacd8cc46e2">◆ </a></span>PyDatabase()</h2>
+<a id="aaaa9bf6e0367ae09fa8b3471a25dd50e"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#aaaa9bf6e0367ae09fa8b3471a25dd50e">◆ </a></span>PyDatabase()</h2>
<div class="memitem">
<div class="memproto">
@@ -425,7 +441,13 @@ Additional Inherited Members</h2></td></tr>
<td class="paramkey"></td>
<td></td>
<td class="paramtype"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a34efc3d18473d179b13332abe5c63324">PyDatabaseNode::FSize</a> </td>
- <td class="paramname"><em>f_size</em> </td>
+ <td class="paramname"><em>f_size</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype"><a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> </td>
+ <td class="paramname"><em>mod_eq_name</em> = <code>"structural"</code> </td>
</tr>
<tr>
<td></td>
@@ -452,6 +474,7 @@ Additional Inherited Members</h2></td></tr>
<tr><td class="paramname">f_query_schedule</td><td>The packed function of <code>QuerySchedule</code>. </td></tr>
<tr><td class="paramname">f_query_ir_module</td><td>The packed function of <code>QueryIRModule</code>. </td></tr>
<tr><td class="paramname">f_size</td><td>The packed function of <code>Size</code>. </td></tr>
+ <tr><td class="paramname">mod_eq_name</td><td>A string to specify the module equality testing and hashing method. </td></tr>
</table>
</dd>
</dl>
@@ -459,8 +482,8 @@ Additional Inherited Members</h2></td></tr>
</div>
</div>
-<a id="afd3cddb62e6fad7974e457b708c895a4"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#afd3cddb62e6fad7974e457b708c895a4">◆ </a></span>ScheduleFnDatabase()</h2>
+<a id="a75f998ad3493c37b4ad2a1d93f454443"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a75f998ad3493c37b4ad2a1d93f454443">◆ </a></span>ScheduleFnDatabase()</h2>
<div class="memitem">
<div class="memproto">
@@ -472,8 +495,18 @@ Additional Inherited Members</h2></td></tr>
<td class="memname">static <a class="el" href="classtvm_1_1meta__schedule_1_1Database.html">Database</a> tvm::meta_schedule::Database::ScheduleFnDatabase </td>
<td>(</td>
<td class="paramtype"><a class="el" href="classtvm_1_1runtime_1_1TypedPackedFunc.html">runtime::TypedPackedFunc</a>< bool(<a class="el" href="classtvm_1_1tir_1_1Schedule.html">tir::Schedule</a>)> </td>
- <td class="paramname"><em>schedule_fn</em></td><td>)</td>
+ <td class="paramname"><em>schedule_fn</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
<td></td>
+ <td class="paramtype"><a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> </td>
+ <td class="paramname"><em>mod_eq_name</em> = <code>"structural"</code> </td>
+ </tr>
+ <tr>
+ <td></td>
+ <td>)</td>
+ <td></td><td></td>
</tr>
</table>
</td>
@@ -487,6 +520,7 @@ Additional Inherited Members</h2></td></tr>
<dl class="params"><dt>Parameters</dt><dd>
<table class="params">
<tr><td class="paramname">schedule_fn</td><td>The function to do scheduling, which takes a TIR schedule, and returns a boolean indicating if the schedule is successful. </td></tr>
+ <tr><td class="paramname">mod_eq_name</td><td>A string to specify the module equality testing and hashing method. </td></tr>
</table>
</dd>
</dl>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode-members.html b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode-members.html
index a2bd828e82..1a03d6e8da 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode-members.html
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode-members.html
@@ -80,10 +80,12 @@ $(function() {
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#af9d87608cd8203220feb6bc032b00729">_type_key</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a391ebf6b8bd4d6a1e564ad807f336fd4">CommitTuningRecord</a>(const TuningRecord &record)=0</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">pure virtual</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a833752c0069eb976028425751df6e9ac">CommitWorkload</a>(const IRModule &mod)=0</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">pure virtual</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a70fb5361147634605d6595bb89381f03">DecRef</a>()</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">protected</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#af4407d2b59132e803ff791482dbe0145">deleter_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a9e84841ca982bff376a978ade0132631">FDeleter</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#aa0b963a127d514ff78295b4f75c88872">GetAllTuningRecords</a>()=0</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">pure virtual</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#adfd6210675fcb9dd7778d77b8ce56782">DatabaseNode</a>(String mod_eq_name="structural")</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">explicit</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a70fb5361147634605d6595bb89381f03">DecRef</a>()</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">protected</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#af4407d2b59132e803ff791482dbe0145">deleter_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a9e84841ca982bff376a978ade0132631">FDeleter</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#aa0b963a127d514ff78295b4f75c88872">GetAllTuningRecords</a>()=0</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">pure virtual</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a491ad91dbd0a811712c5013bade724cb">GetModuleEquality</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a726972ff315c446192df94027ddea032">GetOrAllocRuntimeTypeIndex</a>(const std::string &key, uint32_t static_tindex, uint32_t parent_tindex, uint32_t type_child_slots, bool type_child_slots_can_overflow)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span><span class="mlabel">static</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a27a9519109fd970572be75bc277a1fb2">GetTopK</a>(const Workload &workload, int top_k)=0</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">pure virtual</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a4d951e51832081b85875669eac90e940">GetTypeKey</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
@@ -110,7 +112,7 @@ $(function() {
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a6ee32a02dd44257da105fbbe5d9c8622">TypeIndex2KeyHash</a>(uint32_t tindex)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a6841f97e06e6614dd7e82c6dd41b818a">TypeKey2Index</a>(const std::string &key)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#afd548730a6139d19fe24473ad66026d7">unique</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a776359f44ac6b51e337d4a1efc3f04a9">~DatabaseNode</a>()=default</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">virtual</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a48707424816f7ad0fd6f59a7111d9a09">~DatabaseNode</a>()</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">virtual</span></td></tr>
</table></div><!-- contents -->
<!-- start footer part -->
<hr class="footer"/><address class="footer"><small>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode.html b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode.html
index a08b1d2e8c..f17154b1e2 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode.html
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode.html
@@ -75,21 +75,24 @@ $(function() {
<div class="dynheader">
Inheritance diagram for tvm::meta_schedule::DatabaseNode:</div>
<div class="dyncontent">
-<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1DatabaseNode__inherit__graph.svg" width="290" height="1248"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
+<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1DatabaseNode__inherit__graph.svg" width="290" height="1292"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
</div>
</div>
<div class="dynheader">
Collaboration diagram for tvm::meta_schedule::DatabaseNode:</div>
<div class="dyncontent">
-<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1DatabaseNode__coll__graph.svg" width="364" height="816"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
+<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1DatabaseNode__coll__graph.svg" width="364" height="846"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
</div>
</div>
<table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-methods"></a>
Public Member Functions</h2></td></tr>
-<tr class="memitem:a776359f44ac6b51e337d4a1efc3f04a9"><td class="memItemLeft" align="right" valign="top">virtual </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a776359f44ac6b51e337d4a1efc3f04a9">~DatabaseNode</a> ()=default</td></tr>
-<tr class="memdesc:a776359f44ac6b51e337d4a1efc3f04a9"><td class="mdescLeft"> </td><td class="mdescRight">Default destructor. <a href="#a776359f44ac6b51e337d4a1efc3f04a9">More...</a><br /></td></tr>
-<tr class="separator:a776359f44ac6b51e337d4a1efc3f04a9"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:adfd6210675fcb9dd7778d77b8ce56782"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#adfd6210675fcb9dd7778d77b8ce56782">DatabaseNode</a> (<a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> mod_eq_name="structural")</td></tr>
+<tr class="memdesc:adfd6210675fcb9dd7778d77b8ce56782"><td class="mdescLeft"> </td><td class="mdescRight"><a class="el" href="classtvm_1_1Constructor.html" title="Managed reference to ConstructorNode. ">Constructor</a>. <a href="#adfd6210675fcb9dd7778d77b8ce56782">More...</a><br /></td></tr>
+<tr class="separator:adfd6210675fcb9dd7778d77b8ce56782"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a48707424816f7ad0fd6f59a7111d9a09"><td class="memItemLeft" align="right" valign="top">virtual </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a48707424816f7ad0fd6f59a7111d9a09">~DatabaseNode</a> ()</td></tr>
+<tr class="memdesc:a48707424816f7ad0fd6f59a7111d9a09"><td class="mdescLeft"> </td><td class="mdescRight">Default destructor. <a href="#a48707424816f7ad0fd6f59a7111d9a09">More...</a><br /></td></tr>
+<tr class="separator:a48707424816f7ad0fd6f59a7111d9a09"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a04b2ddf6acb509d5cc848c8636f9619d"><td class="memItemLeft" align="right" valign="top">virtual bool </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a04b2ddf6acb509d5cc848c8636f9619d">HasWorkload</a> (const <a class="el" href="classtvm_1_1IRModule.html">IRModule</a> &mod)=0</td></tr>
<tr class="memdesc:a04b2ddf6acb509d5cc848c8636f9619d"><td class="mdescLeft"> </td><td class="mdescRight">Check if the database has the given workload. <a href="#a04b2ddf6acb509d5cc848c8636f9619d">More...</a><br /></td></tr>
<tr class="separator:a04b2ddf6acb509d5cc848c8636f9619d"><td class="memSeparator" colspan="2"> </td></tr>
@@ -117,6 +120,9 @@ Public Member Functions</h2></td></tr>
<tr class="memitem:aeb4101db551afa93ea144b9b173783a0"><td class="memItemLeft" align="right" valign="top">virtual <a class="el" href="classtvm_1_1runtime_1_1Optional.html">Optional</a>< <a class="el" href="classtvm_1_1IRModule.html">IRModule</a> > </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#aeb4101db551afa93ea144b9b173783a0">QueryIRModule</a> (const <a class="el" href="classtvm_1_1IRModule.html">IRModule</a> [...]
<tr class="memdesc:aeb4101db551afa93ea144b9b173783a0"><td class="mdescLeft"> </td><td class="mdescRight">Query the best <a class="el" href="classtvm_1_1IRModule.html" title="Managed reference class to IRModuleNode. ">IRModule</a> of the given workload from the database. <a href="#aeb4101db551afa93ea144b9b173783a0">More...</a><br /></td></tr>
<tr class="separator:aeb4101db551afa93ea144b9b173783a0"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a491ad91dbd0a811712c5013bade724cb"><td class="memItemLeft" align="right" valign="top">const ModuleEquality & </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a491ad91dbd0a811712c5013bade724cb">GetModuleEquality</a> () const</td></tr>
+<tr class="memdesc:a491ad91dbd0a811712c5013bade724cb"><td class="mdescLeft"> </td><td class="mdescRight">Return a reference to the owned module equality method instance. <a href="#a491ad91dbd0a811712c5013bade724cb">More...</a><br /></td></tr>
+<tr class="separator:a491ad91dbd0a811712c5013bade724cb"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:ae7ea55bfa3703dfb6452573afc31a45e"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#ae7ea55bfa3703dfb6452573afc31a45e">TVM_DECLARE_BASE_OBJECT_INFO</a> (<a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">DatabaseNode</a>, <a class="el" href="classtvm_1_1runtime_1_1Object.html">runtime::Object</a>)</td></tr>
<tr class="separator:ae7ea55bfa3703dfb6452573afc31a45e"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="inherit_header pub_methods_classtvm_1_1runtime_1_1Object"><td colspan="2" onclick="javascript:toggleInherit('pub_methods_classtvm_1_1runtime_1_1Object')"><img src="closed.png" alt="-"/> Public Member Functions inherited from <a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td></tr>
@@ -209,8 +215,45 @@ Additional Inherited Members</h2></td></tr>
<tr class="separator:af4407d2b59132e803ff791482dbe0145 inherit pro_attribs_classtvm_1_1runtime_1_1Object"><td class="memSeparator" colspan="2"> </td></tr>
</table>
<h2 class="groupheader">Constructor & Destructor Documentation</h2>
-<a id="a776359f44ac6b51e337d4a1efc3f04a9"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a776359f44ac6b51e337d4a1efc3f04a9">◆ </a></span>~DatabaseNode()</h2>
+<a id="adfd6210675fcb9dd7778d77b8ce56782"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#adfd6210675fcb9dd7778d77b8ce56782">◆ </a></span>DatabaseNode()</h2>
+
+<div class="memitem">
+<div class="memproto">
+<table class="mlabels">
+ <tr>
+ <td class="mlabels-left">
+ <table class="memname">
+ <tr>
+ <td class="memname">tvm::meta_schedule::DatabaseNode::DatabaseNode </td>
+ <td>(</td>
+ <td class="paramtype"><a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> </td>
+ <td class="paramname"><em>mod_eq_name</em> = <code>"structural"</code></td><td>)</td>
+ <td></td>
+ </tr>
+ </table>
+ </td>
+ <td class="mlabels-right">
+<span class="mlabels"><span class="mlabel">explicit</span></span> </td>
+ </tr>
+</table>
+</div><div class="memdoc">
+
+<p><a class="el" href="classtvm_1_1Constructor.html" title="Managed reference to ConstructorNode. ">Constructor</a>. </p>
+<dl class="params"><dt>Parameters</dt><dd>
+ <table class="params">
+ <tr><td class="paramname">mod_eq_name</td><td>A string to specify the module equality testing and hashing method. It must be one of the followings:<ul>
+<li>"structural": Use StructuralEqual/Hash </li>
+</ul>
+</td></tr>
+ </table>
+ </dd>
+</dl>
+
+</div>
+</div>
+<a id="a48707424816f7ad0fd6f59a7111d9a09"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a48707424816f7ad0fd6f59a7111d9a09">◆ </a></span>~DatabaseNode()</h2>
<div class="memitem">
<div class="memproto">
@@ -227,7 +270,7 @@ Additional Inherited Members</h2></td></tr>
</table>
</td>
<td class="mlabels-right">
-<span class="mlabels"><span class="mlabel">virtual</span><span class="mlabel">default</span></span> </td>
+<span class="mlabels"><span class="mlabel">virtual</span></span> </td>
</tr>
</table>
</div><div class="memdoc">
@@ -338,6 +381,33 @@ Additional Inherited Members</h2></td></tr>
<p>Implemented in <a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad07d7d9e78771eaa2e6e65f84e032401">tvm::meta_schedule::PyDatabaseNode</a>.</p>
+</div>
+</div>
+<a id="a491ad91dbd0a811712c5013bade724cb"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a491ad91dbd0a811712c5013bade724cb">◆ </a></span>GetModuleEquality()</h2>
+
+<div class="memitem">
+<div class="memproto">
+<table class="mlabels">
+ <tr>
+ <td class="mlabels-left">
+ <table class="memname">
+ <tr>
+ <td class="memname">const ModuleEquality& tvm::meta_schedule::DatabaseNode::GetModuleEquality </td>
+ <td>(</td>
+ <td class="paramname"></td><td>)</td>
+ <td> const</td>
+ </tr>
+ </table>
+ </td>
+ <td class="mlabels-right">
+<span class="mlabels"><span class="mlabel">inline</span></span> </td>
+ </tr>
+</table>
+</div><div class="memdoc">
+
+<p>Return a reference to the owned module equality method instance. </p>
+
</div>
</div>
<a id="a27a9519109fd970572be75bc277a1fb2"></a>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__coll__graph.svg b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__coll__graph.svg
index aa74f47728..fd3713d66c 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__coll__graph.svg
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__coll__graph.svg
@@ -4,87 +4,89 @@
<!-- Generated by graphviz version 2.40.1 (20161225.0304)
-->
<!-- Title: tvm::meta_schedule::DatabaseNode Pages: 1 -->
-<svg width="273pt" height="612pt"
- viewBox="0.00 0.00 273.00 612.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
-<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 608)">
+<svg width="273pt" height="634pt"
+ viewBox="0.00 0.00 273.00 634.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 630)">
<title>tvm::meta_schedule::DatabaseNode</title>
-<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-608 269,-608 269,4 -4,4"/>
+<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-630 269,-630 269,4 -4,4"/>
<!-- Node3 -->
<g id="node1" class="node">
<title>Node3</title>
-<polygon fill="#bfbfbf" stroke="#000000" points="0,-.5 0,-178.5 207,-178.5 207,-.5 0,-.5"/>
-<text text-anchor="start" x="8" y="-166.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
-<text text-anchor="middle" x="103.5" y="-155.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::DatabaseNode</text>
-<polyline fill="none" stroke="#000000" points="0,-148.5 207,-148.5 "/>
-<text text-anchor="start" x="8" y="-136.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<polyline fill="none" stroke="#000000" points="0,-129.5 207,-129.5 "/>
-<text text-anchor="start" x="8" y="-117.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ ~DatabaseNode()</text>
-<text text-anchor="start" x="8" y="-106.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
-<text text-anchor="start" x="8" y="-95.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
-<text text-anchor="start" x="8" y="-84.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitTuningRecord()</text>
-<text text-anchor="start" x="8" y="-73.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTopK()</text>
-<text text-anchor="start" x="8" y="-62.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetAllTuningRecords()</text>
-<text text-anchor="start" x="8" y="-51.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Size()</text>
-<text text-anchor="start" x="8" y="-40.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryTuningRecord()</text>
-<text text-anchor="start" x="8" y="-29.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QuerySchedule()</text>
-<text text-anchor="start" x="8" y="-18.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryIRModule()</text>
+<polygon fill="#bfbfbf" stroke="#000000" points="0,-.5 0,-200.5 207,-200.5 207,-.5 0,-.5"/>
+<text text-anchor="start" x="8" y="-188.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
+<text text-anchor="middle" x="103.5" y="-177.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::DatabaseNode</text>
+<polyline fill="none" stroke="#000000" points="0,-170.5 207,-170.5 "/>
+<text text-anchor="start" x="8" y="-158.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<polyline fill="none" stroke="#000000" points="0,-151.5 207,-151.5 "/>
+<text text-anchor="start" x="8" y="-139.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ DatabaseNode()</text>
+<text text-anchor="start" x="8" y="-128.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ ~DatabaseNode()</text>
+<text text-anchor="start" x="8" y="-117.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
+<text text-anchor="start" x="8" y="-106.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
+<text text-anchor="start" x="8" y="-95.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitTuningRecord()</text>
+<text text-anchor="start" x="8" y="-84.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTopK()</text>
+<text text-anchor="start" x="8" y="-73.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetAllTuningRecords()</text>
+<text text-anchor="start" x="8" y="-62.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Size()</text>
+<text text-anchor="start" x="8" y="-51.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryTuningRecord()</text>
+<text text-anchor="start" x="8" y="-40.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QuerySchedule()</text>
+<text text-anchor="start" x="8" y="-29.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryIRModule()</text>
+<text text-anchor="start" x="8" y="-18.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetModuleEquality()</text>
<text text-anchor="start" x="8" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TVM_DECLARE_BASE_OBJECT_INFO()</text>
</g>
<!-- Node4 -->
<g id="node2" class="node">
<title>Node4</title>
<g id="a_node2"><a xlink:href="classtvm_1_1runtime_1_1Object.html" target="_top" xlink:title="base class of all object containers. ">
-<polygon fill="#ffffff" stroke="#000000" points="12,-216.5 12,-603.5 195,-603.5 195,-216.5 12,-216.5"/>
-<text text-anchor="middle" x="103.5" y="-591.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::Object</text>
-<polyline fill="none" stroke="#000000" points="12,-584.5 195,-584.5 "/>
-<text text-anchor="start" x="20" y="-572.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<text text-anchor="start" x="20" y="-561.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_final</text>
-<text text-anchor="start" x="20" y="-550.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots</text>
-<text text-anchor="start" x="20" y="-539.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots_can</text>
-<text text-anchor="start" x="20" y="-528.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_overflow</text>
-<text text-anchor="start" x="20" y="-517.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_visit</text>
-<text text-anchor="start" x="20" y="-506.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_attrs</text>
-<text text-anchor="start" x="20" y="-495.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_sequal</text>
+<polygon fill="#ffffff" stroke="#000000" points="12,-238.5 12,-625.5 195,-625.5 195,-238.5 12,-238.5"/>
+<text text-anchor="middle" x="103.5" y="-613.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::Object</text>
+<polyline fill="none" stroke="#000000" points="12,-606.5 195,-606.5 "/>
+<text text-anchor="start" x="20" y="-594.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<text text-anchor="start" x="20" y="-583.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_final</text>
+<text text-anchor="start" x="20" y="-572.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots</text>
+<text text-anchor="start" x="20" y="-561.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots_can</text>
+<text text-anchor="start" x="20" y="-550.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_overflow</text>
+<text text-anchor="start" x="20" y="-539.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_visit</text>
+<text text-anchor="start" x="20" y="-528.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_attrs</text>
+<text text-anchor="start" x="20" y="-517.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_sequal</text>
+<text text-anchor="start" x="20" y="-506.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
+<text text-anchor="start" x="20" y="-495.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_shash</text>
<text text-anchor="start" x="20" y="-484.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
-<text text-anchor="start" x="20" y="-473.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_shash</text>
-<text text-anchor="start" x="20" y="-462.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
-<text text-anchor="start" x="20" y="-451.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_index</text>
-<text text-anchor="start" x="20" y="-440.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># type_index_</text>
-<text text-anchor="start" x="20" y="-429.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># ref_counter_</text>
-<polyline fill="none" stroke="#000000" points="12,-422.5 195,-422.5 "/>
-<text text-anchor="start" x="20" y="-410.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ type_index()</text>
-<text text-anchor="start" x="20" y="-399.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKey()</text>
-<text text-anchor="start" x="20" y="-388.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKeyHash()</text>
-<text text-anchor="start" x="20" y="-377.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ IsInstance()</text>
-<text text-anchor="start" x="20" y="-366.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ unique()</text>
+<text text-anchor="start" x="20" y="-473.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_index</text>
+<text text-anchor="start" x="20" y="-462.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># type_index_</text>
+<text text-anchor="start" x="20" y="-451.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># ref_counter_</text>
+<polyline fill="none" stroke="#000000" points="12,-444.5 195,-444.5 "/>
+<text text-anchor="start" x="20" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ type_index()</text>
+<text text-anchor="start" x="20" y="-421.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKey()</text>
+<text text-anchor="start" x="20" y="-410.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKeyHash()</text>
+<text text-anchor="start" x="20" y="-399.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ IsInstance()</text>
+<text text-anchor="start" x="20" y="-388.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ unique()</text>
+<text text-anchor="start" x="20" y="-377.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
+<text text-anchor="start" x="20" y="-366.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
<text text-anchor="start" x="20" y="-355.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="20" y="-344.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="20" y="-333.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="20" y="-322.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
-<text text-anchor="start" x="20" y="-311.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
-<text text-anchor="start" x="20" y="-300.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2Key()</text>
-<text text-anchor="start" x="20" y="-289.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2KeyHash()</text>
-<text text-anchor="start" x="20" y="-278.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeKey2Index()</text>
-<text text-anchor="start" x="20" y="-267.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _GetOrAllocRuntimeTypeIndex()</text>
-<text text-anchor="start" x="20" y="-256.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ RuntimeTypeIndex()</text>
-<text text-anchor="start" x="20" y="-245.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># IncRef()</text>
-<text text-anchor="start" x="20" y="-234.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># DecRef()</text>
-<text text-anchor="start" x="20" y="-223.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># GetOrAllocRuntimeTypeIndex()</text>
+<text text-anchor="start" x="20" y="-344.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
+<text text-anchor="start" x="20" y="-333.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
+<text text-anchor="start" x="20" y="-322.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2Key()</text>
+<text text-anchor="start" x="20" y="-311.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2KeyHash()</text>
+<text text-anchor="start" x="20" y="-300.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeKey2Index()</text>
+<text text-anchor="start" x="20" y="-289.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _GetOrAllocRuntimeTypeIndex()</text>
+<text text-anchor="start" x="20" y="-278.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ RuntimeTypeIndex()</text>
+<text text-anchor="start" x="20" y="-267.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># IncRef()</text>
+<text text-anchor="start" x="20" y="-256.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># DecRef()</text>
+<text text-anchor="start" x="20" y="-245.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># GetOrAllocRuntimeTypeIndex()</text>
</a>
</g>
</g>
<!-- Node4->Node3 -->
<g id="edge1" class="edge">
<title>Node4->Node3</title>
-<path fill="none" stroke="#191970" d="M103.5,-205.9464C103.5,-196.5963 103.5,-187.4618 103.5,-178.684"/>
-<polygon fill="none" stroke="#191970" points="100.0001,-206.1701 103.5,-216.1701 107.0001,-206.1701 100.0001,-206.1701"/>
+<path fill="none" stroke="#191970" d="M103.5,-227.9868C103.5,-218.6259 103.5,-209.4373 103.5,-200.5493"/>
+<polygon fill="none" stroke="#191970" points="100.0001,-228.2059 103.5,-238.2059 107.0001,-228.206 100.0001,-228.2059"/>
</g>
<!-- Node4->Node4 -->
<g id="edge2" class="edge">
<title>Node4->Node4</title>
-<path fill="none" stroke="#404040" d="M195.3625,-443.9248C206.0482,-437.6637 213,-426.3555 213,-410 213,-399.0112 209.8618,-390.3007 204.5615,-383.8687"/>
-<polygon fill="none" stroke="#404040" points="204.5184,-383.8322 197.3548,-383.0056 195.3625,-376.0752 202.5261,-376.9017 204.5184,-383.8322"/>
-<text text-anchor="middle" x="239" y="-407.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> #deleter_</text>
+<path fill="none" stroke="#404040" d="M195.3625,-465.9248C206.0482,-459.6637 213,-448.3555 213,-432 213,-421.0112 209.8618,-412.3007 204.5615,-405.8687"/>
+<polygon fill="none" stroke="#404040" points="204.5184,-405.8322 197.3548,-405.0056 195.3625,-398.0752 202.5261,-398.9017 204.5184,-405.8322"/>
+<text text-anchor="middle" x="239" y="-429.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> #deleter_</text>
</g>
</g>
</svg>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__inherit__graph.svg b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__inherit__graph.svg
index 781234be73..9f9d308bd5 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__inherit__graph.svg
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1DatabaseNode__inherit__graph.svg
@@ -4,51 +4,54 @@
<!-- Generated by graphviz version 2.40.1 (20161225.0304)
-->
<!-- Title: tvm::meta_schedule::DatabaseNode Pages: 1 -->
-<svg width="217pt" height="936pt"
- viewBox="0.00 0.00 217.00 936.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
-<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 932)">
+<svg width="217pt" height="969pt"
+ viewBox="0.00 0.00 217.00 969.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 965)">
<title>tvm::meta_schedule::DatabaseNode</title>
-<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-932 213,-932 213,4 -4,4"/>
+<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-965 213,-965 213,4 -4,4"/>
<!-- Node0 -->
<g id="node1" class="node">
<title>Node0</title>
-<polygon fill="#bfbfbf" stroke="#000000" points="1,-314.5 1,-492.5 208,-492.5 208,-314.5 1,-314.5"/>
-<text text-anchor="start" x="9" y="-480.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
-<text text-anchor="middle" x="104.5" y="-469.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::DatabaseNode</text>
-<polyline fill="none" stroke="#000000" points="1,-462.5 208,-462.5 "/>
-<text text-anchor="start" x="9" y="-450.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<polyline fill="none" stroke="#000000" points="1,-443.5 208,-443.5 "/>
-<text text-anchor="start" x="9" y="-431.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ ~DatabaseNode()</text>
-<text text-anchor="start" x="9" y="-420.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
-<text text-anchor="start" x="9" y="-409.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
-<text text-anchor="start" x="9" y="-398.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitTuningRecord()</text>
-<text text-anchor="start" x="9" y="-387.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTopK()</text>
-<text text-anchor="start" x="9" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetAllTuningRecords()</text>
-<text text-anchor="start" x="9" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Size()</text>
-<text text-anchor="start" x="9" y="-354.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryTuningRecord()</text>
-<text text-anchor="start" x="9" y="-343.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QuerySchedule()</text>
-<text text-anchor="start" x="9" y="-332.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryIRModule()</text>
-<text text-anchor="start" x="9" y="-321.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TVM_DECLARE_BASE_OBJECT_INFO()</text>
+<polygon fill="#bfbfbf" stroke="#000000" points="1,-325.5 1,-525.5 208,-525.5 208,-325.5 1,-325.5"/>
+<text text-anchor="start" x="9" y="-513.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
+<text text-anchor="middle" x="104.5" y="-502.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::DatabaseNode</text>
+<polyline fill="none" stroke="#000000" points="1,-495.5 208,-495.5 "/>
+<text text-anchor="start" x="9" y="-483.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<polyline fill="none" stroke="#000000" points="1,-476.5 208,-476.5 "/>
+<text text-anchor="start" x="9" y="-464.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ DatabaseNode()</text>
+<text text-anchor="start" x="9" y="-453.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ ~DatabaseNode()</text>
+<text text-anchor="start" x="9" y="-442.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
+<text text-anchor="start" x="9" y="-431.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
+<text text-anchor="start" x="9" y="-420.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitTuningRecord()</text>
+<text text-anchor="start" x="9" y="-409.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTopK()</text>
+<text text-anchor="start" x="9" y="-398.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetAllTuningRecords()</text>
+<text text-anchor="start" x="9" y="-387.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Size()</text>
+<text text-anchor="start" x="9" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryTuningRecord()</text>
+<text text-anchor="start" x="9" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QuerySchedule()</text>
+<text text-anchor="start" x="9" y="-354.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryIRModule()</text>
+<text text-anchor="start" x="9" y="-343.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetModuleEquality()</text>
+<text text-anchor="start" x="9" y="-332.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TVM_DECLARE_BASE_OBJECT_INFO()</text>
</g>
<!-- Node2 -->
<g id="node3" class="node">
<title>Node2</title>
<g id="a_node3"><a xlink:href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html" target="_top" xlink:title="The database with customized methods on the python-side. ">
-<polygon fill="#ffffff" stroke="#000000" points="0,-.5 0,-277.5 209,-277.5 209,-.5 0,-.5"/>
-<text text-anchor="start" x="8" y="-265.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
-<text text-anchor="middle" x="104.5" y="-254.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::PyDatabaseNode</text>
-<polyline fill="none" stroke="#000000" points="0,-247.5 209,-247.5 "/>
-<text text-anchor="start" x="8" y="-235.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_has_workload</text>
-<text text-anchor="start" x="8" y="-224.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_commit_workload</text>
-<text text-anchor="start" x="8" y="-213.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_commit_tuning_record</text>
-<text text-anchor="start" x="8" y="-202.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_get_top_k</text>
-<text text-anchor="start" x="8" y="-191.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_get_all_tuning_records</text>
-<text text-anchor="start" x="8" y="-180.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_query_tuning_record</text>
-<text text-anchor="start" x="8" y="-169.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_query_schedule</text>
-<text text-anchor="start" x="8" y="-158.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_query_ir_module</text>
-<text text-anchor="start" x="8" y="-147.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_size</text>
-<text text-anchor="start" x="8" y="-136.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<polyline fill="none" stroke="#000000" points="0,-129.5 209,-129.5 "/>
+<polygon fill="#ffffff" stroke="#000000" points="0,-.5 0,-288.5 209,-288.5 209,-.5 0,-.5"/>
+<text text-anchor="start" x="8" y="-276.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
+<text text-anchor="middle" x="104.5" y="-265.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::PyDatabaseNode</text>
+<polyline fill="none" stroke="#000000" points="0,-258.5 209,-258.5 "/>
+<text text-anchor="start" x="8" y="-246.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_has_workload</text>
+<text text-anchor="start" x="8" y="-235.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_commit_workload</text>
+<text text-anchor="start" x="8" y="-224.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_commit_tuning_record</text>
+<text text-anchor="start" x="8" y="-213.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_get_top_k</text>
+<text text-anchor="start" x="8" y="-202.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_get_all_tuning_records</text>
+<text text-anchor="start" x="8" y="-191.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_query_tuning_record</text>
+<text text-anchor="start" x="8" y="-180.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_query_schedule</text>
+<text text-anchor="start" x="8" y="-169.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_query_ir_module</text>
+<text text-anchor="start" x="8" y="-158.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ f_size</text>
+<text text-anchor="start" x="8" y="-147.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<polyline fill="none" stroke="#000000" points="0,-140.5 209,-140.5 "/>
+<text text-anchor="start" x="8" y="-128.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ PyDatabaseNode()</text>
<text text-anchor="start" x="8" y="-117.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ VisitAttrs()</text>
<text text-anchor="start" x="8" y="-106.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
<text text-anchor="start" x="8" y="-95.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
@@ -66,58 +69,58 @@
<!-- Node0->Node2 -->
<g id="edge2" class="edge">
<title>Node0->Node2</title>
-<path fill="none" stroke="#191970" d="M104.5,-304.2113C104.5,-295.5113 104.5,-286.6081 104.5,-277.6657"/>
-<polygon fill="none" stroke="#191970" points="101.0001,-304.3211 104.5,-314.3211 108.0001,-304.3211 101.0001,-304.3211"/>
+<path fill="none" stroke="#191970" d="M104.5,-315.1784C104.5,-306.5354 104.5,-297.7326 104.5,-288.9091"/>
+<polygon fill="none" stroke="#191970" points="101.0001,-315.2426 104.5,-325.2426 108.0001,-315.2427 101.0001,-315.2426"/>
</g>
<!-- Node1 -->
<g id="node2" class="node">
<title>Node1</title>
<g id="a_node2"><a xlink:href="classtvm_1_1runtime_1_1Object.html" target="_top" xlink:title="base class of all object containers. ">
-<polygon fill="#ffffff" stroke="#000000" points="13,-529.5 13,-927.5 196,-927.5 196,-529.5 13,-529.5"/>
-<text text-anchor="middle" x="104.5" y="-915.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::Object</text>
-<polyline fill="none" stroke="#000000" points="13,-908.5 196,-908.5 "/>
-<text text-anchor="start" x="21" y="-896.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<text text-anchor="start" x="21" y="-885.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_final</text>
-<text text-anchor="start" x="21" y="-874.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots</text>
-<text text-anchor="start" x="21" y="-863.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots_can</text>
-<text text-anchor="start" x="21" y="-852.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_overflow</text>
-<text text-anchor="start" x="21" y="-841.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_visit</text>
-<text text-anchor="start" x="21" y="-830.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_attrs</text>
-<text text-anchor="start" x="21" y="-819.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_sequal</text>
-<text text-anchor="start" x="21" y="-808.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
-<text text-anchor="start" x="21" y="-797.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_shash</text>
-<text text-anchor="start" x="21" y="-786.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
-<text text-anchor="start" x="21" y="-775.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_index</text>
-<text text-anchor="start" x="21" y="-764.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># type_index_</text>
-<text text-anchor="start" x="21" y="-753.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># ref_counter_</text>
-<text text-anchor="start" x="21" y="-742.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># deleter_</text>
-<polyline fill="none" stroke="#000000" points="13,-735.5 196,-735.5 "/>
-<text text-anchor="start" x="21" y="-723.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ type_index()</text>
-<text text-anchor="start" x="21" y="-712.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKey()</text>
-<text text-anchor="start" x="21" y="-701.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKeyHash()</text>
-<text text-anchor="start" x="21" y="-690.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ IsInstance()</text>
-<text text-anchor="start" x="21" y="-679.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ unique()</text>
-<text text-anchor="start" x="21" y="-668.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="21" y="-657.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="21" y="-646.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="21" y="-635.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
-<text text-anchor="start" x="21" y="-624.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
-<text text-anchor="start" x="21" y="-613.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2Key()</text>
-<text text-anchor="start" x="21" y="-602.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2KeyHash()</text>
-<text text-anchor="start" x="21" y="-591.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeKey2Index()</text>
-<text text-anchor="start" x="21" y="-580.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _GetOrAllocRuntimeTypeIndex()</text>
-<text text-anchor="start" x="21" y="-569.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ RuntimeTypeIndex()</text>
-<text text-anchor="start" x="21" y="-558.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># IncRef()</text>
-<text text-anchor="start" x="21" y="-547.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># DecRef()</text>
-<text text-anchor="start" x="21" y="-536.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># GetOrAllocRuntimeTypeIndex()</text>
+<polygon fill="#ffffff" stroke="#000000" points="13,-562.5 13,-960.5 196,-960.5 196,-562.5 13,-562.5"/>
+<text text-anchor="middle" x="104.5" y="-948.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::Object</text>
+<polyline fill="none" stroke="#000000" points="13,-941.5 196,-941.5 "/>
+<text text-anchor="start" x="21" y="-929.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<text text-anchor="start" x="21" y="-918.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_final</text>
+<text text-anchor="start" x="21" y="-907.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots</text>
+<text text-anchor="start" x="21" y="-896.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots_can</text>
+<text text-anchor="start" x="21" y="-885.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_overflow</text>
+<text text-anchor="start" x="21" y="-874.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_visit</text>
+<text text-anchor="start" x="21" y="-863.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_attrs</text>
+<text text-anchor="start" x="21" y="-852.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_sequal</text>
+<text text-anchor="start" x="21" y="-841.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
+<text text-anchor="start" x="21" y="-830.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_shash</text>
+<text text-anchor="start" x="21" y="-819.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
+<text text-anchor="start" x="21" y="-808.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_index</text>
+<text text-anchor="start" x="21" y="-797.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># type_index_</text>
+<text text-anchor="start" x="21" y="-786.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># ref_counter_</text>
+<text text-anchor="start" x="21" y="-775.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># deleter_</text>
+<polyline fill="none" stroke="#000000" points="13,-768.5 196,-768.5 "/>
+<text text-anchor="start" x="21" y="-756.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ type_index()</text>
+<text text-anchor="start" x="21" y="-745.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKey()</text>
+<text text-anchor="start" x="21" y="-734.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKeyHash()</text>
+<text text-anchor="start" x="21" y="-723.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ IsInstance()</text>
+<text text-anchor="start" x="21" y="-712.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ unique()</text>
+<text text-anchor="start" x="21" y="-701.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
+<text text-anchor="start" x="21" y="-690.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
+<text text-anchor="start" x="21" y="-679.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
+<text text-anchor="start" x="21" y="-668.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
+<text text-anchor="start" x="21" y="-657.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
+<text text-anchor="start" x="21" y="-646.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2Key()</text>
+<text text-anchor="start" x="21" y="-635.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2KeyHash()</text>
+<text text-anchor="start" x="21" y="-624.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeKey2Index()</text>
+<text text-anchor="start" x="21" y="-613.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _GetOrAllocRuntimeTypeIndex()</text>
+<text text-anchor="start" x="21" y="-602.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ RuntimeTypeIndex()</text>
+<text text-anchor="start" x="21" y="-591.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># IncRef()</text>
+<text text-anchor="start" x="21" y="-580.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># DecRef()</text>
+<text text-anchor="start" x="21" y="-569.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># GetOrAllocRuntimeTypeIndex()</text>
</a>
</g>
</g>
<!-- Node1->Node0 -->
<g id="edge1" class="edge">
<title>Node1->Node0</title>
-<path fill="none" stroke="#191970" d="M104.5,-518.883C104.5,-509.8603 104.5,-501.0496 104.5,-492.5763"/>
-<polygon fill="none" stroke="#191970" points="101.0001,-519.1535 104.5,-529.1535 108.0001,-519.1535 101.0001,-519.1535"/>
+<path fill="none" stroke="#191970" d="M104.5,-552.0222C104.5,-543.097 104.5,-534.3384 104.5,-525.856"/>
+<polygon fill="none" stroke="#191970" points="101.0001,-552.164 104.5,-562.1641 108.0001,-552.1641 101.0001,-552.164"/>
</g>
</g>
</svg>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode-members.html b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode-members.html
index 0d17cee104..52074beb4e 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode-members.html
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode-members.html
@@ -80,28 +80,30 @@ $(function() {
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ada16907864ae1df2b570f94788c4e06f">_type_key</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a00614b22bd1cae147dc0d83cdd071187">CommitTuningRecord</a>(const TuningRecord &record) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a12d64aa44d3d2ab37a6600d17be3e708">CommitWorkload</a>(const IRModule &mod) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a70fb5361147634605d6595bb89381f03">DecRef</a>()</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">protected</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#af4407d2b59132e803ff791482dbe0145">deleter_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a7d7d50ae9b6c43e064b38db91bc8123d">f_commit_tuning_record</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#aa0bd64983cde2fc6ef1547448352981c">f_commit_workload</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#aafdd0874be052072521b2aa8a6c56d5f">f_get_all_tuning_records</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a52fb1116090619e95fb6b28352308eed">f_get_top_k</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#add146bf1e2006f72ed1534b2004bcb06">f_has_workload</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#abd9fc8fc83bc6c252465ffdbcb310bfc">f_query_ir_module</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a4a03c70569c9a18059861dfb5c90e845">f_query_schedule</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a65fcb9b59b8ce6e685fb62c4459c57ba">f_query_tuning_record</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ac7ae1a05fe5c7858f5860133a82bc7b7">f_size</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a653d04c0c6349350489c0ea5f68563f1">FCommitTuningRecord</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a44b8d5e2721f12bdaf1a457b85f23124">FCommitWorkload</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a9e84841ca982bff376a978ade0132631">FDeleter</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a574d90736eda21019540d4a26c155b28">FGetAllTuningRecords</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad5e04e950cd2a63f439d95285b5674b6">FGetTopK</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#afb177ebca29227e94c3b3036b4908bad">FHasWorkload</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a713ae7e8634c0aedc366dffda2c899df">FQueryIRModule</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a16c17595db4a845b3511d6d7fa0f741d">FQuerySchedule</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#acd7fb3619d530c0ae85fb1d6e94f6e7d">FQueryTuningRecord</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a34efc3d18473d179b13332abe5c63324">FSize</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad07d7d9e78771eaa2e6e65f84e032401">GetAllTuningRecords</a>() final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#adfd6210675fcb9dd7778d77b8ce56782">DatabaseNode</a>(String mod_eq_name="structural")</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">explicit</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a70fb5361147634605d6595bb89381f03">DecRef</a>()</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">protected</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#af4407d2b59132e803ff791482dbe0145">deleter_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a7d7d50ae9b6c43e064b38db91bc8123d">f_commit_tuning_record</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#aa0bd64983cde2fc6ef1547448352981c">f_commit_workload</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#aafdd0874be052072521b2aa8a6c56d5f">f_get_all_tuning_records</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a52fb1116090619e95fb6b28352308eed">f_get_top_k</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#add146bf1e2006f72ed1534b2004bcb06">f_has_workload</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#abd9fc8fc83bc6c252465ffdbcb310bfc">f_query_ir_module</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a4a03c70569c9a18059861dfb5c90e845">f_query_schedule</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a65fcb9b59b8ce6e685fb62c4459c57ba">f_query_tuning_record</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ac7ae1a05fe5c7858f5860133a82bc7b7">f_size</a></td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a653d04c0c6349350489c0ea5f68563f1">FCommitTuningRecord</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a44b8d5e2721f12bdaf1a457b85f23124">FCommitWorkload</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a9e84841ca982bff376a978ade0132631">FDeleter</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a574d90736eda21019540d4a26c155b28">FGetAllTuningRecords</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad5e04e950cd2a63f439d95285b5674b6">FGetTopK</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#afb177ebca29227e94c3b3036b4908bad">FHasWorkload</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a713ae7e8634c0aedc366dffda2c899df">FQueryIRModule</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a16c17595db4a845b3511d6d7fa0f741d">FQuerySchedule</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#acd7fb3619d530c0ae85fb1d6e94f6e7d">FQueryTuningRecord</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a34efc3d18473d179b13332abe5c63324">FSize</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad07d7d9e78771eaa2e6e65f84e032401">GetAllTuningRecords</a>() final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a491ad91dbd0a811712c5013bade724cb">GetModuleEquality</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a726972ff315c446192df94027ddea032">GetOrAllocRuntimeTypeIndex</a>(const std::string &key, uint32_t static_tindex, uint32_t parent_tindex, uint32_t type_child_slots, bool type_child_slots_can_overflow)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span><span class="mlabel">static</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#adebe3a6bfb55e5ce0807b97f12a6c39e">GetTopK</a>(const Workload &workload, int top_k) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a4d951e51832081b85875669eac90e940">GetTypeKey</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
@@ -114,23 +116,24 @@ $(function() {
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#aa1612f69ea5b4225d4cda759cd517323">Object</a>(Object &&other)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
<tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a69c32fbd96181f5c21d2c878ab285e4f">operator=</a>(const Object &other)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
<tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#ae341e561272ff43cdcbc927bc29ac50d">operator=</a>(Object &&other)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a4a21df0e4369b208e8d0332c0dcdfee3">QueryIRModule</a>(const IRModule &mod, const Target &target, const String &workload_name) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a340ce2715f3f9be3ded8a4560a45f5d3">QuerySchedule</a>(const IRModule &mod, const Target &target, const String &workload_name) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a76186192f9e7e52d8c9f1e3b53fe0e60">QueryTuningRecord</a>(const IRModule &mod, const Target &target, const String &workload_name) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a0d492efee331e2239a093f4b2017c10f">ref_counter_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a55549a6c23987890246248682560a03d">RefCounterType</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#ad94d79729ac85aa7c976e23d39066383">RuntimeTypeIndex</a>()</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">static</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a36817d04978253571fef7d01427ce9c0">Size</a>() final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#ae7ea55bfa3703dfb6452573afc31a45e">TVM_DECLARE_BASE_OBJECT_INFO</a>(DatabaseNode, runtime::Object)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad4256c83aa0af3514af963a073811b2b">TVM_DECLARE_FINAL_OBJECT_INFO</a>(PyDatabaseNode, DatabaseNode)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a481f01923b14e1851ebd38506e9c66ea">type_index</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a4bfc2586cb55f2af47728187b3256255">type_index_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a817ba6c23b7ee1821c48a75edf255a30">TypeIndex2Key</a>(uint32_t tindex)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a6ee32a02dd44257da105fbbe5d9c8622">TypeIndex2KeyHash</a>(uint32_t tindex)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a6841f97e06e6614dd7e82c6dd41b818a">TypeKey2Index</a>(const std::string &key)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#afd548730a6139d19fe24473ad66026d7">unique</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
- <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a1a24b977e3d25a7ffab704e12457fb66">VisitAttrs</a>(tvm::AttrVisitor *v)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
- <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a776359f44ac6b51e337d4a1efc3f04a9">~DatabaseNode</a>()=default</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">virtual</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a0f8533a4896ab37bd2f099460b0e2938">PyDatabaseNode</a>(String mod_eq_name="structural")</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">explicit</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a4a21df0e4369b208e8d0332c0dcdfee3">QueryIRModule</a>(const IRModule &mod, const Target &target, const String &workload_name) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a340ce2715f3f9be3ded8a4560a45f5d3">QuerySchedule</a>(const IRModule &mod, const Target &target, const String &workload_name) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a76186192f9e7e52d8c9f1e3b53fe0e60">QueryTuningRecord</a>(const IRModule &mod, const Target &target, const String &workload_name) final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a0d492efee331e2239a093f4b2017c10f">ref_counter_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a55549a6c23987890246248682560a03d">RefCounterType</a> typedef</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#ad94d79729ac85aa7c976e23d39066383">RuntimeTypeIndex</a>()</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">static</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a36817d04978253571fef7d01427ce9c0">Size</a>() final</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span><span class="mlabel">virtual</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#ae7ea55bfa3703dfb6452573afc31a45e">TVM_DECLARE_BASE_OBJECT_INFO</a>(DatabaseNode, runtime::Object)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad4256c83aa0af3514af963a073811b2b">TVM_DECLARE_FINAL_OBJECT_INFO</a>(PyDatabaseNode, DatabaseNode)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a481f01923b14e1851ebd38506e9c66ea">type_index</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a4bfc2586cb55f2af47728187b3256255">type_index_</a></td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">protected</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a817ba6c23b7ee1821c48a75edf255a30">TypeIndex2Key</a>(uint32_t tindex)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a6ee32a02dd44257da105fbbe5d9c8622">TypeIndex2KeyHash</a>(uint32_t tindex)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#a6841f97e06e6614dd7e82c6dd41b818a">TypeKey2Index</a>(const std::string &key)</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">static</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html#afd548730a6139d19fe24473ad66026d7">unique</a>() const</td><td class="entry"><a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
+ <tr><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a1a24b977e3d25a7ffab704e12457fb66">VisitAttrs</a>(tvm::AttrVisitor *v)</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">tvm::meta_schedule::PyDatabaseNode</a></td><td class="entry"><span class="mlabel">inline</span></td></tr>
+ <tr class="even"><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a48707424816f7ad0fd6f59a7111d9a09">~DatabaseNode</a>()</td><td class="entry"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td><td class="entry"><span class="mlabel">virtual</span></td></tr>
</table></div><!-- contents -->
<!-- start footer part -->
<hr class="footer"/><address class="footer"><small>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode.html b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode.html
index 233aeddfe0..d20d74b3ce 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode.html
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode.html
@@ -80,13 +80,13 @@ $(function() {
<div class="dynheader">
Inheritance diagram for tvm::meta_schedule::PyDatabaseNode:</div>
<div class="dyncontent">
-<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1PyDatabaseNode__inherit__graph.svg" width="290" height="1248"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
+<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1PyDatabaseNode__inherit__graph.svg" width="290" height="1292"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
</div>
</div>
<div class="dynheader">
Collaboration diagram for tvm::meta_schedule::PyDatabaseNode:</div>
<div class="dyncontent">
-<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1PyDatabaseNode__coll__graph.svg" width="2527" height="1118"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
+<div class="center"><iframe scrolling="no" frameborder="0" src="classtvm_1_1meta__schedule_1_1PyDatabaseNode__coll__graph.svg" width="2527" height="1162"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
</div>
</div>
<table class="memberdecls">
@@ -128,6 +128,9 @@ Public Types</h2></td></tr>
</table><table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="pub-methods"></a>
Public Member Functions</h2></td></tr>
+<tr class="memitem:a0f8533a4896ab37bd2f099460b0e2938"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a0f8533a4896ab37bd2f099460b0e2938">PyDatabaseNode</a> (<a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> mod_eq_name="structural")</td></tr>
+<tr class="memdesc:a0f8533a4896ab37bd2f099460b0e2938"><td class="mdescLeft"> </td><td class="mdescRight"><a class="el" href="classtvm_1_1Constructor.html" title="Managed reference to ConstructorNode. ">Constructor</a>. <a href="#a0f8533a4896ab37bd2f099460b0e2938">More...</a><br /></td></tr>
+<tr class="separator:a0f8533a4896ab37bd2f099460b0e2938"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a1a24b977e3d25a7ffab704e12457fb66"><td class="memItemLeft" align="right" valign="top">void </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a1a24b977e3d25a7ffab704e12457fb66">VisitAttrs</a> (<a class="el" href="classtvm_1_1AttrVisitor.html">tvm::AttrVisitor</a> *v)</td></tr>
<tr class="separator:a1a24b977e3d25a7ffab704e12457fb66"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a38195f2c3aea9ae4890477393ab2c624"><td class="memItemLeft" align="right" valign="top">bool </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#a38195f2c3aea9ae4890477393ab2c624">HasWorkload</a> (const <a class="el" href="classtvm_1_1IRModule.html">IRModule</a> &mod) final</td></tr>
@@ -160,9 +163,15 @@ Public Member Functions</h2></td></tr>
<tr class="memitem:ad4256c83aa0af3514af963a073811b2b"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html#ad4256c83aa0af3514af963a073811b2b">TVM_DECLARE_FINAL_OBJECT_INFO</a> (<a class="el" href="classtvm_1_1meta__schedule_1_1PyDatabaseNode.html">PyDatabaseNode</a>, <a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">DatabaseNode</a>)</td></tr>
<tr class="separator:ad4256c83aa0af3514af963a073811b2b"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="inherit_header pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td colspan="2" onclick="javascript:toggleInherit('pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode')"><img src="closed.png" alt="-"/> Public Member Functions inherited from <a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">tvm::meta_schedule::DatabaseNode</a></td></tr>
-<tr class="memitem:a776359f44ac6b51e337d4a1efc3f04a9 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memItemLeft" align="right" valign="top">virtual </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a776359f44ac6b51e337d4a1efc3f04a9">~DatabaseNode</a> ()=default</td></tr>
-<tr class="memdesc:a776359f44ac6b51e337d4a1efc3f04a9 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="mdescLeft"> </td><td class="mdescRight">Default destructor. <a href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a776359f44ac6b51e337d4a1efc3f04a9">More...</a><br /></td></tr>
-<tr class="separator:a776359f44ac6b51e337d4a1efc3f04a9 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:adfd6210675fcb9dd7778d77b8ce56782 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#adfd6210675fcb9dd7778d77b8ce56782">DatabaseNode</a> (<a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> mod_eq_name="structural")</td></tr>
+<tr class="memdesc:adfd6210675fcb9dd7778d77b8ce56782 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="mdescLeft"> </td><td class="mdescRight"><a class="el" href="classtvm_1_1Constructor.html" title="Managed reference to ConstructorNode. ">Constructor</a>. <a href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#adfd6210675fcb9dd7778d77b8ce56782">More...</a><br /></td></tr>
+<tr class="separator:adfd6210675fcb9dd7778d77b8ce56782 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a48707424816f7ad0fd6f59a7111d9a09 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memItemLeft" align="right" valign="top">virtual </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a48707424816f7ad0fd6f59a7111d9a09">~DatabaseNode</a> ()</td></tr>
+<tr class="memdesc:a48707424816f7ad0fd6f59a7111d9a09 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="mdescLeft"> </td><td class="mdescRight">Default destructor. <a href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a48707424816f7ad0fd6f59a7111d9a09">More...</a><br /></td></tr>
+<tr class="separator:a48707424816f7ad0fd6f59a7111d9a09 inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a491ad91dbd0a811712c5013bade724cb inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memItemLeft" align="right" valign="top">const ModuleEquality & </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a491ad91dbd0a811712c5013bade724cb">GetModuleEquality</a> () const</td></tr>
+<tr class="memdesc:a491ad91dbd0a811712c5013bade724cb inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="mdescLeft"> </td><td class="mdescRight">Return a reference to the owned module equality method instance. <a href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#a491ad91dbd0a811712c5013bade724cb">More...</a><br /></td></tr>
+<tr class="separator:a491ad91dbd0a811712c5013bade724cb inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:ae7ea55bfa3703dfb6452573afc31a45e inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memItemLeft" align="right" valign="top"> </td><td class="memItemRight" valign="bottom"><a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html#ae7ea55bfa3703dfb6452573afc31a45e">TVM_DECLARE_BASE_OBJECT_INFO</a> (<a class="el" href="classtvm_1_1meta__schedule_1_1DatabaseNode.html">DatabaseNode</a>, <a class="el" href="classtvm_1_1runtime_1_1Objec [...]
<tr class="separator:ae7ea55bfa3703dfb6452573afc31a45e inherit pub_methods_classtvm_1_1meta__schedule_1_1DatabaseNode"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="inherit_header pub_methods_classtvm_1_1runtime_1_1Object"><td colspan="2" onclick="javascript:toggleInherit('pub_methods_classtvm_1_1runtime_1_1Object')"><img src="closed.png" alt="-"/> Public Member Functions inherited from <a class="el" href="classtvm_1_1runtime_1_1Object.html">tvm::runtime::Object</a></td></tr>
@@ -483,6 +492,44 @@ Additional Inherited Members</h2></td></tr>
<p>The function type of <code>Size</code> method. </p>
<dl class="section return"><dt>Returns</dt><dd>The size of the database. </dd></dl>
+</div>
+</div>
+<h2 class="groupheader">Constructor & Destructor Documentation</h2>
+<a id="a0f8533a4896ab37bd2f099460b0e2938"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a0f8533a4896ab37bd2f099460b0e2938">◆ </a></span>PyDatabaseNode()</h2>
+
+<div class="memitem">
+<div class="memproto">
+<table class="mlabels">
+ <tr>
+ <td class="mlabels-left">
+ <table class="memname">
+ <tr>
+ <td class="memname">tvm::meta_schedule::PyDatabaseNode::PyDatabaseNode </td>
+ <td>(</td>
+ <td class="paramtype"><a class="el" href="classtvm_1_1runtime_1_1String.html">String</a> </td>
+ <td class="paramname"><em>mod_eq_name</em> = <code>"structural"</code></td><td>)</td>
+ <td></td>
+ </tr>
+ </table>
+ </td>
+ <td class="mlabels-right">
+<span class="mlabels"><span class="mlabel">explicit</span></span> </td>
+ </tr>
+</table>
+</div><div class="memdoc">
+
+<p><a class="el" href="classtvm_1_1Constructor.html" title="Managed reference to ConstructorNode. ">Constructor</a>. </p>
+<dl class="params"><dt>Parameters</dt><dd>
+ <table class="params">
+ <tr><td class="paramname">mod_eq_name</td><td>A string to specify the module equality testing and hashing method. It must be one of the followings:<ul>
+<li>"structural": Use StructuralEqual/Hash </li>
+</ul>
+</td></tr>
+ </table>
+ </dd>
+</dl>
+
</div>
</div>
<h2 class="groupheader">Member Function Documentation</h2>
diff --git a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode__coll__graph.svg b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode__coll__graph.svg
index e6d6349bc5..fabb4d3388 100644
--- a/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode__coll__graph.svg
+++ b/docs/reference/api/doxygen/classtvm_1_1meta__schedule_1_1PyDatabaseNode__coll__graph.svg
@@ -4,20 +4,21 @@
<!-- Generated by graphviz version 2.40.1 (20161225.0304)
-->
<!-- Title: tvm::meta_schedule::PyDatabaseNode Pages: 1 -->
-<svg width="1895pt" height="838pt"
- viewBox="0.00 0.00 1895.00 838.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
-<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 834)">
+<svg width="1895pt" height="871pt"
+ viewBox="0.00 0.00 1895.00 871.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 867)">
<title>tvm::meta_schedule::PyDatabaseNode</title>
-<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-834 1891,-834 1891,4 -4,4"/>
+<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-867 1891,-867 1891,4 -4,4"/>
<!-- Node3 -->
<g id="node1" class="node">
<title>Node3</title>
-<polygon fill="#bfbfbf" stroke="#000000" points="827,-.5 827,-178.5 1036,-178.5 1036,-.5 827,-.5"/>
-<text text-anchor="start" x="835" y="-166.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
-<text text-anchor="middle" x="931.5" y="-155.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::PyDatabaseNode</text>
-<polyline fill="none" stroke="#000000" points="827,-148.5 1036,-148.5 "/>
-<text text-anchor="start" x="835" y="-136.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<polyline fill="none" stroke="#000000" points="827,-129.5 1036,-129.5 "/>
+<polygon fill="#bfbfbf" stroke="#000000" points="827,-.5 827,-189.5 1036,-189.5 1036,-.5 827,-.5"/>
+<text text-anchor="start" x="835" y="-177.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
+<text text-anchor="middle" x="931.5" y="-166.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::PyDatabaseNode</text>
+<polyline fill="none" stroke="#000000" points="827,-159.5 1036,-159.5 "/>
+<text text-anchor="start" x="835" y="-147.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<polyline fill="none" stroke="#000000" points="827,-140.5 1036,-140.5 "/>
+<text text-anchor="start" x="835" y="-128.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ PyDatabaseNode()</text>
<text text-anchor="start" x="835" y="-117.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ VisitAttrs()</text>
<text text-anchor="start" x="835" y="-106.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
<text text-anchor="start" x="835" y="-95.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
@@ -33,284 +34,286 @@
<!-- Node4 -->
<g id="node2" class="node">
<title>Node4</title>
-<g id="a_node2"><a xlink:href="classtvm_1_1meta__schedule_1_1DatabaseNode.html" target="_top" xlink:title="{tvm::meta_schedule\l::DatabaseNode\n|+ _type_key\l|+ ~DatabaseNode()\l+ HasWorkload()\l+ CommitWorkload()\l+ CommitTuningRecord()\l+ GetTopK()\l+ GetAllTuningRecords()\l+ Size()\l+ QueryTuningRecord()\l+ QuerySchedule()\l+ QueryIRModule()\l+ TVM_DECLARE_BASE_OBJECT_INFO()\l}">
-<polygon fill="#ffffff" stroke="#000000" points="0,-226.5 0,-404.5 207,-404.5 207,-226.5 0,-226.5"/>
-<text text-anchor="start" x="8" y="-392.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
-<text text-anchor="middle" x="103.5" y="-381.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::DatabaseNode</text>
-<polyline fill="none" stroke="#000000" points="0,-374.5 207,-374.5 "/>
-<text text-anchor="start" x="8" y="-362.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<polyline fill="none" stroke="#000000" points="0,-355.5 207,-355.5 "/>
-<text text-anchor="start" x="8" y="-343.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ ~DatabaseNode()</text>
-<text text-anchor="start" x="8" y="-332.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
-<text text-anchor="start" x="8" y="-321.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
-<text text-anchor="start" x="8" y="-310.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitTuningRecord()</text>
-<text text-anchor="start" x="8" y="-299.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTopK()</text>
-<text text-anchor="start" x="8" y="-288.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetAllTuningRecords()</text>
-<text text-anchor="start" x="8" y="-277.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Size()</text>
-<text text-anchor="start" x="8" y="-266.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryTuningRecord()</text>
-<text text-anchor="start" x="8" y="-255.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QuerySchedule()</text>
-<text text-anchor="start" x="8" y="-244.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryIRModule()</text>
-<text text-anchor="start" x="8" y="-233.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TVM_DECLARE_BASE_OBJECT_INFO()</text>
+<g id="a_node2"><a xlink:href="classtvm_1_1meta__schedule_1_1DatabaseNode.html" target="_top" xlink:title="{tvm::meta_schedule\l::DatabaseNode\n|+ _type_key\l|+ DatabaseNode()\l+ ~DatabaseNode()\l+ HasWorkload()\l+ CommitWorkload()\l+ CommitTuningRecord()\l+ GetTopK()\l+ GetAllTuningRecords()\l+ Size()\l+ QueryTuningRecord()\l+ QuerySchedule()\l+ QueryIRModule()\l+ GetModuleEquality()\l+ TVM_DECLARE_BASE_OBJECT_INFO()\l}">
+<polygon fill="#ffffff" stroke="#000000" points="0,-237.5 0,-437.5 207,-437.5 207,-237.5 0,-237.5"/>
+<text text-anchor="start" x="8" y="-425.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::meta_schedule</text>
+<text text-anchor="middle" x="103.5" y="-414.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">::DatabaseNode</text>
+<polyline fill="none" stroke="#000000" points="0,-407.5 207,-407.5 "/>
+<text text-anchor="start" x="8" y="-395.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<polyline fill="none" stroke="#000000" points="0,-388.5 207,-388.5 "/>
+<text text-anchor="start" x="8" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ DatabaseNode()</text>
+<text text-anchor="start" x="8" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ ~DatabaseNode()</text>
+<text text-anchor="start" x="8" y="-354.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ HasWorkload()</text>
+<text text-anchor="start" x="8" y="-343.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitWorkload()</text>
+<text text-anchor="start" x="8" y="-332.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ CommitTuningRecord()</text>
+<text text-anchor="start" x="8" y="-321.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTopK()</text>
+<text text-anchor="start" x="8" y="-310.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetAllTuningRecords()</text>
+<text text-anchor="start" x="8" y="-299.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Size()</text>
+<text text-anchor="start" x="8" y="-288.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryTuningRecord()</text>
+<text text-anchor="start" x="8" y="-277.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QuerySchedule()</text>
+<text text-anchor="start" x="8" y="-266.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ QueryIRModule()</text>
+<text text-anchor="start" x="8" y="-255.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetModuleEquality()</text>
+<text text-anchor="start" x="8" y="-244.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TVM_DECLARE_BASE_OBJECT_INFO()</text>
</a>
</g>
</g>
<!-- Node4->Node3 -->
<g id="edge1" class="edge">
<title>Node4->Node3</title>
-<path fill="none" stroke="#191970" d="M216.3934,-225.8991C426.5422,-131.5736 685.0934,-102.8631 826.8539,-93.7431"/>
-<polygon fill="none" stroke="#191970" points="214.7057,-222.8217 207.0537,-230.1495 217.6053,-229.1929 214.7057,-222.8217"/>
+<path fill="none" stroke="#191970" d="M216.2752,-237.096C338.3904,-181.3925 659.487,-131.9979 826.9366,-108.7923"/>
+<polygon fill="none" stroke="#191970" points="214.6549,-233.9911 207.1425,-241.4621 217.6742,-240.3066 214.6549,-233.9911"/>
</g>
<!-- Node5 -->
<g id="node3" class="node">
<title>Node5</title>
<g id="a_node3"><a xlink:href="classtvm_1_1runtime_1_1Object.html" target="_top" xlink:title="base class of all object containers. ">
-<polygon fill="#ffffff" stroke="#000000" points="12,-442.5 12,-829.5 195,-829.5 195,-442.5 12,-442.5"/>
-<text text-anchor="middle" x="103.5" y="-817.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::Object</text>
-<polyline fill="none" stroke="#000000" points="12,-810.5 195,-810.5 "/>
-<text text-anchor="start" x="20" y="-798.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
-<text text-anchor="start" x="20" y="-787.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_final</text>
-<text text-anchor="start" x="20" y="-776.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots</text>
-<text text-anchor="start" x="20" y="-765.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots_can</text>
-<text text-anchor="start" x="20" y="-754.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_overflow</text>
-<text text-anchor="start" x="20" y="-743.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_visit</text>
-<text text-anchor="start" x="20" y="-732.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_attrs</text>
-<text text-anchor="start" x="20" y="-721.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_sequal</text>
-<text text-anchor="start" x="20" y="-710.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
-<text text-anchor="start" x="20" y="-699.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_shash</text>
-<text text-anchor="start" x="20" y="-688.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
-<text text-anchor="start" x="20" y="-677.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_index</text>
-<text text-anchor="start" x="20" y="-666.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># type_index_</text>
-<text text-anchor="start" x="20" y="-655.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># ref_counter_</text>
-<polyline fill="none" stroke="#000000" points="12,-648.5 195,-648.5 "/>
-<text text-anchor="start" x="20" y="-636.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ type_index()</text>
-<text text-anchor="start" x="20" y="-625.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKey()</text>
-<text text-anchor="start" x="20" y="-614.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKeyHash()</text>
-<text text-anchor="start" x="20" y="-603.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ IsInstance()</text>
-<text text-anchor="start" x="20" y="-592.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ unique()</text>
-<text text-anchor="start" x="20" y="-581.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="20" y="-570.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="20" y="-559.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
-<text text-anchor="start" x="20" y="-548.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
-<text text-anchor="start" x="20" y="-537.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
-<text text-anchor="start" x="20" y="-526.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2Key()</text>
-<text text-anchor="start" x="20" y="-515.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2KeyHash()</text>
-<text text-anchor="start" x="20" y="-504.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeKey2Index()</text>
-<text text-anchor="start" x="20" y="-493.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _GetOrAllocRuntimeTypeIndex()</text>
-<text text-anchor="start" x="20" y="-482.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ RuntimeTypeIndex()</text>
-<text text-anchor="start" x="20" y="-471.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># IncRef()</text>
-<text text-anchor="start" x="20" y="-460.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># DecRef()</text>
-<text text-anchor="start" x="20" y="-449.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># GetOrAllocRuntimeTypeIndex()</text>
+<polygon fill="#ffffff" stroke="#000000" points="12,-475.5 12,-862.5 195,-862.5 195,-475.5 12,-475.5"/>
+<text text-anchor="middle" x="103.5" y="-850.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::Object</text>
+<polyline fill="none" stroke="#000000" points="12,-843.5 195,-843.5 "/>
+<text text-anchor="start" x="20" y="-831.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_key</text>
+<text text-anchor="start" x="20" y="-820.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_final</text>
+<text text-anchor="start" x="20" y="-809.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots</text>
+<text text-anchor="start" x="20" y="-798.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_child_slots_can</text>
+<text text-anchor="start" x="20" y="-787.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_overflow</text>
+<text text-anchor="start" x="20" y="-776.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_visit</text>
+<text text-anchor="start" x="20" y="-765.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_attrs</text>
+<text text-anchor="start" x="20" y="-754.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_sequal</text>
+<text text-anchor="start" x="20" y="-743.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
+<text text-anchor="start" x="20" y="-732.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_has_method_shash</text>
+<text text-anchor="start" x="20" y="-721.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_reduce</text>
+<text text-anchor="start" x="20" y="-710.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _type_index</text>
+<text text-anchor="start" x="20" y="-699.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># type_index_</text>
+<text text-anchor="start" x="20" y="-688.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># ref_counter_</text>
+<polyline fill="none" stroke="#000000" points="12,-681.5 195,-681.5 "/>
+<text text-anchor="start" x="20" y="-669.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ type_index()</text>
+<text text-anchor="start" x="20" y="-658.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKey()</text>
+<text text-anchor="start" x="20" y="-647.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ GetTypeKeyHash()</text>
+<text text-anchor="start" x="20" y="-636.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ IsInstance()</text>
+<text text-anchor="start" x="20" y="-625.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ unique()</text>
+<text text-anchor="start" x="20" y="-614.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
+<text text-anchor="start" x="20" y="-603.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
+<text text-anchor="start" x="20" y="-592.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ Object()</text>
+<text text-anchor="start" x="20" y="-581.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
+<text text-anchor="start" x="20" y="-570.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ operator=()</text>
+<text text-anchor="start" x="20" y="-559.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2Key()</text>
+<text text-anchor="start" x="20" y="-548.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeIndex2KeyHash()</text>
+<text text-anchor="start" x="20" y="-537.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ TypeKey2Index()</text>
+<text text-anchor="start" x="20" y="-526.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ _GetOrAllocRuntimeTypeIndex()</text>
+<text text-anchor="start" x="20" y="-515.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">+ RuntimeTypeIndex()</text>
+<text text-anchor="start" x="20" y="-504.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># IncRef()</text>
+<text text-anchor="start" x="20" y="-493.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># DecRef()</text>
+<text text-anchor="start" x="20" y="-482.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"># GetOrAllocRuntimeTypeIndex()</text>
</a>
</g>
</g>
<!-- Node5->Node4 -->
<g id="edge2" class="edge">
<title>Node5->Node4</title>
-<path fill="none" stroke="#191970" d="M103.5,-431.9464C103.5,-422.5963 103.5,-413.4618 103.5,-404.684"/>
-<polygon fill="none" stroke="#191970" points="100.0001,-432.1701 103.5,-442.1701 107.0001,-432.1701 100.0001,-432.1701"/>
+<path fill="none" stroke="#191970" d="M103.5,-464.9868C103.5,-455.6259 103.5,-446.4373 103.5,-437.5493"/>
+<polygon fill="none" stroke="#191970" points="100.0001,-465.2059 103.5,-475.2059 107.0001,-465.206 100.0001,-465.2059"/>
</g>
<!-- Node5->Node5 -->
<g id="edge3" class="edge">
<title>Node5->Node5</title>
-<path fill="none" stroke="#404040" d="M195.3625,-669.9248C206.0482,-663.6637 213,-652.3555 213,-636 213,-625.0112 209.8618,-616.3007 204.5615,-609.8687"/>
-<polygon fill="none" stroke="#404040" points="204.5184,-609.8322 197.3548,-609.0056 195.3625,-602.0752 202.5261,-602.9017 204.5184,-609.8322"/>
-<text text-anchor="middle" x="239" y="-633.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> #deleter_</text>
+<path fill="none" stroke="#404040" d="M195.3625,-702.9248C206.0482,-696.6637 213,-685.3555 213,-669 213,-658.0112 209.8618,-649.3007 204.5615,-642.8687"/>
+<polygon fill="none" stroke="#404040" points="204.5184,-642.8322 197.3548,-642.0056 195.3625,-635.0752 202.5261,-635.9017 204.5184,-642.8322"/>
+<text text-anchor="middle" x="239" y="-666.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> #deleter_</text>
</g>
<!-- Node6 -->
<g id="node4" class="node">
<title>Node6</title>
<g id="a_node4"><a xlink:href="classtvm_1_1runtime_1_1TypedPackedFunc.html" target="_top" xlink:title="{tvm::runtime::TypedPacked\lFunc\< Optional\< TuningRecord\l \>(const IRModule &, const\l Target &, const String &)\>\n||}">
-<polygon fill="#ffffff" stroke="#000000" points="225,-270.5 225,-360.5 392,-360.5 392,-270.5 225,-270.5"/>
-<text text-anchor="start" x="233" y="-348.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
-<text text-anchor="start" x="233" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Optional< TuningRecord</text>
-<text text-anchor="start" x="233" y="-326.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> >(const IRModule &, const</text>
-<text text-anchor="middle" x="308.5" y="-315.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> Target &, const String &)></text>
-<polyline fill="none" stroke="#000000" points="225,-308.5 392,-308.5 "/>
-<text text-anchor="middle" x="308.5" y="-296.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
-<polyline fill="none" stroke="#000000" points="225,-289.5 392,-289.5 "/>
-<text text-anchor="middle" x="308.5" y="-277.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polygon fill="#ffffff" stroke="#000000" points="225,-292.5 225,-382.5 392,-382.5 392,-292.5 225,-292.5"/>
+<text text-anchor="start" x="233" y="-370.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
+<text text-anchor="start" x="233" y="-359.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Optional< TuningRecord</text>
+<text text-anchor="start" x="233" y="-348.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> >(const IRModule &, const</text>
+<text text-anchor="middle" x="308.5" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> Target &, const String &)></text>
+<polyline fill="none" stroke="#000000" points="225,-330.5 392,-330.5 "/>
+<text text-anchor="middle" x="308.5" y="-318.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polyline fill="none" stroke="#000000" points="225,-311.5 392,-311.5 "/>
+<text text-anchor="middle" x="308.5" y="-299.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
</a>
</g>
</g>
<!-- Node6->Node3 -->
<g id="edge4" class="edge">
<title>Node6->Node3</title>
-<path fill="none" stroke="#404040" d="M347.4664,-270.4891C362.7811,-254.8218 381.2886,-238.1463 400.5,-226 528.0961,-145.3282 701.3945,-112.2226 814.7972,-98.709"/>
-<polygon fill="none" stroke="#404040" points="815.0286,-98.6824 820.5312,-94.0218 826.9497,-97.3085 821.4472,-101.9692 815.0286,-98.6824"/>
-<text text-anchor="middle" x="513" y="-200" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_query_tuning_record</text>
+<path fill="none" stroke="#404040" d="M341.796,-292.4352C357.8068,-273.2635 378.33,-251.9322 400.5,-237 526.3125,-152.2611 700.6126,-118.0074 814.678,-104.2147"/>
+<polygon fill="none" stroke="#404040" points="814.9825,-104.1792 820.478,-99.5102 826.9015,-102.7871 821.4061,-107.4561 814.9825,-104.1792"/>
+<text text-anchor="middle" x="499" y="-211" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_query_tuning_record</text>
</g>
<!-- Node7 -->
<g id="node5" class="node">
<title>Node7</title>
<g id="a_node5"><a xlink:href="classtvm_1_1runtime_1_1TypedPackedFunc.html" target="_top" xlink:title="{tvm::runtime::TypedPacked\lFunc\< int64_t()\>\n||}">
-<polygon fill="#ffffff" stroke="#000000" points="410,-281.5 410,-349.5 559,-349.5 559,-281.5 410,-281.5"/>
-<text text-anchor="start" x="418" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
-<text text-anchor="middle" x="484.5" y="-326.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< int64_t()></text>
-<polyline fill="none" stroke="#000000" points="410,-319.5 559,-319.5 "/>
-<text text-anchor="middle" x="484.5" y="-307.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
-<polyline fill="none" stroke="#000000" points="410,-300.5 559,-300.5 "/>
-<text text-anchor="middle" x="484.5" y="-288.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polygon fill="#ffffff" stroke="#000000" points="410,-303.5 410,-371.5 559,-371.5 559,-303.5 410,-303.5"/>
+<text text-anchor="start" x="418" y="-359.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
+<text text-anchor="middle" x="484.5" y="-348.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< int64_t()></text>
+<polyline fill="none" stroke="#000000" points="410,-341.5 559,-341.5 "/>
+<text text-anchor="middle" x="484.5" y="-329.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polyline fill="none" stroke="#000000" points="410,-322.5 559,-322.5 "/>
+<text text-anchor="middle" x="484.5" y="-310.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
</a>
</g>
</g>
<!-- Node7->Node3 -->
<g id="edge5" class="edge">
<title>Node7->Node3</title>
-<path fill="none" stroke="#404040" d="M509.8187,-281.2095C524.8734,-262.7895 545.3265,-240.8758 567.5,-226 643.1785,-175.2286 740.289,-140.0413 815.1694,-118.1232"/>
-<polygon fill="none" stroke="#404040" points="815.4583,-118.04 820.117,-112.5357 826.9896,-114.719 822.331,-120.2233 815.4583,-118.04"/>
-<text text-anchor="middle" x="633.5" y="-200" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_size</text>
+<path fill="none" stroke="#404040" d="M506.0097,-303.3676C521.0196,-281.8733 542.7475,-254.9043 567.5,-237 641.9736,-183.1307 739.5873,-146.5049 814.9805,-123.9606"/>
+<polygon fill="none" stroke="#404040" points="815.3698,-123.8463 819.9985,-118.3168 826.8829,-120.4628 822.2542,-125.9923 815.3698,-123.8463"/>
+<text text-anchor="middle" x="632.5" y="-211" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_size</text>
</g>
<!-- Node8 -->
<g id="node6" class="node">
<title>Node8</title>
<g id="a_node6"><a xlink:href="classtvm_1_1runtime_1_1TypedPackedFunc.html" target="_top" xlink:title="{tvm::runtime::TypedPacked\lFunc\< bool(const IRModule &)\>\n||}">
-<polygon fill="#ffffff" stroke="#000000" points="577,-281.5 577,-349.5 748,-349.5 748,-281.5 577,-281.5"/>
-<text text-anchor="start" x="585" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
-<text text-anchor="middle" x="662.5" y="-326.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< bool(const IRModule &)></text>
-<polyline fill="none" stroke="#000000" points="577,-319.5 748,-319.5 "/>
-<text text-anchor="middle" x="662.5" y="-307.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
-<polyline fill="none" stroke="#000000" points="577,-300.5 748,-300.5 "/>
-<text text-anchor="middle" x="662.5" y="-288.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polygon fill="#ffffff" stroke="#000000" points="577,-303.5 577,-371.5 748,-371.5 748,-303.5 577,-303.5"/>
+<text text-anchor="start" x="585" y="-359.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
+<text text-anchor="middle" x="662.5" y="-348.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< bool(const IRModule &)></text>
+<polyline fill="none" stroke="#000000" points="577,-341.5 748,-341.5 "/>
+<text text-anchor="middle" x="662.5" y="-329.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polyline fill="none" stroke="#000000" points="577,-322.5 748,-322.5 "/>
+<text text-anchor="middle" x="662.5" y="-310.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
</a>
</g>
</g>
<!-- Node8->Node3 -->
<g id="edge6" class="edge">
<title>Node8->Node3</title>
-<path fill="none" stroke="#404040" d="M697.5361,-281.2596C715.215,-264.3013 737.1719,-243.7165 757.5,-226 776.3654,-209.5583 796.9035,-192.6407 817.0396,-176.5551"/>
-<polygon fill="none" stroke="#404040" points="817.1745,-176.4479 819.3821,-169.583 826.568,-168.9805 824.3604,-175.8454 817.1745,-176.4479"/>
-<text text-anchor="middle" x="834.5" y="-200" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_has_workload</text>
+<path fill="none" stroke="#404040" d="M692.6959,-303.4095C710.8716,-283.4435 734.817,-258.0891 757.5,-237 776.3201,-219.5024 796.9824,-201.6026 817.2909,-184.6674"/>
+<polygon fill="none" stroke="#404040" points="817.3507,-184.6178 819.417,-177.7091 826.589,-176.9592 824.5228,-183.8679 817.3507,-184.6178"/>
+<text text-anchor="middle" x="833.5" y="-211" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_has_workload</text>
</g>
<!-- Node9 -->
<g id="node7" class="node">
<title>Node9</title>
<g id="a_node7"><a xlink:href="classtvm_1_1runtime_1_1TypedPackedFunc.html" target="_top" xlink:title="{tvm::runtime::TypedPacked\lFunc\< Optional\< IRModule\l \>(const IRModule &, const\l Target &, const String &)\>\n||}">
-<polygon fill="#ffffff" stroke="#000000" points="766,-270.5 766,-360.5 915,-360.5 915,-270.5 766,-270.5"/>
-<text text-anchor="start" x="774" y="-348.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
-<text text-anchor="start" x="774" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Optional< IRModule</text>
-<text text-anchor="start" x="774" y="-326.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> >(const IRModule &, const</text>
-<text text-anchor="middle" x="840.5" y="-315.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> Target &, const String &)></text>
-<polyline fill="none" stroke="#000000" points="766,-308.5 915,-308.5 "/>
-<text text-anchor="middle" x="840.5" y="-296.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
-<polyline fill="none" stroke="#000000" points="766,-289.5 915,-289.5 "/>
-<text text-anchor="middle" x="840.5" y="-277.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polygon fill="#ffffff" stroke="#000000" points="766,-292.5 766,-382.5 915,-382.5 915,-292.5 766,-292.5"/>
+<text text-anchor="start" x="774" y="-370.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
+<text text-anchor="start" x="774" y="-359.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Optional< IRModule</text>
+<text text-anchor="start" x="774" y="-348.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> >(const IRModule &, const</text>
+<text text-anchor="middle" x="840.5" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> Target &, const String &)></text>
+<polyline fill="none" stroke="#000000" points="766,-330.5 915,-330.5 "/>
+<text text-anchor="middle" x="840.5" y="-318.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polyline fill="none" stroke="#000000" points="766,-311.5 915,-311.5 "/>
+<text text-anchor="middle" x="840.5" y="-299.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
</a>
</g>
</g>
<!-- Node9->Node3 -->
<g id="edge7" class="edge">
<title>Node9->Node3</title>
-<path fill="none" stroke="#404040" d="M858.6965,-270.3087C868.0375,-247.1102 879.8148,-217.8611 891.1588,-189.6881"/>
-<polygon fill="none" stroke="#404040" points="891.1644,-189.6739 889.695,-182.6141 895.6467,-178.5424 897.116,-185.6022 891.1644,-189.6739"/>
-<text text-anchor="middle" x="937" y="-200" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_query_ir_module</text>
+<path fill="none" stroke="#404040" d="M857.4031,-292.4559C867.1046,-266.6032 879.7229,-232.9773 891.7892,-200.8228"/>
+<polygon fill="none" stroke="#404040" points="891.8124,-200.7609 890.1754,-193.738 896.0284,-189.5259 897.6654,-196.5487 891.8124,-200.7609"/>
+<text text-anchor="middle" x="940" y="-211" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_query_ir_module</text>
</g>
<!-- Node10 -->
<g id="node8" class="node">
<title>Node10</title>
<g id="a_node8"><a xlink:href="classtvm_1_1runtime_1_1TypedPackedFunc.html" target="_top" xlink:title="{tvm::runtime::TypedPacked\lFunc\< Array\< TuningRecord \>()\>\n||}">
-<polygon fill="#ffffff" stroke="#000000" points="933,-281.5 933,-349.5 1112,-349.5 1112,-281.5 933,-281.5"/>
-<text text-anchor="start" x="941" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
-<text text-anchor="middle" x="1022.5" y="-326.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Array< TuningRecord >()></text>
-<polyline fill="none" stroke="#000000" points="933,-319.5 1112,-319.5 "/>
-<text text-anchor="middle" x="1022.5" y="-307.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
-<polyline fill="none" stroke="#000000" points="933,-300.5 1112,-300.5 "/>
-<text text-anchor="middle" x="1022.5" y="-288.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polygon fill="#ffffff" stroke="#000000" points="933,-303.5 933,-371.5 1112,-371.5 1112,-303.5 933,-303.5"/>
+<text text-anchor="start" x="941" y="-359.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
+<text text-anchor="middle" x="1022.5" y="-348.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Array< TuningRecord >()></text>
+<polyline fill="none" stroke="#000000" points="933,-341.5 1112,-341.5 "/>
+<text text-anchor="middle" x="1022.5" y="-329.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polyline fill="none" stroke="#000000" points="933,-322.5 1112,-322.5 "/>
+<text text-anchor="middle" x="1022.5" y="-310.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
</a>
</g>
</g>
<!-- Node10->Node3 -->
<g id="edge8" class="edge">
<title>Node10->Node3</title>
-<path fill="none" stroke="#404040" d="M1016.3733,-281.2803C1011.4537,-257.1896 1003.5372,-224.4703 992.5,-197 991.5469,-194.6277 990.5503,-192.2434 989.5165,-189.8534"/>
-<polygon fill="none" stroke="#404040" points="989.3905,-189.5758 983.2676,-185.7665 984.4289,-178.6495 990.5518,-182.4588 989.3905,-189.5758"/>
-<text text-anchor="middle" x="1061" y="-200" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_get_all_tuning_records</text>
+<path fill="none" stroke="#404040" d="M1018.3385,-303.4233C1014.3958,-276.8262 1007.2852,-239.2685 995.5,-208 994.5992,-205.6101 993.6529,-203.2104 992.6667,-200.8065"/>
+<polygon fill="none" stroke="#404040" points="992.5579,-200.5556 986.5007,-196.6428 987.7829,-189.5465 993.8401,-193.4594 992.5579,-200.5556"/>
+<text text-anchor="middle" x="1064" y="-211" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_get_all_tuning_records</text>
</g>
<!-- Node11 -->
<g id="node9" class="node">
<title>Node11</title>
<g id="a_node9"><a xlink:href="classtvm_1_1runtime_1_1TypedPackedFunc.html" target="_top" xlink:title="{tvm::runtime::TypedPacked\lFunc\< Array\< TuningRecord\l \>(const Workload &, int)\>\n||}">
-<polygon fill="#ffffff" stroke="#000000" points="1130.5,-276 1130.5,-355 1282.5,-355 1282.5,-276 1130.5,-276"/>
-<text text-anchor="start" x="1138.5" y="-343" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
-<text text-anchor="start" x="1138.5" y="-332" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Array< TuningRecord</text>
-<text text-anchor="middle" x="1206.5" y="-321" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> >(const Workload &, int)></text>
-<polyline fill="none" stroke="#000000" points="1130.5,-314 1282.5,-314 "/>
-<text text-anchor="middle" x="1206.5" y="-302" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
-<polyline fill="none" stroke="#000000" points="1130.5,-295 1282.5,-295 "/>
-<text text-anchor="middle" x="1206.5" y="-283" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polygon fill="#ffffff" stroke="#000000" points="1130.5,-298 1130.5,-377 1282.5,-377 1282.5,-298 1130.5,-298"/>
+<text text-anchor="start" x="1138.5" y="-365" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
+<text text-anchor="start" x="1138.5" y="-354" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Array< TuningRecord</text>
+<text text-anchor="middle" x="1206.5" y="-343" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> >(const Workload &, int)></text>
+<polyline fill="none" stroke="#000000" points="1130.5,-336 1282.5,-336 "/>
+<text text-anchor="middle" x="1206.5" y="-324" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
+<polyline fill="none" stroke="#000000" points="1130.5,-317 1282.5,-317 "/>
+<text text-anchor="middle" x="1206.5" y="-305" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
</a>
</g>
</g>
<!-- Node11->Node3 -->
<g id="edge9" class="edge">
<title>Node11->Node3</title>
-<path fill="none" stroke="#404040" d="M1187.6774,-275.8672C1174.3542,-250.9317 1154.5993,-219.3568 1130.5,-197 1106.1432,-174.4043 1076.1489,-155.0019 1046.9043,-139.1503"/>
-<polygon fill="none" stroke="#404040" points="1046.7816,-139.0854 1039.6073,-139.8138 1036.1757,-133.4716 1043.3499,-132.7432 1046.7816,-139.0854"/>
-<text text-anchor="middle" x="1175" y="-200" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_get_top_k</text>
+<path fill="none" stroke="#404040" d="M1191.0131,-297.837C1178.6629,-270.1057 1159.1824,-233.5986 1133.5,-208 1108.6776,-183.2586 1077.3225,-162.2596 1046.7672,-145.3247"/>
+<polygon fill="none" stroke="#404040" points="1046.5602,-145.2129 1039.3798,-145.8773 1036.0048,-139.5047 1043.1852,-138.8403 1046.5602,-145.2129"/>
+<text text-anchor="middle" x="1176" y="-211" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> +f_get_top_k</text>
</g>
<!-- Node12 -->
<g id="node10" class="node">
<title>Node12</title>
<g id="a_node10"><a xlink:href="classtvm_1_1runtime_1_1TypedPackedFunc.html" target="_top" xlink:title="{tvm::runtime::TypedPacked\lFunc\< Workload(const IRModule &)\>\n||}">
-<polygon fill="#ffffff" stroke="#000000" points="1300.5,-281.5 1300.5,-349.5 1496.5,-349.5 1496.5,-281.5 1300.5,-281.5"/>
-<text text-anchor="start" x="1308.5" y="-337.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm::runtime::TypedPacked</text>
-<text text-anchor="middle" x="1398.5" y="-326.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">Func< Workload(const IRModule &)></text>
-<polyline fill="none" stroke="#000000" points="1300.5,-319.5 1496.5,-319.5 "/>
-<text text-anchor="middle" x="1398.5" y="-307.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000"> </text>
-<polyline fill="none" stroke="#000000" points="1300.5,-300.5 1496.5,-300.5 "/>
... 14048 lines suppressed ...