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 2022/10/11 21:27:39 UTC

[tvm-site] branch asf-site updated: deploying docs (apache/tvm@cd077575f365cc2e7fb20b263d7e533b202518e4)

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 234d850bef deploying docs (apache/tvm@cd077575f365cc2e7fb20b263d7e533b202518e4)
234d850bef is described below

commit 234d850befb756610256271c0cf55ff2998433bd
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Tue Oct 11 21:27:31 2022 +0000

    deploying docs (apache/tvm@cd077575f365cc2e7fb20b263d7e533b202518e4)
---
 docs/_images/sphx_glr_micro_train_001.png          |  Bin 324216 -> 332672 bytes
 docs/_images/sphx_glr_micro_train_thumb.png        |  Bin 23634 -> 24174 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   |    8 +-
 .../how_to/extend_tvm/use_pass_instrument.rst.txt  |   16 +-
 .../optimize_operators/opt_conv_cuda.rst.txt       |    2 +-
 .../optimize_operators/opt_conv_tensorcore.rst.txt |    2 +-
 .../how_to/optimize_operators/opt_gemm.rst.txt     |   16 +-
 .../optimize_operators/sg_execution_times.rst.txt  |    8 +-
 .../sg_execution_times.rst.txt                     |   14 +-
 .../tune_conv2d_layer_cuda.rst.txt                 | 2491 +++++++++----------
 .../tune_network_cuda.rst.txt                      |    4 +-
 .../tune_network_x86.rst.txt                       |    4 +-
 .../tune_sparse_x86.rst.txt                        |  155 +-
 .../tune_with_autotvm/sg_execution_times.rst.txt   |    6 +-
 .../tune_with_autotvm/tune_conv2d_cuda.rst.txt     |  344 +--
 .../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 |   16 +-
 .../how_to/work_with_schedules/tensorize.rst.txt   |    2 +-
 .../tutorials/autotvm/sg_execution_times.rst.txt   |    6 +-
 .../frontend/deploy_classification.rst.txt         |    2 +-
 .../tutorials/frontend/deploy_detection.rst.txt    |    2 +-
 .../tutorials/frontend/sg_execution_times.rst.txt  |    6 +-
 .../tutorials/optimize/sg_execution_times.rst.txt  |    6 +-
 .../topic/vta/tutorials/sg_execution_times.rst.txt |    6 +-
 .../tutorial/auto_scheduler_matmul_x86.rst.txt     |    4 +-
 docs/_sources/tutorial/autotvm_matmul_x86.rst.txt  |   20 +-
 docs/_sources/tutorial/autotvm_relay_x86.rst.txt   |   60 +-
 .../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       |   40 +-
 docs/_static/documentation_options.js              |    2 +-
 docs/arch/benchmark.html                           |   11 +-
 docs/arch/convert_layout.html                      |   11 +-
 docs/arch/debugger.html                            |   11 +-
 docs/arch/device_target_interactions.html          |   11 +-
 docs/arch/frontend/tensorflow.html                 |   11 +-
 docs/arch/hybrid_script.html                       |   11 +-
 docs/arch/index.html                               |   11 +-
 docs/arch/inferbound.html                          |   11 +-
 .../arch/introduction_to_module_serialization.html |   11 +-
 docs/arch/microtvm_design.html                     |   11 +-
 docs/arch/microtvm_project_api.html                |   11 +-
 docs/arch/model_library_format.html                |   11 +-
 docs/arch/pass_infra.html                          |   11 +-
 docs/arch/relay_intro.html                         |   11 +-
 docs/arch/relay_op_strategy.html                   |   11 +-
 docs/arch/runtime.html                             |   11 +-
 docs/arch/runtimes/vulkan.html                     |   11 +-
 docs/arch/security.html                            |   11 +-
 docs/arch/virtual_machine.html                     |   11 +-
 docs/commit_hash                                   |    2 +-
 docs/contribute/ci.html                            |   11 +-
 docs/contribute/code_guide.html                    |   11 +-
 docs/contribute/code_review.html                   |   11 +-
 docs/contribute/committer_guide.html               |   11 +-
 docs/contribute/community.html                     |   11 +-
 docs/contribute/document.html                      |   11 +-
 docs/contribute/error_handling.html                |   11 +-
 docs/contribute/git_howto.html                     |   11 +-
 docs/contribute/index.html                         |   11 +-
 docs/contribute/pull_request.html                  |   11 +-
 docs/contribute/release_process.html               |   11 +-
 docs/dev/how_to/debugging_tvm.html                 |   11 +-
 docs/dev/how_to/how_to.html                        |   11 +-
 docs/dev/how_to/pytest_target_parametrization.html |   11 +-
 docs/dev/how_to/relay_add_op.html                  |   11 +-
 docs/dev/how_to/relay_add_pass.html                |   11 +-
 docs/dev/how_to/relay_bring_your_own_codegen.html  |   11 +-
 docs/dev/tutorial/codebase_walkthrough.html        |   11 +-
 docs/dev/tutorial/index.html                       |   11 +-
 docs/errors.html                                   |   11 +-
 docs/faq.html                                      |   11 +-
 docs/genindex.html                                 |   11 +-
 docs/how_to/compile_models/from_coreml.html        |   11 +-
 docs/how_to/compile_models/from_darknet.html       |   13 +-
 docs/how_to/compile_models/from_keras.html         |   13 +-
 docs/how_to/compile_models/from_mxnet.html         |   13 +-
 docs/how_to/compile_models/from_oneflow.html       |   18 +-
 docs/how_to/compile_models/from_onnx.html          |   11 +-
 docs/how_to/compile_models/from_paddle.html        |   11 +-
 docs/how_to/compile_models/from_pytorch.html       |   18 +-
 docs/how_to/compile_models/from_tensorflow.html    |   13 +-
 docs/how_to/compile_models/from_tflite.html        |   11 +-
 docs/how_to/compile_models/index.html              |   11 +-
 docs/how_to/compile_models/sg_execution_times.html |   33 +-
 docs/how_to/deploy/android.html                    |   11 +-
 docs/how_to/deploy/arm_compute_lib.html            |   11 +-
 docs/how_to/deploy/bnns.html                       |   11 +-
 docs/how_to/deploy/cpp_deploy.html                 |   11 +-
 docs/how_to/deploy/hls.html                        |   11 +-
 docs/how_to/deploy/index.html                      |   11 +-
 docs/how_to/deploy/integrate.html                  |   11 +-
 docs/how_to/deploy/tensorrt.html                   |   11 +-
 docs/how_to/deploy/vitis_ai.html                   |   11 +-
 .../deploy_models/deploy_model_on_android.html     |   13 +-
 .../how_to/deploy_models/deploy_model_on_nano.html |   11 +-
 .../how_to/deploy_models/deploy_model_on_rasp.html |   11 +-
 .../deploy_object_detection_pytorch.html           |   45 +-
 docs/how_to/deploy_models/deploy_prequantized.html |   19 +-
 .../deploy_models/deploy_prequantized_tflite.html  |   15 +-
 docs/how_to/deploy_models/deploy_quantized.html    |   13 +-
 docs/how_to/deploy_models/deploy_sparse.html       |   11 +-
 docs/how_to/deploy_models/deploy_ssd_gluoncv.html  |   50 +-
 docs/how_to/deploy_models/index.html               |   11 +-
 docs/how_to/deploy_models/sg_execution_times.html  |   29 +-
 .../extend_tvm/bring_your_own_datatypes.html       |   13 +-
 docs/how_to/extend_tvm/index.html                  |   11 +-
 docs/how_to/extend_tvm/low_level_custom_pass.html  |   11 +-
 docs/how_to/extend_tvm/sg_execution_times.html     |   19 +-
 docs/how_to/extend_tvm/use_pass_infra.html         |   11 +-
 docs/how_to/extend_tvm/use_pass_instrument.html    |   27 +-
 docs/how_to/index.html                             |   11 +-
 docs/how_to/optimize_operators/index.html          |   11 +-
 docs/how_to/optimize_operators/opt_conv_cuda.html  |   13 +-
 .../optimize_operators/opt_conv_tensorcore.html    |   13 +-
 docs/how_to/optimize_operators/opt_gemm.html       |   27 +-
 .../optimize_operators/sg_execution_times.html     |   19 +-
 docs/how_to/profile/index.html                     |   11 +-
 docs/how_to/profile/papi.html                      |   11 +-
 docs/how_to/tune_with_autoscheduler/index.html     |   11 +-
 .../sg_execution_times.html                        |   25 +-
 .../tune_conv2d_layer_cuda.html                    | 2502 +++++++++-----------
 .../tune_with_autoscheduler/tune_network_arm.html  |   11 +-
 .../tune_with_autoscheduler/tune_network_cuda.html |   15 +-
 .../tune_with_autoscheduler/tune_network_mali.html |   11 +-
 .../tune_with_autoscheduler/tune_network_x86.html  |   15 +-
 .../tune_with_autoscheduler/tune_sparse_x86.html   |  166 +-
 docs/how_to/tune_with_autotvm/index.html           |   11 +-
 .../tune_with_autotvm/sg_execution_times.html      |   17 +-
 .../how_to/tune_with_autotvm/tune_conv2d_cuda.html |  355 +--
 docs/how_to/tune_with_autotvm/tune_relay_arm.html  |   11 +-
 docs/how_to/tune_with_autotvm/tune_relay_cuda.html |   11 +-
 .../tune_with_autotvm/tune_relay_mobile_gpu.html   |   11 +-
 docs/how_to/tune_with_autotvm/tune_relay_x86.html  |   11 +-
 docs/how_to/work_with_microtvm/index.html          |   11 +-
 docs/how_to/work_with_microtvm/micro_aot.html      |   11 +-
 docs/how_to/work_with_microtvm/micro_autotune.html |   27 +-
 docs/how_to/work_with_microtvm/micro_ethosu.html   |   11 +-
 .../work_with_microtvm/micro_reference_vm.html     |   11 +-
 docs/how_to/work_with_microtvm/micro_tflite.html   |   11 +-
 docs/how_to/work_with_microtvm/micro_train.html    |   27 +-
 docs/how_to/work_with_microtvm/micro_tvmc.html     |   11 +-
 .../work_with_microtvm/sg_execution_times.html     |   21 +-
 docs/how_to/work_with_relay/build_gcn.html         |   11 +-
 docs/how_to/work_with_relay/index.html             |   11 +-
 .../how_to/work_with_relay/sg_execution_times.html |   19 +-
 .../how_to/work_with_relay/using_external_lib.html |   11 +-
 .../work_with_relay/using_pipeline_executor.html   |   11 +-
 docs/how_to/work_with_relay/using_relay_viz.html   |   11 +-
 docs/how_to/work_with_schedules/extern_op.html     |   11 +-
 docs/how_to/work_with_schedules/index.html         |   11 +-
 docs/how_to/work_with_schedules/intrin_math.html   |   13 +-
 docs/how_to/work_with_schedules/reduction.html     |   11 +-
 docs/how_to/work_with_schedules/scan.html          |   11 +-
 .../work_with_schedules/schedule_primitives.html   |   11 +-
 .../work_with_schedules/sg_execution_times.html    |   27 +-
 docs/how_to/work_with_schedules/tedd.html          |   11 +-
 docs/how_to/work_with_schedules/tensorize.html     |   13 +-
 docs/how_to/work_with_schedules/tuple_inputs.html  |   11 +-
 docs/index.html                                    |   11 +-
 docs/install/docker.html                           |   11 +-
 docs/install/from_source.html                      |   11 +-
 docs/install/index.html                            |   11 +-
 docs/install/nnpack.html                           |   11 +-
 docs/objects.inv                                   |  Bin 23564 -> 23564 bytes
 docs/py-modindex.html                              |   11 +-
 docs/reference/api/doxygen/c__runtime__api_8h.html |    4 +-
 .../api/doxygen/c__runtime__api_8h_source.html     |    2 +-
 docs/reference/api/links.html                      |   11 +-
 docs/reference/api/python/auto_scheduler.html      |   15 +-
 docs/reference/api/python/autotvm.html             |   11 +-
 docs/reference/api/python/contrib.html             |   11 +-
 docs/reference/api/python/driver.html              |   11 +-
 docs/reference/api/python/error.html               |   11 +-
 docs/reference/api/python/graph_executor.html      |   11 +-
 docs/reference/api/python/index.html               |   11 +-
 docs/reference/api/python/ir.html                  |   11 +-
 docs/reference/api/python/micro.html               |   11 +-
 docs/reference/api/python/ndarray.html             |   11 +-
 docs/reference/api/python/relay/analysis.html      |   11 +-
 docs/reference/api/python/relay/backend.html       |   11 +-
 .../api/python/relay/dataflow_pattern.html         |   11 +-
 docs/reference/api/python/relay/frontend.html      |   11 +-
 docs/reference/api/python/relay/image.html         |   11 +-
 docs/reference/api/python/relay/index.html         |   11 +-
 docs/reference/api/python/relay/nn.html            |   11 +-
 docs/reference/api/python/relay/testing.html       |   11 +-
 docs/reference/api/python/relay/transform.html     |   11 +-
 docs/reference/api/python/relay/vision.html        |   11 +-
 docs/reference/api/python/rpc.html                 |   11 +-
 docs/reference/api/python/runtime.html             |   11 +-
 docs/reference/api/python/target.html              |   11 +-
 docs/reference/api/python/te.html                  |   11 +-
 docs/reference/api/python/tir.html                 |   11 +-
 docs/reference/api/python/topi.html                |   11 +-
 docs/reference/api/python/vta/index.html           |   11 +-
 .../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/reference/langref/hybrid_script.html          |   11 +-
 docs/reference/langref/index.html                  |   11 +-
 docs/reference/langref/relay_adt.html              |   11 +-
 docs/reference/langref/relay_expr.html             |   11 +-
 docs/reference/langref/relay_op.html               |   11 +-
 docs/reference/langref/relay_pattern.html          |   11 +-
 docs/reference/langref/relay_type.html             |   11 +-
 docs/reference/publications.html                   |   11 +-
 docs/search.html                                   |   11 +-
 docs/searchindex.js                                |    2 +-
 docs/topic/microtvm/index.html                     |   11 +-
 docs/topic/vta/dev/config.html                     |   11 +-
 docs/topic/vta/dev/hardware.html                   |   11 +-
 docs/topic/vta/dev/index.html                      |   11 +-
 docs/topic/vta/index.html                          |   11 +-
 docs/topic/vta/install.html                        |   11 +-
 docs/topic/vta/tutorials/autotvm/index.html        |   11 +-
 .../vta/tutorials/autotvm/sg_execution_times.html  |   17 +-
 docs/topic/vta/tutorials/autotvm/tune_alu_vta.html |   11 +-
 .../vta/tutorials/autotvm/tune_relay_vta.html      |   11 +-
 .../tutorials/frontend/deploy_classification.html  |   13 +-
 .../vta/tutorials/frontend/deploy_detection.html   |   13 +-
 docs/topic/vta/tutorials/frontend/index.html       |   11 +-
 .../vta/tutorials/frontend/sg_execution_times.html |   17 +-
 docs/topic/vta/tutorials/index.html                |   11 +-
 docs/topic/vta/tutorials/matrix_multiply.html      |   11 +-
 .../vta/tutorials/optimize/convolution_opt.html    |   11 +-
 docs/topic/vta/tutorials/optimize/index.html       |   11 +-
 .../tutorials/optimize/matrix_multiply_opt.html    |   11 +-
 .../vta/tutorials/optimize/sg_execution_times.html |   17 +-
 docs/topic/vta/tutorials/sg_execution_times.html   |   17 +-
 docs/topic/vta/tutorials/vta_get_started.html      |   11 +-
 docs/tutorial/auto_scheduler_matmul_x86.html       |   15 +-
 docs/tutorial/autotvm_matmul_x86.html              |   31 +-
 docs/tutorial/autotvm_relay_x86.html               |  283 +--
 docs/tutorial/cross_compilation_and_rpc.html       |   13 +-
 docs/tutorial/index.html                           |   11 +-
 docs/tutorial/install.html                         |   11 +-
 docs/tutorial/intro_topi.html                      |   13 +-
 docs/tutorial/introduction.html                    |   11 +-
 docs/tutorial/relay_quick_start.html               |   11 +-
 docs/tutorial/sg_execution_times.html              |   39 +-
 docs/tutorial/tensor_expr_get_started.html         |   51 +-
 docs/tutorial/tensor_ir_blitz_course.html          |   11 +-
 docs/tutorial/tvmc_command_line_driver.html        |   11 +-
 docs/tutorial/tvmc_python.html                     |   11 +-
 docs/tutorial/uma.html                             |   11 +-
 287 files changed, 4981 insertions(+), 4855 deletions(-)

diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 58230570fb..3f59a37dab 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 c7f45c5bc4..6746d44e45 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 57bb8d9415..3153d30f4a 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  11.050 seconds)
+   **Total running time of the script:** ( 1 minutes  8.358 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 30629a6d70..112869d658 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 1s/step
+
    1/1 [==============================] - ETA: 0s
    1/1 [==============================] - 1s 963ms/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 9c4004c6b2..4c6ec542ce 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.zipa1c06891-93d7-4365-abc9-377bcd6779e3 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip360dc28d-d1b8-47e5-843b-e602cdf08c95 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 7d1351024c..f8cda04cd2 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]
     39%|###8      | 16.0M/41.5M [00:00<00:00, 136MB/s]
     70%|######9   | 29.0M/41.5M [00:00<00:00, 132MB/s]
    100%|##########| 41.5M/41.5M [00:00<00:00, 117MB/s]
+
      0%|          | 0.00/41.5M [00:00<?, ?B/s]
     31%|###1      | 13.0M/41.5M [00:00<00:00, 136MB/s]
     63%|######2   | 26.0M/41.5M [00:00<00:00, 100MB/s]
     87%|########6 | 36.1M/41.5M [00:00<00:00, 97.2MB/s]
    100%|##########| 41.5M/41.5M [00:00<00:00, 92.6MB/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 66349a52c0..dd4dc32c39 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]
     38%|###8      | 17.0M/44.7M [00:00<00:00, 177MB/s]
     76%|#######5  | 33.9M/44.7M [00:00<00:00, 122MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 120MB/s]
+
      0%|          | 0.00/44.7M [00:00<?, ?B/s]
     28%|##8       | 12.7M/44.7M [00:00<00:00, 133MB/s]
     57%|#####6    | 25.4M/44.7M [00:00<00:00, 117MB/s]
     82%|########2 | 36.7M/44.7M [00:00<00:00, 100MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 110MB/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 12f8ac9aa4..8948bad20e 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  13.246 seconds)
+   **Total running time of the script:** ( 1 minutes  9.408 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 a6d52a5dbd..630ba0b89a 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
 =================
-**05:51.552** total execution time for **how_to_compile_models** files:
+**05:37.660** total execution time for **how_to_compile_models** files:
 
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:13.246 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:09.408 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:11.050 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:08.358 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:47.116 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:45.374 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:32.464 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:31.466 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:30.984 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:30.137 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:27.484 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:26.074 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:24.871 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:25.004 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:22.612 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:21.674 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:19.266 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:17.808 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.460 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.358 | 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 a385e1a28d..4e5bc96ec2 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)  
-      16.3367      16.3350      16.4466      16.2776       0.0429   
+      16.4607      16.6100      16.9983      15.9131       0.4308   
                
 
 
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 92584d300d..9d69954200 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]
      6%|5         | 10.1M/170M [00:00<00:01, 94.5MB/s]
     14%|#3        | 23.0M/170M [00:00<00:01, 118MB/s] 
     20%|##        | 34.4M/170M [00:00<00:01, 109MB/s]
     26%|##6       | 44.9M/170M [00:00<00:01, 106MB/s]
     32%|###2      | 55.0M/170M [00:00<00:01, 104MB/s]
     38%|###8      | 64.9M/170M [00:00<00:01, 103MB/s]
     44%|####4     | 74.7M/170M [00:00<00:00, 101MB/s]
     50%|####9     | 84.4M/170M [00:00<00:00, 101MB/s]
     55%|#####5    | 94.2M/170M [00:00<00:00, 101MB/s]
     61%|######1   | 104M/170M [00:01<00:00, 99.7MB/s]
     67%|######6   | 114M/170M [00:01<00:00, 101MB/s] 
     73%|#######2  | 123M/170M [00:01<00:00, 101MB/s]
     78%|#######8  | 133M/170M [00:01<00:00, 98.0MB/s]
     84%|########4 | 143M/170M [00:01<00:00, 101MB/s] 
     90%|######### | 153M/170M [00:01<00:00, 101MB/s]
     96%|#########5| 163M/170M [00:01<00:00, 100MB/s]
    100%|##########| 170M/170M [00:01<00:00, 102MB/s]
+
      0%|          | 0.00/170M [00:00<?, ?B/s]
      9%|9         | 15.6M/170M [00:00<00:00, 164MB/s]
     18%|#8        | 31.2M/170M [00:00<00:01, 119MB/s]
     25%|##5       | 43.2M/170M [00:00<00:01, 111MB/s]
     32%|###1      | 54.1M/170M [00:00<00:01, 98.8MB/s]
     39%|###8      | 66.1M/170M [00:00<00:01, 107MB/s] 
     45%|####5     | 76.6M/170M [00:00<00:01, 83.6MB/s]
     55%|#####4    | 93.0M/170M [00:00<00:00, 105MB/s] 
     61%|######1   | 104M/170M [00:01<00:00, 107MB/s] 
     68%|######7   | 115M/170M [00:01<00:00, 106MB/s]
     74%|#######4  | 126M/170M [00:01<00:00, 104MB/s]
     80%|########  | 136M/170M [00:01<00:00, 101MB/s]
     86%|########6 | 146M/170M [00:01<00:00, 101MB/s]
     92%|#########2| 156M/170M [00:01<00:00, 103MB/s]
     98%|#########7| 166M/170M [00:01<00:00, 102MB/s]
    100%|##########| 170M/170M [00:01<00:00, 103MB/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  22.709 seconds)
+   **Total running time of the script:** ( 3 minutes  13.999 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 64e39c0b6e..310dc55ca7 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  | 9.95M/13.6M [00:00<00:00, 104MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 106MB/s]
+
      0%|          | 0.00/13.6M [00:00<?, ?B/s]
     59%|#####8    | 7.99M/13.6M [00:00<00:00, 83.7MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 95.2MB/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)  
-      90.6393      90.5738      92.2934      90.3124       0.3266   
+      90.3089      90.2385      91.0332      89.9026       0.2178   
                
 
 
@@ -463,7 +463,7 @@ TODO
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  23.780 seconds)
+   **Total running time of the script:** ( 1 minutes  20.538 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 587e2bdd2b..b157881124 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)  
-      123.3837     123.2487     129.0543     122.0951      0.6967   
+      119.4024     119.3994     120.7627     118.5256      0.4341   
                
 
 
@@ -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:** ( 1 minutes  57.380 seconds)
+   **Total running time of the script:** ( 1 minutes  57.169 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 cfc38fdc24..82ce23ab64 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  17.982 seconds)
+   **Total running time of the script:** ( 1 minutes  18.619 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 729aabd784..34305a59a3 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]
      3%|3         | 4247/132723 [00:00<00:03, 42463.36KB/s]
      8%|8         | 10914/132723 [00:00<00:02, 56694.36KB/s]
     14%|#3        | 18483/132723 [00:00<00:01, 65350.78KB/s]
     20%|#9        | 26049/132723 [00:00<00:01, 69418.39KB/s]
     25%|##5       | 33535/132723 [00:00<00:01, 71378.07KB/s]
     31%|###       | 41122/132723 [00:00<00:01, 72900.80KB/s]
     37%|###6      | 48769/132723 [00:00<00:01, 74066.15KB/s]
     42%|####2     | 56311/132723 [00:00<00:01, 74495.72KB/s]
     48%|####8     | 63891/132723 [00:00<00:00, 74900.54KB/s]
     54%|#####3    | 71573/132723 [00:01<00:00, 75491.70KB/s]
     60%|#####9    | 79208/132723 [00:01<00:00, 75750.19KB/s]
     65%|######5   | 86784/132723 [00:01<00:00, 75638.49KB/s]
     71%|#######1  | 94348/132723 [00:01<00:00, 75430.55KB/s]
     77%|#######6  | 101927/132723 [00:01<00:00, 75536.51KB/s]
     82%|########2 | 109481/132723 [00:01<00:00, 75330.47KB/s]
     88%|########8
  | 117046/132723 [00:01<00:00, 75423.93KB/s]
     94%|#########3| 124589/132723 [00:01<00:00, 75392.61KB/s]
    100%|#########9| 132161/132723 [00:01<00:00, 75489.36KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 73266.12KB/s]
+
      0%|          | 0/132723 [00:00<?, ?KB/s]
      3%|3         | 4246/132723 [00:00<00:03, 42454.27KB/s]
      8%|8         | 11251/132723 [00:00<00:02, 58682.07KB/s]
     13%|#3        | 17523/132723 [00:00<00:01, 60523.67KB/s]
     19%|#9        | 25482/132723 [00:00<00:01, 68048.67KB/s]
     25%|##5       | 33433/132723 [00:00<00:01, 72173.81KB/s]
     31%|###1      | 41313/132723 [00:00<00:01, 74423.00KB/s]
     37%|###7      | 49360/132723 [00:00<00:01, 76395.80KB/s]
     43%|####3     | 57461/132723 [00:00<00:00, 77861.30KB/s]
     49%|####9     | 65496/132723 [00:00<00:00, 78637.38KB/s]
     55%|#####5    | 73531/132723 [00:01<00:00, 79160.48KB/s]
     61%|######1   | 81579/132723 [00:01<00:00, 79563.14KB/s]
     68%|######7   | 89680/132723 [00:01<00:00, 80002.17KB/s]
     74%|#######3  | 97681/132723 [00:01<00:00, 79978.00KB/s]
     80%|#######9  | 105679/132723 [00:01<00:00, 79769.93KB/s]
     86%|########5 | 113657/132723 [00:01<00:00, 79676.52KB/s]
     92%|#########
 1| 121713/132723 [00:01<00:00, 79937.75KB/s]
     98%|#########7| 129720/132723 [00:01<00:00, 79974.61KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 76338.18KB/s]
 
 
 
@@ -242,7 +242,7 @@ Display result
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  3.620 seconds)
+   **Total running time of the script:** ( 2 minutes  55.549 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 791102fb47..5072a96889 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
 =================
-**12:33.619** total execution time for **how_to_deploy_models** files:
+**12:11.026** 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``) | 03:22.709 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:13.999 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 03:03.620 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 02:55.549 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 01:57.380 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 01:57.169 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:23.780 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:20.538 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:17.982 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:18.619 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:37.144 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:35.487 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:25.669 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:25.094 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:25.329 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:24.564 | 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 935fc701ca..3e8f2a9239 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.zipf45eae96-4cc9-400c-8240-b8a923568747 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+    Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip3503d0bd-f202-4194-8f6a-5d7cb2e726e8 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 7fa93e3584..2a9800554c 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
 =================
-**00:48.829** total execution time for **how_to_extend_tvm** files:
+**00:46.929** 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``) | 00:45.210 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:43.432 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.518 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.439 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.093 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.050 | 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 563f1e2c78..354cc2de68 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: 7310us [7310us] (46.99%; 46.99%)
-    FoldScaleAxis: 8245us [7us] (53.01%; 53.01%)
-            FoldConstant: 8238us [1682us] (52.96%; 99.91%)
-                    InferType: 6556us [6556us] (42.15%; 79.58%)
+    InferType: 6841us [6841us] (46.30%; 46.30%)
+    FoldScaleAxis: 7935us [6us] (53.70%; 53.70%)
+            FoldConstant: 7929us [1650us] (53.66%; 99.93%)
+                    InferType: 6280us [6280us] (42.50%; 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: 6729us [6729us] (44.76%; 44.76%)
-    FoldScaleAxis: 8304us [8us] (55.24%; 55.24%)
-            FoldConstant: 8297us [1723us] (55.19%; 99.91%)
-                    InferType: 6574us [6574us] (43.73%; 79.23%)
+    InferType: 6329us [6329us] (42.30%; 42.30%)
+    FoldScaleAxis: 8632us [5us] (57.70%; 57.70%)
+            FoldConstant: 8628us [1690us] (57.67%; 99.95%)
+                    InferType: 6938us [6938us] (46.37%; 80.41%)
 
 
 
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 3903b9add1..33ba0d09ab 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -340,7 +340,7 @@ latency of convolution.
 
  .. code-block:: none
 
-    Convolution: 48.016288 ms
+    Convolution: 34.771232 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 c80f084aa8..1d5da45fad 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: 10.185052 ms
+    conv2d with tensor core: 13.361171 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 89f59d6c16..887ad8e010 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: 0.020022
-    Baseline: 3.553984
+    Numpy running time: 0.018603
+    Baseline: 3.452391
 
 
 
@@ -239,7 +239,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
 
  .. code-block:: none
 
-    Opt1: 0.333014
+    Opt1: 0.309239
 
 
 
@@ -342,7 +342,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
 
  .. code-block:: none
 
-    Opt2: 0.354947
+    Opt2: 0.342060
 
 
 
@@ -438,7 +438,7 @@ the access pattern for A matrix is more cache friendly.
 
  .. code-block:: none
 
-    Opt3: 0.136089
+    Opt3: 0.115616
 
 
 
@@ -563,7 +563,7 @@ flattening.
 
  .. code-block:: none
 
-    Opt4: 0.111404
+    Opt4: 0.110118
 
 
 
@@ -685,7 +685,7 @@ write to C when all the block results are ready.
 
  .. code-block:: none
 
-    Opt5: 0.113366
+    Opt5: 0.112093
 
 
 
@@ -810,7 +810,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
 
  .. code-block:: none
 
-    Opt6: 0.148662
+    Opt6: 0.147396
 
 
 
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 fe0f713108..753e1baf44 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
 =================
-**00:36.294** total execution time for **how_to_optimize_operators** files:
+**00:35.052** total execution time for **how_to_optimize_operators** files:
 
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:33.906 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:32.687 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.331 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.354 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.058 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.012 | 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 b7ee24e147..6878d93aef 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
 =================
-**08:52.975** total execution time for **how_to_tune_with_autoscheduler** files:
+**08:32.196** 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``) | 05:22.781 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 05:08.147 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:34.218 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:32.684 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:04.203 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:02.782 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:29.074 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:27.068 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:11.443 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:10.772 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:11.256 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:10.743 | 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 d541673cd6..7365f35cbb 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
@@ -240,11 +240,11 @@ 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" = 16;
+      attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 64;
       allocate(conv2d_nchw: Pointer(local float32), float32, [7]), 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" = 224 {
+      allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
+      allocate(kernel.shared: Pointer(shared float32), float32, [192]), storage_scope = shared;
+      attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56 {
         conv2d_nchw_1: Buffer(conv2d_nchw, float32, [7], [], scope="local", align=16)[0] = 0f32
         conv2d_nchw_1[1] = 0f32
         conv2d_nchw_1[2] = 0f32
@@ -252,673 +252,602 @@ cooperative fetching, unrolling and operator fusion.
         conv2d_nchw_1[4] = 0f32
         conv2d_nchw_1[5] = 0f32
         conv2d_nchw_1[6] = 0f32
-        for (rc.outer.outer: int32, 0, 32) {
-          let cse_var_1: int32 = (rc.outer.outer*144)
+        for (rc.outer.outer: int32, 0, 64) {
+          let cse_var_2: int32 = (rc.outer.outer*392)
+          let cse_var_1: int32 = (rc.outer.outer*72)
            {
-            attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224 {
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1296], [], scope="shared")[(threadIdx.x_1*36)] = 0f32
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 1)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1*4), 9)) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 7)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 2)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1*4), 9)) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 6)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 3)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1*4), 9)) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 5)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 4)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1*4), 9)) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 4)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 5)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1*4), 9)) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 3)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 6)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1*4), 9)) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 2)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 7)] = @tir.if_then_else(((1 <= floormod((threadIdx.x_1*4), 9)) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 1)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 8)] = 0f32
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 9)] = 0f32
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 10)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 1), 9)) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 7)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 11)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 1), 9)) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 6)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 12)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 1), 9)) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 5)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 13)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 1), 9)) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 4)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 14)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 1), 9)) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 3)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 15)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 1), 9)) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 2)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 16)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 1), 9)) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 1)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 17)] = 0f32
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 18)] = 0f32
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 19)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 2), 9)) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 7)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 20)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 2), 9)) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 6)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 21)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 2), 9)) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 5)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 22)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 2), 9)) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 4)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 23)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 2), 9)) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 3)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 24)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 2), 9)) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 2)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 25)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 2), 9)) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 1)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 26)] = 0f32
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 27)] = 0f32
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 28)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 3), 9)) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 7)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 29)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 3), 9)) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 6)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 30)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 3), 9)) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 5)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 31)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 3), 9)) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 4)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 32)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 3), 9)) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 3)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 33)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 3), 9)) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 2)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 34)] = @tir.if_then_else(((1 <= floormod(((threadIdx.x_1*4) + 3), 9)) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 1)], 0f32, dtype=float32)
-              }
-              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
-                pad_temp.shared_1[((threadIdx.x_1*36) + 35)] = 0f32
-              }
+            attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((7 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 7))), data[((cse_var_2 + threadIdx.x_1) - 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 8), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 8), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 8), 9)*7)) + floormod(threadIdx.x_1, 7)) - 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 4), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 4), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 4), 9)*7)) + floormod(threadIdx.x_1, 7)) - 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((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 336), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 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(((threadIdx.x_1 < 49) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 448), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1: Buffer(kernel.shared, float32, [192], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 56), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 112), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            if @tir.likely((threadIdx.x_2 < 24), dtype=bool) {
+              kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[((((blockIdx.x*36864) + cse_var_1) + (threadIdx.x_2*3)) + 32256)]
             }
-            attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            kernel.shared_1: Buffer(kernel.shared, float32, [4608], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 144)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 144))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 8), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 7), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 672), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 96), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 3)*3)) + floormod(threadIdx.x_2, 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 5), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 4), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1344), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 48), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 1), 3)*3)) + floormod(threadIdx.x_2, 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 144)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 144)) + 64512)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 8), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 7), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2688), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 96), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 3)*3)) + floormod(threadIdx.x_2, 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 5), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 4), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            kernel.shared_1[(threadIdx.x_2 + 3360)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3360), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 48), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 1), 3)*3)) + floormod(threadIdx.x_2, 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 144)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 144)) + 129024)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 8), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-            if @tir.likely((threadIdx.x_2 < 128), dtype=bool) {
-              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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 7), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((7 <= threadIdx.x_1), data[((cse_var_2 + threadIdx.x_1) - 7)], 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(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 8), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 8), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 8), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 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(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 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(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 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(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 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(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 4), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 4), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 4), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 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((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 336), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 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(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 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((threadIdx.x_1 < 49), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 448), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + 1)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[((((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 56), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3)) + 1)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 112), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + 1)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            if @tir.likely((threadIdx.x_2 < 24), dtype=bool) {
+              kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[((((blockIdx.x*36864) + cse_var_1) + (threadIdx.x_2*3)) + 32257)]
             }
-            for (rc.outer.inner: int32, 0, 2) {
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9))]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 1)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 1)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 3)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 1)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 4)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 1)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 5)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 1)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 6)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 1)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 7)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 1)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 2)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 3)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 2)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 4)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 2)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 5)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 2)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 6)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 2)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 7)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 2)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 8)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 2)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 9)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 3)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 10)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 3)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 11)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 3)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 12)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 3)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 13)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 3)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 14)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 3)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 15)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 3)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 10)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 4)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 11)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 4)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 12)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 4)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 13)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 4)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 14)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 4)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 15)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 4)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 16)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 4)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 11)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 5)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 12)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 5)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 13)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 5)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 14)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 5)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 15)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 5)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 16)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 5)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 17)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 5)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 18)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 6)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 19)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 6)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 20)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 6)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 21)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 6)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 22)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 6)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 23)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 6)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 24)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 6)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 19)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 7)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 20)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 7)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 21)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 7)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 22)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 7)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 23)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 7)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 24)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 7)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 25)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 7)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 20)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 8)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 21)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 8)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 22)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 8)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 23)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 8)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 24)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 8)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 25)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 8)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 26)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 8)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 81)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 9)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 82)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 9)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 83)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 9)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 84)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 9)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 85)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 9)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 86)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 9)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 87)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 9)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 82)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 10)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 83)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 10)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 84)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 10)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 85)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 10)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 86)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 10)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 87)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 10)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 88)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 10)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 83)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 11)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 84)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 11)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 85)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 11)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 86)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 11)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 87)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 11)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 88)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 11)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 89)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 11)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 90)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 12)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 91)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 12)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 92)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 12)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 93)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 12)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 94)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 12)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 95)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 12)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 96)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 12)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 91)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 13)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 92)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 13)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 93)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 13)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 94)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 13)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 95)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 13)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 96)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 13)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 97)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 13)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 92)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 14)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 93)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 14)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 94)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 14)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 95)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 14)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 96)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 14)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 97)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 14)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 98)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 14)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 99)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 15)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 100)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 15)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 101)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 15)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 102)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 15)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 103)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 15)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 104)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 15)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 105)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 15)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 100)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 16)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 101)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 16)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 102)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 16)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 103)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 16)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 104)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 16)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 105)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 16)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 106)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 16)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 101)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 17)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 102)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 17)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 103)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 17)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 104)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 17)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 105)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 17)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 106)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 17)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 107)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 17)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 162)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 18)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 163)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 18)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 164)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 18)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 165)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 18)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 166)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 18)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 167)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 18)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 168)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 18)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 163)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 19)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 164)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 19)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 165)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 19)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 166)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 19)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 167)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 19)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 168)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 19)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 169)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 19)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 164)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 20)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 165)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 20)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 166)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 20)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 167)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 20)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 168)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 20)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 169)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 20)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 170)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 20)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 171)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 21)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 172)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 21)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 173)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 21)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 174)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 21)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 175)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 21)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 176)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 21)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 177)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 21)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 172)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 22)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 173)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 22)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 174)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 22)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 175)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 22)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 176)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 22)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 177)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 22)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 178)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 22)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 173)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 23)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 174)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 23)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 175)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 23)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 176)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 23)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 177)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 23)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 178)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 23)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 179)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 23)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 180)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 24)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 181)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 24)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 182)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 24)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 183)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 24)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 184)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 24)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 185)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 24)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 186)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 24)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 181)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 25)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 182)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 25)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 183)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 25)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 184)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 25)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 185)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 25)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 186)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 25)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 187)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 25)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 182)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 26)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 183)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 26)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 184)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 26)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 185)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 26)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 186)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 26)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 187)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 26)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 188)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 26)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 243)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 27)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 244)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 27)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 245)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 27)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 246)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 27)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 247)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 27)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 248)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 27)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 249)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 27)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 244)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 28)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 245)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 28)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 246)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 28)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 247)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 28)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 248)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 28)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 249)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 28)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 250)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 28)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 245)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 29)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 246)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 29)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 247)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 29)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 248)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 29)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 249)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 29)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 250)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 29)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 251)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 29)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 252)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 30)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 30)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 30)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 255)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 30)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 256)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 30)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 257)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 30)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 258)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 30)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 253)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 31)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 31)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 255)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 31)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 256)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 31)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 257)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 31)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 258)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 31)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 259)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 31)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 254)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 32)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 255)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 32)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 256)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 32)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 257)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 32)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 258)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 32)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 259)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 32)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 260)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 32)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 261)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 33)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 262)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 33)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 263)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 33)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 264)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 33)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 265)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 33)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 266)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 33)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 267)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 33)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 262)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 34)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 263)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 34)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 264)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 34)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 265)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 34)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 266)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 34)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 267)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 34)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 268)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 34)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 263)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 35)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 264)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 35)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 265)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 35)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 266)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 35)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 267)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 35)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 268)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 35)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 269)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 35)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 324)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 36)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 325)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 36)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 326)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 36)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 327)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 36)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 328)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 36)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 329)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 36)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 330)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 36)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 325)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 37)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 326)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 37)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 327)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 37)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 328)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 37)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 329)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 37)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 330)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 37)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 331)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 37)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 326)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 38)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 327)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 38)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 328)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 38)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 329)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 38)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 330)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 38)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 331)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 38)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 332)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 38)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 333)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 39)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 334)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 39)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 335)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 39)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 336)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 39)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 337)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 39)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 338)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 39)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 339)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 39)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 334)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 40)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 335)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 40)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 336)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 40)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 337)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 40)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 338)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 40)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 339)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 40)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 340)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 40)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 335)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 41)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 336)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 41)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 337)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 41)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 338)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 41)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 339)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 41)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 340)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 41)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 341)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 41)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 342)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 42)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 343)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 42)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 344)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 42)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 345)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 42)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 346)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 42)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 347)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 42)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 348)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 42)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 343)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 43)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 344)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 43)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 345)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 43)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 346)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 43)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 347)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 43)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 348)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 43)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 349)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 43)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 344)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 44)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 345)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 44)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 346)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 44)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 347)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 44)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 348)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 44)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 349)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 44)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 350)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 44)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 405)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 45)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 406)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 45)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 407)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 45)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 408)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 45)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 409)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 45)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 410)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 45)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 411)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 45)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 406)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 46)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 407)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 46)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 408)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 46)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 409)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 46)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 410)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 46)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 411)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 46)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 412)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 46)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 407)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 47)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 408)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 47)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 409)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 47)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 410)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 47)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 411)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 47)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 412)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 47)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 413)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 47)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 414)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 48)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 415)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 48)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 416)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 48)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 417)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 48)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 418)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 48)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 419)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 48)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 420)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 48)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 415)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 49)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 416)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 49)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 417)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 49)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 418)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 49)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 419)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 49)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 420)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 49)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 421)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 49)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 416)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 50)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 417)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 50)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 418)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 50)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 419)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 50)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 420)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 50)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 421)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 50)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 422)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 50)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 423)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 51)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 424)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 51)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 425)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 51)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 426)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 51)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 427)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 51)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 428)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 51)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 429)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 51)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 424)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 52)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 425)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 52)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 426)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 52)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 427)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 52)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 428)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 52)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 429)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 52)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 430)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 52)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 425)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 53)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 426)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 53)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 427)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 53)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 428)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 53)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 429)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 53)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 430)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 53)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 431)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 53)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 486)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 54)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 487)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 54)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 488)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 54)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 489)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 54)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 490)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 54)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 491)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 54)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 492)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 54)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 487)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 55)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 488)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 55)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 489)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 55)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 490)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 55)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 491)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 55)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 492)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 55)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 493)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 55)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 488)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 56)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 489)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 56)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 490)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 56)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 491)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 56)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 492)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 56)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 493)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 56)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 494)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 56)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 495)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 57)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 496)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 57)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 497)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 57)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 498)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 57)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 499)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 57)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 500)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 57)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 501)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 57)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 496)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 58)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 497)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 58)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 498)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 58)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 499)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 58)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 500)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 58)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 501)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 58)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 502)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 58)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 497)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 59)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 498)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 59)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 499)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 59)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 500)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 59)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 501)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 59)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 502)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 59)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 503)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 59)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 504)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 60)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 505)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 60)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 506)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 60)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 507)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 60)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 508)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 60)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 509)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 60)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 510)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 60)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 505)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 61)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 506)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 61)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 507)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 61)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 508)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 61)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 509)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 61)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 510)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 61)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 511)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 61)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 506)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 62)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 507)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 62)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 508)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 62)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 509)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 62)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 510)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 62)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 511)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 62)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 512)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 62)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 567)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 63)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 568)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 63)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 569)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 63)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 570)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 63)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 571)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 63)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 572)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 63)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 573)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 63)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 568)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 64)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 569)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 64)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 570)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 64)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 571)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 64)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 572)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 64)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 573)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 64)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 574)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 64)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 569)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 65)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 570)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 65)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 571)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 65)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 572)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 65)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 573)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 65)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 574)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 65)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 575)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 65)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 576)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 66)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 577)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 66)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 578)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 66)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 579)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 66)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 580)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 66)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 581)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 66)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 582)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 66)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 577)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 67)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 578)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 67)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 579)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 67)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 580)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 67)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 581)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 67)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 582)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 67)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 583)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 67)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 578)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 68)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 579)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 68)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 580)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 68)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 581)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 68)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 582)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 68)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 583)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 68)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 584)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 68)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 585)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 69)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 586)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 69)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 587)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 69)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 588)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 69)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 589)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 69)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 590)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 69)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 591)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 69)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 586)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 70)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 587)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 70)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 588)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 70)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 589)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 70)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 590)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 70)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 591)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 70)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 592)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 70)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 587)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 71)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 588)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 71)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 589)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 71)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 590)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 71)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 591)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 71)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 592)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 71)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 593)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72)) + 71)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((7 <= threadIdx.x_1) && (floormod(threadIdx.x_1, 7) < 6)), data[((cse_var_2 + threadIdx.x_1) - 6)], 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 8), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 8), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 8), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 4), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 4), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 4), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 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((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 336), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 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((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 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(((threadIdx.x_1 < 49) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 448), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + 2)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[((((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 56), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3)) + 2)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 112), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + 2)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 56;
+            if @tir.likely((threadIdx.x_2 < 24), dtype=bool) {
+              kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[((((blockIdx.x*36864) + cse_var_1) + (threadIdx.x_2*3)) + 32258)]
             }
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
           }
         }
-        for (i3.inner: int32, 0, 7) {
-          compute[(((blockIdx.x*1568) + (threadIdx.x*7)) + i3.inner)] = max((conv2d_nchw_1[i3.inner] + bias[((blockIdx.x*32) + floordiv(threadIdx.x, 7))]), 0f32)
+        for (i2.inner: int32, 0, 7) {
+          compute[((((blockIdx.x*392) + (floordiv(threadIdx.x, 7)*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[i2.inner] + bias[((blockIdx.x*8) + floordiv(threadIdx.x, 7))]), 0f32)
         }
       }
     }
@@ -973,7 +902,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 0.239 ms
+    Execution time of this operator: 0.190 ms
 
 
 
@@ -1023,34 +952,34 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
     conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
     conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
-    conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=32)
+    conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
     conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
-    conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
+    conv2d_nchw_yy_o_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=7)
+    conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
     conv2d_nchw_yy_o_o_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=7)
+    conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
     conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
-    conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
+    conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
     conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
     conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=8)
-    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=1)
     conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
     conv2d_nchw_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=3)
+    conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
     conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
     s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
     compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
     compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
     compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
     compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
-    compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=32)
+    compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
     compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
-    compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
-    compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+    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)
@@ -1070,14 +999,14 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
     kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
     s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-    kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=224)
+    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=36)
+    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=224)
+    pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
     s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
-    s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+    s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
     s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
 
     CUDA source code:
@@ -1095,10 +1024,10 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
       #define int64_t long long
       #define uint64_t unsigned long long
     #endif
-    extern "C" __global__ void __launch_bounds__(224) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+    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[7];
-      __shared__ float pad_temp_shared[1296];
-      __shared__ float kernel_shared[4608];
+      __shared__ float pad_temp_shared[504];
+      __shared__ float kernel_shared[192];
       conv2d_nchw[0] = 0.000000e+00f;
       conv2d_nchw[1] = 0.000000e+00f;
       conv2d_nchw[2] = 0.000000e+00f;
@@ -1106,649 +1035,565 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
       conv2d_nchw[4] = 0.000000e+00f;
       conv2d_nchw[5] = 0.000000e+00f;
       conv2d_nchw[6] = 0.000000e+00f;
-      for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
+      for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
         __syncthreads();
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[(((int)threadIdx.x) * 36)] = 0.000000e+00f;
+        pad_temp_shared[((int)threadIdx.x)] = (((7 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((rc_outer_outer * 392) + ((int)threadIdx.x)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 56)] = ((((1 <= (((((int)threadIdx.x) / 7) + 8) % 9)) && ((((((int)threadIdx.x) / 7) + 8) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 56) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 8) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 112)] = ((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 112) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 168) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 224)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 224) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 280)] = ((((1 <= (((((int)threadIdx.x) / 7) + 4) % 9)) && ((((((int)threadIdx.x) / 7) + 4) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 280) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 4) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 336)] = ((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 336) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 392)] = ((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((int)threadIdx.x) < 49) && (1 <= (((int)threadIdx.x) % 7))) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 448) / 63) * 49)) + ((int)threadIdx.x)) - 1)] : 0.000000e+00f);
+        kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3))];
+        kernel_shared[(((int)threadIdx.x) + 56)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 56) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (((((int)threadIdx.x) + 2) % 3) * 3))];
+        kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 112) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3))];
+        if (((int)threadIdx.x) < 24) {
+          kernel_shared[(((int)threadIdx.x) + 168)] = kernel[((((((int)blockIdx.x) * 36864) + (rc_outer_outer * 72)) + (((int)threadIdx.x) * 3)) + 32256)];
         }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 1)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 7)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 2)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 6)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 3)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 5)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 4)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 4)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 5)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 3)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 6)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 2)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 7)] = (((1 <= ((((int)threadIdx.x) * 4) % 9)) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((rc_outer_outer * 784) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (((((int)threadIdx.x) * 4) % 9) * 7)) - 1)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 8)] = 0.000000e+00f;
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 9)] = 0.000000e+00f;
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 10)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 7)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 11)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 6)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 12)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 5)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 13)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 4)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 14)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 3)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 15)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 2)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 16)] = (((1 <= (((((int)threadIdx.x) * 4) + 1) % 9)) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 1) % 9) * 7)) - 1)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 17)] = 0.000000e+00f;
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 18)] = 0.000000e+00f;
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 19)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 7)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 20)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 6)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 21)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 5)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 22)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 4)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 23)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 3)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 24)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 2)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 25)] = (((1 <= (((((int)threadIdx.x) * 4) + 2) % 9)) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 2) % 9) * 7)) - 1)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 26)] = 0.000000e+00f;
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 27)] = 0.000000e+00f;
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 28)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 7)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 29)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 6)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 30)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 5)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 31)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 4)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 32)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 3)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 33)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 2)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 34)] = (((1 <= (((((int)threadIdx.x) * 4) + 3) % 9)) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((rc_outer_outer * 784) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + ((((((int)threadIdx.x) * 4) + 3) % 9) * 7)) - 1)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 36) {
-          pad_temp_shared[((((int)threadIdx.x) * 36) + 35)] = 0.000000e+00f;
-        }
-        kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) % 144))];
-        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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) + 8) % 9) / 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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) + 7) % 9) / 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) + 96) % 144) / 9) * 9)) + ((((((int)threadIdx.x) / 3) + 2) % 3) * 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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) + 5) % 9) / 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) / 9) * 9)) + ((((((int)threadIdx.x) + 4) % 9) / 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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) / 3) + 1) % 3) * 3)) + (((int)threadIdx.x) % 3))];
-        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) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
-        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) % 144) / 9) * 9)) + ((((int)threadIdx.x) + 1) % 9))];
-        kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) % 144)) + 64512)];
-        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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) + 8) % 9) / 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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) + 7) % 9) / 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) + 96) % 144) / 9) * 9)) + ((((((int)threadIdx.x) / 3) + 2) % 3) * 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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) + 5) % 9) / 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) / 9) * 9)) + ((((((int)threadIdx.x) + 4) % 9) / 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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) / 3) + 1) % 3) * 3)) + (((int)threadIdx.x) % 3))];
-        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) / 9) * 9)) + ((((int)threadIdx.x) + 2) % 9))];
-        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) % 144) / 9) * 9)) + ((((int)threadIdx.x) + 1) % 9))];
-        kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 144) * 4608)) + (rc_outer_outer * 144)) + (((int)threadIdx.x) % 144)) + 129024)];
-        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) % 144) / 9) * 9)) + ((((((int)threadIdx.x) + 8) % 9) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
-        if (((int)threadIdx.x) < 128) {
-          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) / 9) * 9)) + ((((((int)threadIdx.x) + 7) % 9) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        __syncthreads();
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 49)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 49)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 112)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 112)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 175)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 175)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 238)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 238)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 301)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 301)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 364)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 364)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 427)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 427)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 490)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 490)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 497)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        __syncthreads();
+        pad_temp_shared[((int)threadIdx.x)] = ((7 <= ((int)threadIdx.x)) ? data[(((rc_outer_outer * 392) + ((int)threadIdx.x)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 56)] = (((1 <= (((((int)threadIdx.x) / 7) + 8) % 9)) && ((((((int)threadIdx.x) / 7) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 56) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 8) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 112)] = (((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 112) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 168)] = (((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 168) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 224)] = (((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 224) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 280)] = (((1 <= (((((int)threadIdx.x) / 7) + 4) % 9)) && ((((((int)threadIdx.x) / 7) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 280) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 4) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 336)] = (((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 336) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 392)] = (((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 448)] = ((((int)threadIdx.x) < 49) ? data[(((rc_outer_outer * 392) + (((((int)threadIdx.x) + 448) / 63) * 49)) + ((int)threadIdx.x))] : 0.000000e+00f);
+        kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + 1)];
+        kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 56) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (((((int)threadIdx.x) + 2) % 3) * 3)) + 1)];
+        kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 112) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + 1)];
+        if (((int)threadIdx.x) < 24) {
+          kernel_shared[(((int)threadIdx.x) + 168)] = kernel[((((((int)blockIdx.x) * 36864) + (rc_outer_outer * 72)) + (((int)threadIdx.x) * 3)) + 32257)];
         }
         __syncthreads();
-        for (int rc_outer_inner = 0; rc_outer_inner < 2; ++rc_outer_inner) {
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9))] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72))]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72))]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72))]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72))]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72))]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72))]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72))]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 1)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 1)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 3)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 1)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 4)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 1)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 5)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 1)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 6)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 1)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 7)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 1)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 2)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 3)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 2)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 4)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 2)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 5)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 2)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 6)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 2)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 7)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 2)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 8)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 2)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 3)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 3)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 3)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 12)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 3)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 13)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 3)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 14)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 3)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 15)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 3)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 4)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 4)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 12)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 4)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 13)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 4)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 14)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 4)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 15)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 4)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 16)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 4)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 5)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 12)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 5)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 13)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 5)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 14)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 5)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 15)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 5)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 16)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 5)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 17)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 5)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 6)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 6)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 6)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 21)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 6)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 22)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 6)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 23)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 6)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 24)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 6)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 7)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 7)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 21)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 7)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 22)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 7)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 23)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 7)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 24)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 7)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 25)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 7)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 8)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 21)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 8)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 22)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 8)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 23)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 8)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 24)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 8)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 25)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 8)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 26)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 8)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 81)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 9)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 9)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 9)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 84)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 9)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 85)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 9)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 86)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 9)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 87)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 9)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 82)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 10)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 10)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 84)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 10)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 85)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 10)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 86)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 10)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 87)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 10)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 88)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 10)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 83)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 11)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 84)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 11)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 85)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 11)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 86)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 11)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 87)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 11)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 88)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 11)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 89)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 11)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 90)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 12)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 12)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 12)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 93)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 12)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 94)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 12)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 95)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 12)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 96)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 12)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 91)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 13)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 13)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 93)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 13)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 94)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 13)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 95)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 13)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 96)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 13)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 97)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 13)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 92)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 14)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 93)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 14)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 94)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 14)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 95)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 14)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 96)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 14)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 97)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 14)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 98)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 14)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 99)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 15)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 15)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 15)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 102)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 15)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 103)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 15)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 104)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 15)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 105)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 15)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 100)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 16)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 16)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 102)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 16)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 103)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 16)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 104)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 16)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 105)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 16)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 106)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 16)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 101)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 17)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 102)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 17)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 103)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 17)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 104)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 17)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 105)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 17)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 106)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 17)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 107)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 17)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 162)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 18)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 163)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 18)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 164)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 18)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 165)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 18)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 166)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 18)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 167)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 18)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 168)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 18)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 163)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 19)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 164)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 19)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 165)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 19)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 166)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 19)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 167)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 19)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 168)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 19)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 169)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 19)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 164)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 20)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 165)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 20)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 166)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 20)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 167)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 20)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 168)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 20)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 169)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 20)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 170)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 20)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 171)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 21)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 172)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 21)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 173)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 21)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 174)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 21)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 175)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 21)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 176)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 21)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 177)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 21)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 172)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 22)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 173)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 22)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 174)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 22)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 175)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 22)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 176)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 22)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 177)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 22)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 178)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 22)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 173)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 23)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 174)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 23)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 175)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 23)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 176)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 23)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 177)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 23)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 178)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 23)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 179)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 23)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 180)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 24)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 181)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 24)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 182)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 24)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 183)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 24)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 184)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 24)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 185)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 24)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 186)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 24)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 181)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 25)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 182)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 25)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 183)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 25)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 184)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 25)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 185)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 25)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 186)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 25)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 187)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 25)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 182)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 26)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 183)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 26)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 184)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 26)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 185)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 26)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 186)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 26)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 187)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 26)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 188)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 26)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 243)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 27)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 244)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 27)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 245)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 27)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 246)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 27)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 247)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 27)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 248)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 27)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 249)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 27)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 244)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 28)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 245)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 28)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 246)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 28)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 247)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 28)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 248)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 28)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 249)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 28)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 250)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 28)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 245)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 29)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 246)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 29)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 247)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 29)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 248)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 29)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 249)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 29)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 250)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 29)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 251)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 29)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 252)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 30)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 30)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 30)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 255)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 30)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 256)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 30)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 257)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 30)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 258)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 30)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 253)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 31)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 31)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 255)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 31)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 256)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 31)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 257)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 31)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 258)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 31)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 259)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 31)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 254)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 32)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 255)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 32)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 256)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 32)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 257)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 32)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 258)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 32)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 259)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 32)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 260)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 32)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 261)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 33)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 262)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 33)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 263)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 33)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 264)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 33)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 265)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 33)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 266)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 33)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 267)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 33)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 262)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 34)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 263)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 34)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 264)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 34)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 265)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 34)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 266)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 34)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 267)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 34)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 268)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 34)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 263)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 35)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 264)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 35)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 265)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 35)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 266)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 35)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 267)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 35)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 268)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 35)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 269)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 35)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 324)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 36)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 325)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 36)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 326)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 36)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 327)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 36)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 328)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 36)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 329)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 36)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 330)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 36)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 325)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 37)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 326)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 37)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 327)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 37)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 328)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 37)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 329)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 37)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 330)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 37)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 331)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 37)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 326)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 38)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 327)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 38)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 328)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 38)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 329)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 38)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 330)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 38)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 331)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 38)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 332)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 38)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 333)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 39)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 334)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 39)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 335)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 39)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 336)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 39)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 337)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 39)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 338)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 39)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 339)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 39)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 334)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 40)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 335)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 40)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 336)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 40)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 337)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 40)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 338)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 40)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 339)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 40)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 340)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 40)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 335)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 41)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 336)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 41)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 337)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 41)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 338)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 41)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 339)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 41)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 340)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 41)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 341)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 41)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 342)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 42)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 343)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 42)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 344)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 42)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 345)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 42)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 346)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 42)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 347)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 42)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 348)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 42)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 343)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 43)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 344)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 43)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 345)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 43)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 346)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 43)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 347)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 43)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 348)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 43)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 349)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 43)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 344)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 44)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 345)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 44)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 346)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 44)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 347)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 44)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 348)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 44)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 349)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 44)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 350)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 44)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 405)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 45)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 406)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 45)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 407)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 45)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 408)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 45)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 409)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 45)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 410)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 45)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 411)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 45)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 406)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 46)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 407)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 46)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 408)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 46)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 409)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 46)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 410)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 46)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 411)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 46)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 412)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 46)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 407)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 47)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 408)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 47)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 409)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 47)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 410)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 47)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 411)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 47)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 412)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 47)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 413)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 47)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 414)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 48)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 415)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 48)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 416)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 48)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 417)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 48)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 418)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 48)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 419)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 48)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 420)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 48)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 415)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 49)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 416)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 49)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 417)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 49)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 418)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 49)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 419)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 49)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 420)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 49)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 421)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 49)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 416)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 50)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 417)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 50)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 418)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 50)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 419)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 50)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 420)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 50)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 421)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 50)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 422)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 50)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 423)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 51)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 424)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 51)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 425)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 51)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 426)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 51)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 427)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 51)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 428)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 51)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 429)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 51)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 424)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 52)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 425)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 52)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 426)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 52)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 427)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 52)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 428)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 52)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 429)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 52)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 430)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 52)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 425)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 53)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 426)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 53)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 427)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 53)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 428)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 53)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 429)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 53)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 430)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 53)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 431)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 53)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 486)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 54)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 487)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 54)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 488)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 54)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 489)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 54)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 490)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 54)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 491)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 54)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 492)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 54)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 487)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 55)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 488)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 55)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 489)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 55)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 490)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 55)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 491)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 55)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 492)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 55)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 493)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 55)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 488)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 56)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 489)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 56)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 490)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 56)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 491)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 56)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 492)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 56)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 493)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 56)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 494)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 56)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 495)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 57)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 496)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 57)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 497)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 57)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 498)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 57)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 499)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 57)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 500)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 57)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 501)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 57)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 496)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 58)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 497)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 58)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 498)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 58)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 499)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 58)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 500)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 58)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 501)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 58)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 502)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 58)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 497)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 59)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 498)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 59)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 499)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 59)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 500)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 59)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 501)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 59)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 502)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 59)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 503)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 59)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 504)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 60)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 505)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 60)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 506)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 60)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 507)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 60)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 508)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 60)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 509)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 60)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 510)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 60)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 505)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 61)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 506)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 61)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 507)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 61)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 508)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 61)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 509)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 61)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 510)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 61)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 511)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 61)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 506)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 62)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 507)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 62)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 508)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 62)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 509)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 62)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 510)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 62)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 511)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 62)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 512)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 62)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 567)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 63)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 568)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 63)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 569)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 63)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 570)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 63)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 571)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 63)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 572)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 63)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 573)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 63)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 568)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 64)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 569)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 64)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 570)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 64)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 571)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 64)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 572)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 64)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 573)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 64)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 574)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 64)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 569)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 65)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 570)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 65)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 571)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 65)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 572)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 65)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 573)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 65)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 574)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 65)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 575)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 65)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 576)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 66)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 577)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 66)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 578)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 66)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 579)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 66)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 580)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 66)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 581)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 66)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 582)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 66)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 577)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 67)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 578)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 67)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 579)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 67)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 580)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 67)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 581)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 67)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 582)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 67)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 583)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 67)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 578)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 68)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 579)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 68)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 580)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 68)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 581)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 68)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 582)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 68)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 583)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 68)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 584)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 68)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 585)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 69)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 586)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 69)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 587)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 69)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 588)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 69)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 589)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 69)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 590)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 69)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 591)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 69)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 586)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 70)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 587)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 70)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 588)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 70)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 589)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 70)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 590)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 70)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 591)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 70)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 592)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 70)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 587)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 71)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 588)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 71)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 589)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 71)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 590)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 71)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 591)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 71)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 592)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 71)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 648) + ((((int)threadIdx.x) % 7) * 9)) + 593)] * kernel_shared[((((((int)threadIdx.x) / 7) * 144) + (rc_outer_inner * 72)) + 71)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 49)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 49)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 112)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 112)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 175)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 175)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 238)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 238)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 301)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 301)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 364)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 364)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 427)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 427)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 490)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 490)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 497)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        __syncthreads();
+        pad_temp_shared[((int)threadIdx.x)] = (((7 <= ((int)threadIdx.x)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((rc_outer_outer * 392) + ((int)threadIdx.x)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 56)] = ((((1 <= (((((int)threadIdx.x) / 7) + 8) % 9)) && ((((((int)threadIdx.x) / 7) + 8) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 56) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 8) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 112)] = ((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 112) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 168)] = ((((1 <= (((((int)threadIdx.x) / 7) + 6) % 9)) && ((((((int)threadIdx.x) / 7) + 6) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 168) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 6) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 224)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 224) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 280)] = ((((1 <= (((((int)threadIdx.x) / 7) + 4) % 9)) && ((((((int)threadIdx.x) / 7) + 4) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 280) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 4) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 336)] = ((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 336) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 392)] = ((((1 <= (((((int)threadIdx.x) / 7) + 2) % 9)) && ((((((int)threadIdx.x) / 7) + 2) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 392) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 2) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((int)threadIdx.x) < 49) && ((((int)threadIdx.x) % 7) < 6)) ? data[((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 448) / 63) * 49)) + ((int)threadIdx.x)) + 1)] : 0.000000e+00f);
+        kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + 2)];
+        kernel_shared[(((int)threadIdx.x) + 56)] = kernel[((((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 56) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (((((int)threadIdx.x) + 2) % 3) * 3)) + 2)];
+        kernel_shared[(((int)threadIdx.x) + 112)] = kernel[((((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 112) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (((((int)threadIdx.x) + 1) % 3) * 3)) + 2)];
+        if (((int)threadIdx.x) < 24) {
+          kernel_shared[(((int)threadIdx.x) + 168)] = kernel[((((((int)blockIdx.x) * 36864) + (rc_outer_outer * 72)) + (((int)threadIdx.x) * 3)) + 32258)];
         }
+        __syncthreads();
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 7)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 49)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 14)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 49)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 56)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 112)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 77)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 91)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 98)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 112)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 119)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 175)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 140)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 154)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 161)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 175)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 182)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 238)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 203)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 217)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 224)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 238)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 245)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 301)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 266)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 280)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 287)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 301)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 308)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 364)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 329)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 343)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 350)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 364)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 371)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 427)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 392)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 406)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 413)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 427)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 434)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 490)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 455)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 469)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 476)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 490)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((int)threadIdx.x) % 7) + 497)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
       }
-      for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
-        compute[(((((int)blockIdx.x) * 1568) + (((int)threadIdx.x) * 7)) + i3_inner)] = max((conv2d_nchw[i3_inner] + bias[((((int)blockIdx.x) * 32) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
+      for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
+        compute[((((((int)blockIdx.x) * 392) + ((((int)threadIdx.x) / 7) * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[i2_inner] + bias[((((int)blockIdx.x) * 8) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
       }
     }
 
@@ -1810,7 +1655,7 @@ 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  22.781 seconds)
+   **Total running time of the script:** ( 5 minutes  8.147 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 0e383ba8bf..1932535e12 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)  
-       8.2464       8.2491       8.2556       8.2345       0.0088   
+       8.2569       8.2527       8.2671       8.2508       0.0073   
                
 
 
@@ -671,7 +671,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  4.203 seconds)
+   **Total running time of the script:** ( 1 minutes  2.782 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 334c64481a..c7e850f188 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)  
-      758.7186     759.2682     761.4577     755.4300      2.4913   
+      749.7971     751.1262     752.0666     746.1984      2.5734   
                
 
 
@@ -690,7 +690,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  34.218 seconds)
+   **Total running time of the script:** ( 1 minutes  32.684 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 679fe67365..cd4985a7c2 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
@@ -386,103 +386,78 @@ 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_5: placeholder_15: Buffer(placeholder_10, float32, [128, 256], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_8: placeholder_16: Buffer(placeholder_13, int32, [33], []), placeholder_7: placeholder_17: Buffer(placeholder_12, int32, [4916], []), placeholder_9: placeholder_18: Buffer(placeholder_14, float32, [128, 512], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
+      preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), placeholder_5: placeholder_18: Buffer(placeholder_10, float32, [128, 256], []), placeholder_9: placeholder_19: Buffer(placeholder_14, float32, [128, 512], [])} {
       for (i0.outer.i1.outer.fused: int32, 0, 32) "parallel" {
         allocate(compute_4: Pointer(global float32), float32, [2048]), storage_scope = global {
           for (i.outer.inner: int32, 0, 2) {
-            for (i.inner.init: int32, 0, 64) {
-              let cse_var_1: int32 = ((i.outer.inner*1024) + (i.inner.init*16))
-               {
-                compute_5: Buffer(compute_4, float32, [2048], [])[cse_var_1] = 0f32
-                compute_5[(cse_var_1 + 1)] = 0f32
-                compute_5[(cse_var_1 + 2)] = 0f32
-                compute_5[(cse_var_1 + 3)] = 0f32
-                compute_5[(cse_var_1 + 4)] = 0f32
-                compute_5[(cse_var_1 + 5)] = 0f32
-                compute_5[(cse_var_1 + 6)] = 0f32
-                compute_5[(cse_var_1 + 7)] = 0f32
-                compute_5[(cse_var_1 + 8)] = 0f32
-                compute_5[(cse_var_1 + 9)] = 0f32
-                compute_5[(cse_var_1 + 10)] = 0f32
-                compute_5[(cse_var_1 + 11)] = 0f32
-                compute_5[(cse_var_1 + 12)] = 0f32
-                compute_5[(cse_var_1 + 13)] = 0f32
-                compute_5[(cse_var_1 + 14)] = 0f32
-                compute_5[(cse_var_1 + 15)] = 0f32
-              }
-            }
-            for (elem_idx: int32, 0, (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])) {
-              for (i.inner: int32, 0, 64) {
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_2: int32 = ((i.outer.inner*1024) + (i.inner*16))
-                  compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16))]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_3: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 1)
-                  compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 1)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_4: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 2)
-                  compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 2)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_5: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 3)
-                  compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 3)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_6: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 4)
-                  compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 4)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_7: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 5)
-                  compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 5)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_8: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 6)
-                  compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 6)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_9: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 7)
-                  compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 7)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+            for (nb_j.inner: int32, 0, 2) {
+              for (i.inner.init: int32, 0, 32) {
+                let cse_var_1: int32 = (((i.outer.inner*1024) + (i.inner.init*32)) + (nb_j.inner*16))
+                 {
+                  compute_5: Buffer(compute_4, float32, [2048], [])[cse_var_1] = 0f32
+                  compute_5[(cse_var_1 + 1)] = 0f32
+                  compute_5[(cse_var_1 + 2)] = 0f32
+                  compute_5[(cse_var_1 + 3)] = 0f32
+                  compute_5[(cse_var_1 + 4)] = 0f32
+                  compute_5[(cse_var_1 + 5)] = 0f32
+                  compute_5[(cse_var_1 + 6)] = 0f32
+                  compute_5[(cse_var_1 + 7)] = 0f32
+                  compute_5[(cse_var_1 + 8)] = 0f32
+                  compute_5[(cse_var_1 + 9)] = 0f32
+                  compute_5[(cse_var_1 + 10)] = 0f32
+                  compute_5[(cse_var_1 + 11)] = 0f32
+                  compute_5[(cse_var_1 + 12)] = 0f32
+                  compute_5[(cse_var_1 + 13)] = 0f32
+                  compute_5[(cse_var_1 + 14)] = 0f32
+                  compute_5[(cse_var_1 + 15)] = 0f32
                 }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_10: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 8)
-                  compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 8)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_11: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 9)
-                  compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 9)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_12: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 10)
-                  compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 10)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_13: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 11)
-                  compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 11)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_14: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 12)
-                  compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 12)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_15: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 13)
-                  compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 13)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_16: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 14)
-                  compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 14)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i0.outer.i1.outer.fused + 1)] - placeholder_3[i0.outer.i1.outer.fused])), dtype=bool) {
-                  let cse_var_17: int32 = (((i.outer.inner*1024) + (i.inner*16)) + 15)
-                  compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[i0.outer.i1.outer.fused]*16) + (elem_idx*16)) + 15)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[i0.outer.i1.outer.fused] + elem_idx)])], 0f32)))
+              }
+              for (elem_idx: int32, 0, let cse_var_2: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_2 + 1)] - placeholder_3[cse_var_2])) {
+                for (i.inner: int32, 0, 32) {
+                  let cse_var_21: int32 = (elem_idx*16)
+                  let cse_var_20: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
+                  let cse_var_19: int32 = (((i.outer.inner*1024) + (i.inner*32)) + (nb_j.inner*16))
+                  let cse_var_18: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*16384) + (i.outer.inner*8192)) + (i.inner*256))
+                  let cse_var_17: int32 = (cse_var_19 + 9)
+                  let cse_var_16: int32 = (cse_var_19 + 8)
+                  let cse_var_15: int32 = (cse_var_19 + 7)
+                  let cse_var_14: int32 = (cse_var_19 + 6)
+                  let cse_var_13: int32 = (cse_var_19 + 5)
+                  let cse_var_12: int32 = (cse_var_19 + 4)
+                  let cse_var_11: int32 = (cse_var_19 + 3)
+                  let cse_var_10: int32 = (cse_var_19 + 2)
+                  let cse_var_9: int32 = (cse_var_19 + 15)
+                  let cse_var_8: int32 = (cse_var_19 + 14)
+                  let cse_var_7: int32 = (cse_var_19 + 13)
+                  let cse_var_6: int32 = (cse_var_19 + 12)
+                  let cse_var_5: int32 = (cse_var_19 + 11)
+                  let cse_var_4: int32 = (cse_var_19 + 10)
+                  let cse_var_3: int32 = (cse_var_19 + 1)
+                   {
+                    compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[((placeholder_3[cse_var_20]*16) + cse_var_21)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 1)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 2)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 3)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 4)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 5)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 6)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 7)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 8)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 9)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 10)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 11)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 12)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 13)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 14)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                    compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 15)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
+                  }
                 }
               }
             }
           }
-          for (i0.inner: int32, 0, 128) {
-            let cse_var_18: int32 = ((i0.inner*512) + (i0.outer.i1.outer.fused*16))
-            compute[ramp(cse_var_18, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_18, 1, 16)]), broadcast(0f32, 16))
+          for (i0.inner: int32, 0, 64) {
+            let cse_var_22: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*32768) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
+            compute[ramp(cse_var_22, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_22, 1, 32)]), broadcast(0f32, 32))
           }
         }
       }
@@ -538,7 +513,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 1.840 ms
+    Execution time of this operator: 1.797 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 5b0043223b..fa8f117be3 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
 
 Computation times
 =================
-**00:33.651** total execution time for **how_to_tune_with_autotvm** files:
+**00:49.173** 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``)           | 00:33.616 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)           | 00:49.137 | 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 508df5ad6d..2ee9734099 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
@@ -277,8 +277,7 @@ for this template
     waiting for device...
     device available
     Get devices for measurement successfully!
-    No: 1   GFLOPS: 176.27/176.27   result: MeasureResult(costs=(0.0013133221477272727,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8962581157684326, timestamp=1665511081.8514898)      [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2715966
-    No: 2   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    No: 1   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -400,8 +399,8 @@ 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, 16, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4012103
-    No: 3   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 2, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4064011
+    No: 2   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -523,8 +522,9 @@ 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, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,380905
-    No: 4   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 128, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7546795
+    No: 3   GFLOPS: 1.99/1.99       result: MeasureResult(costs=(0.11633968350000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.6836817264556885, timestamp=1665518904.414886) [('tile_f', [-1, 2, 4, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1816748
+    No: 4   GFLOPS: 0.00/1.99       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
@@ -646,8 +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, 2, 4, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3473788
-    No: 5   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 4, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7573839
+    No: 5   GFLOPS: 31.07/31.07     result: MeasureResult(costs=(0.007451583857142857,), error_no=MeasureErrorNo.NO_ERROR, all_cost=6.13435697555542, timestamp=1665518910.7434855) [('tile_f', [-1, 1, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5049164
+    No: 6   GFLOPS: 2.10/31.07      result: MeasureResult(costs=(0.11046655250000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.066395998001099, timestamp=1665518913.418771)  [('tile_f', [-1, 4, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1247677
+    No: 7   GFLOPS: 0.00/31.07      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
@@ -769,8 +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, 32, 1, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3064081
-    No: 6   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10014952
+    No: 8   GFLOPS: 13.34/31.07     result: MeasureResult(costs=(0.017360239500000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3935635089874268, timestamp=1665518915.0651538)       [('tile_f', [-1, 4, 8, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5441781
+    No: 9   GFLOPS: 0.00/31.07      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
@@ -892,8 +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, 16, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3631659
-    No: 7   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 1, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3218960
+    No: 10  GFLOPS: 0.00/31.07      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
@@ -1015,8 +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, 4, 1, 128]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10371232
-    No: 8   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 256, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2293599
+    No: 11  GFLOPS: 0.00/31.07      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
@@ -1138,8 +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, 16, 8, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,792251
-    No: 9   GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 32, 16, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,726699
+    No: 12  GFLOPS: 3.44/31.07      result: MeasureResult(costs=(0.06723238349999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.833800315856934, timestamp=1665518924.098887)  [('tile_f', [-1, 1, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9555014
+    No: 13  GFLOPS: 0.00/31.07      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
@@ -1261,8 +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, 2, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4944430
-    No: 10  GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 1, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5482759
+    No: 14  GFLOPS: 0.00/31.07      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
@@ -1384,8 +1388,8 @@ 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, 8, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4291254
-    No: 11  GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 1, 128]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10344172
+    No: 15  GFLOPS: 0.00/31.07      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
@@ -1507,10 +1511,8 @@ 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, 32, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8241991
-    No: 12  GFLOPS: 1.42/176.27     result: MeasureResult(costs=(0.1628832855,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.4308838844299316, timestamp=1665511088.6105936)       [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5457705
-    No: 13  GFLOPS: 78.24/176.27    result: MeasureResult(costs=(0.0029589941714285714,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2493884563446045, timestamp=1665511091.2689445)      [('tile_f', [-1, 1, 1, 8]), ('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', 0)],None,2152176
-    No: 14  GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,521982
+    No: 16  GFLOPS: 0.00/31.07      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
@@ -1632,285 +1634,8 @@ 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, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1984085
-    No: 15  GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 738, in __call__
-        yield remote, remote.load_module(os.path.split(build_result.filename)[1])
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
-        costs = time_f(*args).results
-      File "/workspace/python/tvm/runtime/module.py", line 357, in evaluator
-        blob = feval(*args)
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 262, in tvm._ffi._cy3.core.FuncCall
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 251, in tvm._ffi._cy3.core.FuncCall3
-      File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
-    tvm._ffi.base.TVMError: Traceback (most recent call last):
-      4: TVMFuncCall
-            at ../src/runtime/c_runtime_api.cc:477
-      3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-            at ../include/tvm/runtime/packed_func.h:1217
-      2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-            at ../src/runtime/rpc/rpc_module.cc:129
-      1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
-            at ../src/runtime/rpc/rpc_endpoint.cc:1012
-      0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
-            at ../src/runtime/rpc/rpc_endpoint.cc:804
-      File "../src/runtime/rpc/rpc_endpoint.cc", line 804
-    TVMError: 
-    ---------------------------------------------------------------
-    An error occurred during the execution of TVM.
-    For more information, please see: https://tvm.apache.org/docs/errors.html
-    ---------------------------------------------------------------
-      Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
-
-    During handling of the above exception, another exception occurred:
-
-    Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
-        costs = time_f(*args).results
-      File "/usr/lib/python3.7/contextlib.py", line 130, in __exit__
-        self.gen.throw(type, value, traceback)
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 742, in __call__
-        remote.remove(build_result.filename)
-      File "/workspace/python/tvm/rpc/client.py", line 143, in remove
-        self._remote_funcs["remove"] = self.get_function("tvm.rpc.server.remove")
-      File "/workspace/python/tvm/rpc/client.py", line 71, in get_function
-        return self._sess.get_function(name)
-      File "/workspace/python/tvm/runtime/module.py", line 171, in get_function
-        self.handle, c_str(name), ctypes.c_int(query_imports), ctypes.byref(ret_handle)
-      File "/workspace/python/tvm/_ffi/base.py", line 348, in check_call
-        raise get_last_ffi_error()
-    tvm._ffi.base.TVMError: Traceback (most recent call last):
-      52: 0xffffffffffffffff
-      51: _start
-      50: __libc_start_main
-      49: _Py_UnixMain
-      48: 0x0000000000650da0
-      47: 0x0000000000650afa
-      46: _PyFunction_FastCallDict
-      45: _PyEval_EvalCodeWithName
-      44: _PyEval_EvalFrameDefault
-      43: _PyFunction_FastCallKeywords
-      42: _PyEval_EvalCodeWithName
-      41: _PyEval_EvalFrameDefault
-      40: _PyMethodDef_RawFastCallKeywords
-      39: 0x0000000000546369
-      38: _PyEval_EvalCodeWithName
-      37: _PyEval_EvalFrameDefault
-      36: _PyFunction_FastCallKeywords
-      35: _PyEval_EvalCodeWithName
-      34: _PyEval_EvalFrameDefault
-      33: _PyFunction_FastCallDict
-      32: _PyEval_EvalCodeWithName
-      31: _PyEval_EvalFrameDefault
-      30: _PyObject_FastCallDict
-      29: 0x00000000004c06e1
-      28: _PyFunction_FastCallDict
-      27: _PyEval_EvalFrameDefault
-      26: _PyMethodDescr_FastCallKeywords
-      25: 0x00000000005dcb58
-      24: 0x00000000005dc83f
-      23: 0x00000000004ba127
-      22: _PyEval_EvalFrameDefault
-      21: _PyFunction_FastCallKeywords
-      20: _PyEval_EvalFrameDefault
-      19: _PyFunction_FastCallKeywords
-      18: _PyEval_EvalFrameDefault
-      17: _PyFunction_FastCallKeywords
-      16: _PyEval_EvalCodeWithName
-      15: _PyEval_EvalFrameDefault
-      14: 0x0000000000537c30
-      13: _PyObject_FastCallKeywords
-      12: 0x00007f68ae36dfa2
-      11: _ctypes_callproc
-      10: ffi_call
-      9: ffi_call_unix64
-      8: TVMModGetFunction
-            at ../src/runtime/c_runtime_api.cc:408
-      7: tvm::runtime::ModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)
-            at ../src/runtime/module.cc:66
-      6: tvm::runtime::RPCModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, tvm::runtime::ObjectPtr<tvm::runtime::Object> const&)
-            at ../src/runtime/rpc/rpc_module.cc:185
-      5: tvm::runtime::RPCClientSession::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
-            at ../src/runtime/rpc/rpc_endpoint.cc:1007
-      4: tvm::runtime::TVMRetValue tvm::runtime::RPCEndpoint::SysCallRemote<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(tvm::runtime::RPCCode, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
-            at ../src/runtime/rpc/rpc_endpoint.h:223
-      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(int&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const
-            at ../include/tvm/runtime/packed_func.h: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/rpc/rpc_endpoint.cc:684
-      File "../src/runtime/rpc/rpc_endpoint.cc", line 684
-    TVMError: 
-    ---------------------------------------------------------------
-    An error occurred during the execution of TVM.
-    For more information, please see: https://tvm.apache.org/docs/errors.html
-    ---------------------------------------------------------------
-      Check failed: (code == RPCCode::kReturn) is false: code=1
-
-    Traceback (most recent call last):
-      52: 0xffffffffffffffff
-      51: _start
-      50: __libc_start_main
-      49: _Py_UnixMain
-      48: 0x0000000000650da0
-      47: 0x0000000000650afa
-      46: _PyFunction_FastCallDict
-      45: _PyEval_EvalCodeWithName
-      44: _PyEval_EvalFrameDefault
-      43: _PyFunction_FastCallKeywords
-      42: _PyEval_EvalCodeWithName
-      41: _PyEval_EvalFrameDefault
-      40: _PyMethodDef_RawFastCallKeywords
-      39: 0x0000000000546369
-      38: _PyEval_EvalCodeWithName
-      37: _PyEval_EvalFrameDefault
-      36: _PyFunction_FastCallKeywords
-      35: _PyEval_EvalCodeWithName
-      34: _PyEval_EvalFrameDefault
-      33: _PyFunction_FastCallDict
-      32: _PyEval_EvalCodeWithName
-      31: _PyEval_EvalFrameDefault
-      30: _PyObject_FastCallDict
-      29: 0x00000000004c06e1
-      28: _PyFunction_FastCallDict
-      27: _PyEval_EvalFrameDefault
-      26: _PyMethodDescr_FastCallKeywords
-      25: 0x00000000005dcb58
-      24: 0x00000000005dc83f
-      23: 0x00000000004ba127
-      22: _PyEval_EvalFrameDefault
-      21: _PyFunction_FastCallKeywords
-      20: _PyEval_EvalFrameDefault
-      19: _PyFunction_FastCall      [('tile_f', [-1, 2, 1, 256]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9684837
-    No: 16  GFLOPS: 0.00/176.27     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:379
-      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:365
-      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
-            at ../src/driver/driver_api.cc:260
-      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:379
-      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:365
-      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
-            at ../src/driver/driver_api.cc:260
-      10: tvm::transform::Pass::operator()(tvm::IRModule) const
-            at ../src/ir/transform.cc:258
-      9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
-            at ../src/ir/transform.cc:274
-      8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
-            at ../src/ir/transform.cc:453
-      7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
-            at ../src/ir/transform.cc:274
-      6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
-            at ../src/tir/ir/transform.cc:100
-      5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
-            at ../include/tvm/runtime/packed_func.h:1750
-      4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
-            at ../include/tvm/runtime/packed_func.h:1694
-      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
-            at ../include/tvm/runtime/packed_func.h:1618
-      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-            at ../include/tvm/runtime/packed_func.h:1217
-      1: Call
-            at ../include/tvm/runtime/packed_func.h:1213
-      0: operator()
-            at ../src/runtime/c_runtime_api.cc:534
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
-        raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 32, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,823503
-    No: 17  GFLOPS: 37.37/176.27    result: MeasureResult(costs=(0.006194496117647059,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2431249618530273, timestamp=1665511097.5451643)       [('tile_f', [-1, 16, 4, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1747976
-    No: 18  GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7141096
+    No: 17  GFLOPS: 0.00/31.07      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
@@ -2032,9 +1757,8 @@ 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, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5344025
-    No: 19  GFLOPS: 1.66/176.27     result: MeasureResult(costs=(0.13921144100000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.047910451889038, timestamp=1665511099.6874096) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7916700
-    No: 20  GFLOPS: 0.00/176.27     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 16, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3095874
+    No: 18  GFLOPS: 0.00/31.07      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
@@ -2156,7 +1880,9 @@ 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, 16, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4804888
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3731548
+    No: 19  GFLOPS: 6.14/31.07      result: MeasureResult(costs=(0.0377166175,), error_no=MeasureErrorNo.NO_ERROR, all_cost=7.879955291748047, timestamp=1665518932.3538373)        [('tile_f', [-1, 16, 8, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9137605
+    No: 20  GFLOPS: 45.77/45.77     result: MeasureResult(costs=(0.0050580673,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3132915496826172, timestamp=1665518933.0437977)       [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9682399
 
 
 
@@ -2211,9 +1937,9 @@ and measure running time.
     Finish loading 20 records
 
     Best config:
-    [('tile_f', [-1, 4, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2715966
+    [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9682399
     Finish loading 20 records
-    Time cost of this operator: 0.001632
+    Time cost of this operator: 0.005429
 
 
 
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 311579e490..f35fc6113c 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  312.0     98.735   (1, 2, 10, 10, 3)  2       1        [312.0]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.019     0.956    (1, 6, 10, 10)     1       1        [3.019]           
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.979     0.31     (1, 1, 10, 10, 3)  1       1        [0.979]           
-    Total_time                                    -                                             315.999   -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  309.4     98.726   (1, 2, 10, 10, 3)  2       1        [309.4]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.024     0.965    (1, 6, 10, 10)     1       1        [3.024]           
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.97      0.31     (1, 1, 10, 10, 3)  1       1        [0.97]            
+    Total_time                                    -                                             313.394   -        -                  -       -        -                 
 
 
 
@@ -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  104.7     97.524   (1, 6, 10, 10, 1)  2       1        [104.7]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.793     1.67     (1, 6, 10, 10)     1       1        [1.793]           
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.865     0.806    (1, 3, 10, 10, 1)  1       1        [0.865]           
-    Total_time                                    -                                             107.358   -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  100.1     97.311   (1, 6, 10, 10, 1)  2       1        [100.1]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.8       1.75     (1, 6, 10, 10)     1       1        [1.8]             
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.966     0.939    (1, 1, 10, 10, 3)  1       1        [0.966]           
+    Total_time                                    -                                             102.866   -        -                  -       -        -                 
 
 
 
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 fe70b6db68..8df2c7f3e9 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/tmprro720uo/images/random'
+    '/tmp/tmpje3p4113/images/random'
 
 
 
@@ -316,7 +316,7 @@ objects to other stuff? We can display some examples from our datasets using ``m
 
 
 .. image-sg:: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
-   :alt: [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]
+   :alt: [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]
    :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/tmprro720uo/images/target contains 8144 images
-    /tmp/tmprro720uo/images/random contains 5000 images
+    /tmp/tmpje3p4113/images/target contains 8144 images
+    /tmp/tmpje3p4113/images/random contains 5000 images
 
 
 
@@ -501,13 +501,13 @@ the time on our validation set).
  .. code-block:: none
 
     Epoch 1/3
-    328/328 - 48s - loss: 0.2202 - accuracy: 0.9252 - val_loss: 0.2005 - val_accuracy: 0.9324 - 48s/epoch - 146ms/step
+    328/328 - 46s - loss: 0.2940 - accuracy: 0.9100 - val_loss: 0.1837 - val_accuracy: 0.9358 - 46s/epoch - 141ms/step
     Epoch 2/3
-    328/328 - 44s - loss: 0.0912 - accuracy: 0.9662 - val_loss: 0.1137 - val_accuracy: 0.9634 - 44s/epoch - 135ms/step
+    328/328 - 43s - loss: 0.1072 - accuracy: 0.9608 - val_loss: 0.1506 - val_accuracy: 0.9520 - 43s/epoch - 131ms/step
     Epoch 3/3
-    328/328 - 44s - loss: 0.0617 - accuracy: 0.9772 - val_loss: 0.1131 - val_accuracy: 0.9626 - 44s/epoch - 134ms/step
+    328/328 - 43s - loss: 0.0752 - accuracy: 0.9726 - val_loss: 0.1220 - val_accuracy: 0.9588 - 43s/epoch - 131ms/step
 
-    <keras.callbacks.History object at 0x7f81b50b1750>
+    <keras.callbacks.History object at 0x7f5ee170cdd0>
 
 
 
@@ -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  19.744 seconds)
+   **Total running time of the script:** ( 4 minutes  22.306 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 ae7fa2cb9a..68cc48a9de 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
 =================
-**05:23.703** total execution time for **how_to_work_with_microtvm** files:
+**05:24.119** total execution time for **how_to_work_with_microtvm** files:
 
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)               | 04:19.744 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)               | 04:22.306 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)         | 00:50.685 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)         | 00:49.337 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)                   | 00:09.420 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)                   | 00:08.742 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)             | 00:03.853 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)             | 00:03.731 | 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 40ce4b546d..577e46b71b 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
 =================
-**00:45.060** total execution time for **how_to_work_with_relay** files:
+**00:43.348** total execution time for **how_to_work_with_relay** files:
 
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:33.321 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:31.670 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:10.154 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:10.009 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:01.578 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:01.662 | 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 1eb821c3b9..1dda329e22 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 0x7f813e49e170>
+    <function my_cuda_math_rule at 0x7f5f43dc3170>
 
 
 
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 d392302701..766ab16912 100644
--- a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
@@ -5,22 +5,22 @@
 
 Computation times
 =================
-**00:06.060** total execution time for **how_to_work_with_schedules** files:
+**00:06.773** total execution time for **how_to_work_with_schedules** files:
 
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:03.688 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:04.475 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)                     | 00:01.015 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)                     | 00:00.992 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)                     | 00:00.586 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)                     | 00:00.556 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)                               | 00:00.555 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)                               | 00:00.541 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.120 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.116 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.050 | 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_tedd.py` (``tedd.py``)                               | 00:00.030 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)                               | 00:00.029 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)               | 00:00.016 | 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 88432d9143..9143ce9b9d 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/tmp_deuajs2/input0.cc'\nsource_filename = \"/tmp/tmp_deuajs2/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/tmphb3sswf5/input0.cc'\nsource_filename = \"/tmp/tmphb3sswf5/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 474a1556a2..b4a5548086 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
 =================
-**00:27.293** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:26.110** total execution time for **topic_vta_tutorials_autotvm** files:
 
 +---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:27.287 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:26.103 | 0.0 MB |
 +---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)     | 00:00.007 | 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 a6bd93b414..452cbbdbdc 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 29.62s!
+    resnet18_v1 inference graph built in 27.82s!
 
 
 
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 7d11ce4d1a..6eab8b130f 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 20.20s!
+    yolov3-tiny inference graph built in 19.30s!
 
 
 
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 24453a23d8..a5aa7580f4 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
 =================
-**01:42.016** total execution time for **topic_vta_tutorials_frontend** files:
+**01:38.693** total execution time for **topic_vta_tutorials_frontend** files:
 
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:52.295 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:51.179 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:49.721 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:47.514 | 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 6c9d58b144..bb7d5d2686 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
 =================
-**00:03.134** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.102** total execution time for **topic_vta_tutorials_optimize** files:
 
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.679 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.666 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.455 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.436 | 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 426fd06eda..5a69442f1f 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
 =================
-**00:00.800** total execution time for **topic_vta_tutorials** files:
+**00:00.771** total execution time for **topic_vta_tutorials** files:
 
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.434 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.413 | 0.0 MB |
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.366 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.358 | 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 e6adaa0427..2aa189e8a0 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: 97.079 ms
+    Execution time of this operator: 97.688 ms
 
 
 
@@ -444,7 +444,7 @@ operations.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  18.135 seconds)
+   **Total running time of the script:** ( 1 minutes  14.367 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 1c55948605..7c345749f5 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: 12.43/12.43     result: MeasureResult(costs=(0.021603550399999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.49829840660095215, timestamp=1665509722.5218642)      [('tile_y', [-1, 2]), ('tile_x', [-1, 512])],None,91
-    No: 2   GFLOPS: 2.60/12.43      result: MeasureResult(costs=(0.10322847220000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8344902992248535, timestamp=1665509724.3878837)        [('tile_y', [-1, 512]), ('tile_x', [-1, 8])],None,39
-    No: 3   GFLOPS: 9.74/12.43      result: MeasureResult(costs=(0.0275660862,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5878810882568359, timestamp=1665509725.773855)        [('tile_y', [-1, 512]), ('tile_x', [-1, 32])],None,59
-    No: 4   GFLOPS: 2.06/12.43      result: MeasureResult(costs=(0.13016559460000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2120654582977295, timestamp=1665509728.802026) [('tile_y', [-1, 2]), ('tile_x', [-1, 4])],None,21
-    No: 5   GFLOPS: 0.89/12.43      result: MeasureResult(costs=(0.302196536,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.986441373825073, timestamp=1665509733.9198444) [('tile_y', [-1, 256]), ('tile_x', [-1, 2])],None,18
-    No: 6   GFLOPS: 0.50/12.43      result: MeasureResult(costs=(0.5346095704,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.720899820327759, timestamp=1665509743.4198418)        [('tile_y', [-1, 64]), ('tile_x', [-1, 1])],None,6
-    No: 7   GFLOPS: 10.98/12.43     result: MeasureResult(costs=(0.0244553162,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5333662033081055, timestamp=1665509743.9848523)       [('tile_y', [-1, 1]), ('tile_x', [-1, 512])],None,90
-    No: 8   GFLOPS: 12.34/12.43     result: MeasureResult(costs=(0.0217585114,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5377156734466553, timestamp=1665509744.550449)        [('tile_y', [-1, 32]), ('tile_x', [-1, 512])],None,95
-    No: 9   GFLOPS: 3.88/12.43      result: MeasureResult(costs=(0.06914457859999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2567574977874756, timestamp=1665509745.9207058)        [('tile_y', [-1, 64]), ('tile_x', [-1, 16])],None,46
-    No: 10  GFLOPS: 11.65/12.43     result: MeasureResult(costs=(0.023043408199999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5077106952667236, timestamp=1665509746.4665062)       [('tile_y', [-1, 32]), ('tile_x', [-1, 256])],None,85
+    No: 1   GFLOPS: 0.51/0.51       result: MeasureResult(costs=(0.5274466568,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.58440637588501, timestamp=1665517583.8960118) [('tile_y', [-1, 32]), ('tile_x', [-1, 1])],None,5
+    No: 2   GFLOPS: 3.27/3.27       result: MeasureResult(costs=(0.0821094644,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4680869579315186, timestamp=1665517586.1076708)       [('tile_y', [-1, 32]), ('tile_x', [-1, 8])],None,35
+    No: 3   GFLOPS: 0.87/3.27       result: MeasureResult(costs=(0.30731819540000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.0509421825408936, timestamp=1665517591.1899326)        [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
+    No: 4   GFLOPS: 3.08/3.27       result: MeasureResult(costs=(0.08710418019999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.533874273300171, timestamp=1665517593.4760852) [('tile_y', [-1, 128]), ('tile_x', [-1, 8])],None,37
+    No: 5   GFLOPS: 9.54/9.54       result: MeasureResult(costs=(0.028123793400000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6069321632385254, timestamp=1665517594.2000203)       [('tile_y', [-1, 4]), ('tile_x', [-1, 32])],None,52
+    No: 6   GFLOPS: 0.51/9.54       result: MeasureResult(costs=(0.5268439768000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.571053266525269, timestamp=1665517602.7924662)  [('tile_y', [-1, 128]), ('tile_x', [-1, 1])],None,7
+    No: 7   GFLOPS: 13.15/13.15     result: MeasureResult(costs=(0.0204171246,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.47562265396118164, timestamp=1665517604.0231414)      [('tile_y', [-1, 4]), ('tile_x', [-1, 512])],None,92
+    No: 8   GFLOPS: 10.47/13.15     result: MeasureResult(costs=(0.0256389898,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5563304424285889, timestamp=1665517604.6051786)       [('tile_y', [-1, 512]), ('tile_x', [-1, 256])],None,89
+    No: 9   GFLOPS: 0.90/13.15      result: MeasureResult(costs=(0.2995569776,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.911799192428589, timestamp=1665517609.6503007)        [('tile_y', [-1, 64]), ('tile_x', [-1, 2])],None,16
+    No: 10  GFLOPS: 9.81/13.15      result: MeasureResult(costs=(0.027360711600000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6169931888580322, timestamp=1665517610.2948134)       [('tile_y', [-1, 8]), ('tile_x', [-1, 32])],None,53
 
 
 
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 74f86faa07..42fb04fa4a 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': 521.2336288400002, 'median': 521.1876098499943, 'std': 1.7164045226842666}
+    {'mean': 515.9857380699988, 'median': 515.575389899999, 'std': 2.4389230765799526}
 
 
 
@@ -554,30 +554,30 @@ the tuning data to.
 
  .. code-block:: none
 
-
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  1/25]  Current/Best:    6.93/  16.64 GFLOPS | Progress: (4/20) | 7.48 s
    [Task  1/25]  Current/Best:   11.40/  16.64 GFLOPS | Progress: (8/20) | 11.95 s
    [Task  1/25]  Current/Best:   17.57/  17.57 GFLOPS | Progress: (12/20) | 14.10 s
    [Task  1/25]  Current/Best:   11.02/  17.57 GFLOPS | Progress: (16/20) | 17.65 s
    [Task  1/25]  Current/Best:    8.81/  17.57 GFLOPS | Progress: (20/20) | 20.93 s Done.
-
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:   14.58/  18.83 GFLOPS | Progress: (4/20) | 2.77 s
    [Task  2/25]  Current/Best:   17.04/  18.83 GFLOPS | Progress: (8/20) | 3.88 s
    [Task  2/25]  Current/Best:   17.75/  18.83 GFLOPS | Progress: (12/20) | 5.42 s
    [Task  2/25]  Current/Best:    5.62/  18.83 GFLOPS | Progress: (16/20) | 6.87 s
    [Task  2/25]  Current/Best:   15.17/  18.83 GFLOPS | Progress: (20/20) | 8.18 s Done.
-
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:    8.91/  10.83 GFLOPS | Progress: (4/20) | 4.20 s
    [Task  3/25]  Current/Best:    9.98/  12.00 GFLOPS | Progress: (8/20) | 6.22 s
    [Task  3/25]  Current/Best:   12.15/  16.88 GFLOPS | Progress: (12/20) | 7.87 s
    [Task  3/25]  Current/Best:   12.69/  16.88 GFLOPS | Progress: (16/20) | 10.22 s
    [Task  3/25]  Current/Best:   12.65/  17.94 GFLOPS | Progress: (20/20) | 13.63 s Done.
-
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:   11.10/  16.92 GFLOPS | Progress: (4/20) | 3.59 s
    [Task  4/25]  Current/Best:   14.01/  16.92 GFLOPS | Progress: (8/20) | 6.49 s
    [Task  4/25]  Current/Best:    5.87/  21.28 GFLOPS | Progress: (12/20) | 9.56 s
    [Task  4/25]  Current/Best:   16.94/  21.28 GFLOPS | Progress: (16/20) | 15.99 s
    [Task  4/25]  Current/Best:   18.72/  21.28 GFLOPS | Progress: (20/20) | 17.95 s Done.
-
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  5/25]  Current/Best:    9.44/  15.08 GFLOPS | Progress: (4/20) | 3.72 s
    [Task  5/25]  Current/Best:   12.84/  15.08 GFLOPS | Progress: (8/20) | 5.40 s
    [Task  5/25]  Current/Best:   10.39/  18.42 GFLOPS | Progress: (12/20) | 6.94 s
    [Task  5/25]  Current/Best:   16.94/  18.42 GFLOPS | Progress: (16/20) | 8.74 s
    [Task  5/25]  Current/Best:   12.81/  18.42 GFLOPS | Progress: (20/20) | 10.21 s Done.
-
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:   12.94/  13.17 GFLOPS | Progress: (4/20) | 3.87 s
    [Task  6/25]  Current/Best:   13.78/  18.42 GFLOPS | Progress: (8/20) | 6.27 s
    [Task  6/25]  Current/Best:    1.58/  18.42 GFLOPS | Progress: (12/20) | 9.86 s
    [Task  6/25]  Current/Best:   18.08/  18.42 GFLOPS | Progress: (16/20) | 11.71 s
    [Task  6/25]  Current/Best:   10.15/  18.42 GFLOPS | Progress: (20/20) | 14.11 s Done.
-
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:    9.62/   9.62 GFLOPS | Progress: (4/20) | 4.34 s
    [Task  7/25]  Current/Best:   16.24/  19.00 GFLOPS | Progress: (8/20) | 6.18 s
    [Task  7/25]  Current/Best:   12.26/  19.00 GFLOPS | Progress: (12/20) | 8.33 s
    [Task  7/25]  Current/Best:   18.18/  19.00 GFLOPS | Progress: (16/20) | 10.24 s
    [Task  7/25]  Current/Best:   12.08/  20.39 GFLOPS | Progress: (20/20) | 12.54 s Done.
-
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:    7.36/  11.75 GFLOPS | Progress: (4/20) | 8.80 s
    [Task  8/25]  Current/Best:   16.06/  20.69 GFLOPS | Progress: (8/20) | 10.97 s
    [Task  8/25]  Current/Best:   21.74/  21.74 GFLOPS | Progress: (12/20) | 16.44 s
    [Task  8/25]  Current/Best:   12.24/  21.74 GFLOPS | Progress: (16/20) | 18.52 s
    [Task  8/25]  Current/Best:    8.60/  21.74 GFLOPS | Progress: (20/20) | 21.80 s Done.
-
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:    6.29/  18.79 GFLOPS | Progress: (4/20) | 3.40 s
    [Task  9/25]  Current/Best:   18.21/  18.79 GFLOPS | Progress: (8/20) | 5.02 s
    [Task  9/25]  Current/Best:   19.43/  19.43 GFLOPS | Progress: (12/20) | 7.34 s
    [Task  9/25]  Current/Best:   14.64/  19.43 GFLOPS | Progress: (16/20) | 11.61 s
    [Task  9/25]  Current/Best:    8.28/  20.51 GFLOPS | Progress: (20/20) | 21.88 s Done.
-
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:   14.28/  14.28 GFLOPS | Progress: (4/20) | 4.49 s
    [Task 10/25]  Current/Best:   20.66/  20.66 GFLOPS | Progress: (8/20) | 5.89 s
    [Task 10/25]  Current/Best:   21.13/  21.13 GFLOPS | Progress: (12/20) | 8.35 s
    [Task 10/25]  Current/Best:   14.48/  21.13 GFLOPS | Progress: (16/20) | 10.87 s
    [Task 10/25]  Current/Best:   14.12/  21.13 GFLOPS | Progress: (20/20) | 12.23 s Done.
-
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   17.24/  21.89 GFLOPS | Progress: (4/20) | 3.88 s
    [Task 11/25]  Current/Best:    6.65/  24.10 GFLOPS | Progress: (8/20) | 6.03 s
    [Task 11/25]  Current/Best:   17.82/  24.10 GFLOPS | Progress: (12/20) | 8.58 s
    [Task 11/25]  Current/Best:   16.01/  24.10 GFLOPS | Progress: (16/20) | 10.32 s
    [Task 11/25]  Current/Best:   20.29/  24.10 GFLOPS | Progress: (20/20) | 12.35 s Done.
-
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:    3.34/  16.01 GFLOPS | Progress: (4/20) | 5.73 s
    [Task 12/25]  Current/Best:   12.92/  16.01 GFLOPS | Progress: (8/20) | 9.05 s
    [Task 12/25]  Current/Best:    7.09/  16.01 GFLOPS | Progress: (12/20) | 11.70 s
    [Task 12/25]  Current/Best:    5.18/  19.83 GFLOPS | Progress: (16/20) | 17.46 s
    [Task 12/25]  Current/Best:    5.70/  19.83 GFLOPS | Progress: (20/20) | 19.37 s Done.
-
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:    3.08/  22.01 GFLOPS | Progress: (4/20) | 4.28 s
    [Task 13/25]  Current/Best:   15.43/  22.01 GFLOPS | Progress: (8/20) | 7.39 s
    [Task 13/25]  Current/Best:   14.36/  22.01 GFLOPS | Progress: (12/20) | 9.77 s
    [Task 13/25]  Current/Best:   11.47/  22.01 GFLOPS | Progress: (16/20) | 12.91 s
    [Task 13/25]  Current/Best:    7.46/  22.01 GFLOPS | Progress: (20/20) | 16.58 s Done.
-
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:    9.77/  15.56 GFLOPS | Progress: (4/20) | 4.39 s
    [Task 14/25]  Current/Best:   10.09/  19.73 GFLOPS | Progress: (8/20) | 6.21 s
    [Task 14/25]  Current/Best:   10.61/  19.73 GFLOPS | Progress: (12/20) | 9.23 s
    [Task 14/25]  Current/Best:   12.73/  19.73 GFLOPS | Progress: (16/20) | 11.98 s
    [Task 14/25]  Current/Best:   10.27/  19.73 GFLOPS | Progress: (20/20) | 15.52 s Done.
-
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 15/25]  Current/Best:    6.88/  13.34 GFLOPS | Progress: (4/20) | 6.61 s
    [Task 15/25]  Current/Best:   14.80/  17.59 GFLOPS | Progress: (8/20) | 7.81 s
    [Task 15/25]  Current/Best:   11.15/  17.59 GFLOPS | Progress: (12/20) | 9.91 s
    [Task 15/25]  Current/Best:   14.30/  17.59 GFLOPS | Progress: (16/20) | 12.28 s
    [Task 15/25]  Current/Best:   14.77/  20.15 GFLOPS | Progress: (20/20) | 15.17 s
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:   18.50/  18.50 GFLOPS | Progress: (4/20) | 3.53 s
    [Task 16/25]  Current/Best:   13.50/  18.50 GFLOPS | Progress: (8/20) | 5.37 s
    [Task 16/25]  Current/Best:   16.02/  18.50 GFLOPS | Progress: (12/20) | 6.79 s
    [Task 16/25]  Current/Best:   16.35/  18.97 GFLOPS | Progress: (16/20) | 8.35 s
    [Task 16/25]  Current/Best:   11.54/  18.97 GFLOPS | Progress: (20/20) |
  11.36 s Done.
-
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:   20.44/  20.44 GFLOPS | Progress: (4/20) | 3.55 s
    [Task 17/25]  Current/Best:   13.50/  20.44 GFLOPS | Progress: (8/20) | 6.32 s
    [Task 17/25]  Current/Best:    9.97/  20.44 GFLOPS | Progress: (12/20) | 8.57 s
    [Task 17/25]  Current/Best:    4.71/  20.44 GFLOPS | Progress: (16/20) | 10.59 s
    [Task 17/25]  Current/Best:   12.38/  21.39 GFLOPS | Progress: (20/20) | 12.96 s Done.
-
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:   16.89/  16.89 GFLOPS | Progress: (4/20) | 3.83 s
    [Task 18/25]  Current/Best:   12.28/  18.92 GFLOPS | Progress: (8/20) | 7.41 s
    [Task 18/25]  Current/Best:   10.93/  18.92 GFLOPS | Progress: (12/20) | 10.36 s
    [Task 18/25]  Current/Best:   11.78/  19.37 GFLOPS | Progress: (16/20) | 13.90 s
    [Task 18/25]  Current/Best:   16.15/  19.37 GFLOPS | Progress: (20/20) | 16.59 s Done.
-
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:    5.34/  18.48 GFLOPS | Progress: (4/20) | 5.13 s
    [Task 19/25]  Current/Best:   13.85/  18.48 GFLOPS | Progress: (8/20) | 8.50 s
    [Task 19/25]  Current/Best:    3.08/  18.48 GFLOPS | Progress: (12/20) | 12.69 s
    [Task 19/25]  Current/Best:   12.26/  20.64 GFLOPS | Progress: (16/20) | 14.84 s
    [Task 19/25]  Current/Best:   12.07/  20.64 GFLOPS | Progress: (20/20) | 17.21 s Done.
-
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:   17.30/  17.30 GFLOPS | Progress: (4/20) | 3.32 s
    [Task 20/25]  Current/Best:   16.28/  17.30 GFLOPS | Progress: (8/20) | 5.61 s Done.
-
    [Task 20/25]  Current/Best:    2.32/  17.30 GFLOPS | Progress: (12/20) | 9.79 s
    [Task 20/25]  Current/Best:   13.29/  17.30 GFLOPS | Progress: (16/20) | 12.35 s
    [Task 20/25]  Current/Best:   15.75/  17.30 GFLOPS | Progress: (20/20) | 14.98 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 21/25]  Current/Best:   12.42/  14.04 GFLOPS | Progress: (4/20) | 3.02 s
    [Task 21/25]  Current/Best:   18.32/  18.87 GFLOPS | Progress: (8/20) | 5.46 s
    [Task 21/25]  Current/Best:   21.03/  21.03 GFLOPS | Progress: (12/20) | 7.51 s
    [Task 21/25]  Current/Best:    1.60/  21.03 GFLOPS | Progress: (16/20) | 10.10 s
    [Task 21/25]  Current/Best:    8.73/  21.03 GFLOPS | Progress: (20/20) | 11.52 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:   11.37/  18.78 GFLOPS | Progress: (4/20) | 5.13 s
    [Task 22/25]  Current/Best:   15.41/  18.78 GFLOPS | Progress: (8/20)
  | 6.55 s
    [Task 22/25]  Current/Best:    4.58/  21.52 GFLOPS | Progress: (12/20) | 8.22 s
    [Task 22/25]  Current/Best:   14.80/  21.52 GFLOPS | Progress: (16/20) | 10.28 s Done.
-
    [Task 22/25]  Current/Best:    9.45/  21.53 GFLOPS | Progress: (20/20) | 11.70 s Done.
-
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:   10.41/  20.83 GFLOPS | Progress: (4/20) | 3.78 s
    [Task 23/25]  Current/Best:   11.94/  20.83 GFLOPS | Progress: (8/20) | 6.12 s
    [Task 23/25]  Current/Best:   17.26/  22.74 GFLOPS | Progress: (12/20) | 8.09 s
    [Task 23/25]  Current/Best:   21.35/  22.74 GFLOPS | Progress: (16/20) | 10.89 s
    [Task 23/25]  Current/Best:   18.15/  22.74 GFLOPS | Progress: (20/20) | 13.11 s Done.
-
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    8.00/   8.00 GFLOPS | Progress: (4/20) | 12.05 s
    [Task 24/25]  Current/Best:    3.25/   8.00 GFLOPS | Progress: (8/20) | 23.51 s
    [Task 24/25]  Current/Best:    7.70/   8.00 GFLOPS | Progress: (12/20) | 35.48 s
    [Task 24/25]  Current/Best:    1.90/   9.19 GFLOPS | Progress: (16/20) | 47.38 s
    [Task 24/25]  Current/Best:    2.48/   9.19 GFLOPS | Progress: (20/20) | 57.89 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
    [Task 25/25]  Current/Best:    5.68/   7.52 GFLOPS | Progress: (4/20) | 7.49 s
    [Task 25/25]  Current/Best:    1.54/   8.97 GFLOPS | Progress: (8/20) | 18.24 s
    [Task 25/25]  Current/Best:    8.01/   8.97 GFLOPS | Progress: (12/20) | 29.97 s
    [Task 25/25]  Current/Best:    5.70/   9.20 GFLOPS | Progress: (16/20) | 39.68 s
    [Task 25/25]  Current/Best:    2.95/   9.20 GFLOPS | Progress: (20/20) | 50.22 s
+
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  1/25]  Current/Best:    9.64/  12.58 GFLOPS | Progress: (4/20) | 8.33 s
    [Task  1/25]  Current/Best:    3.39/  17.98 GFLOPS | Progress: (8/20) | 11.84 s
    [Task  1/25]  Current/Best:   12.62/  19.17 GFLOPS | Progress: (12/20) | 13.65 s
    [Task  1/25]  Current/Best:   15.43/  19.17 GFLOPS | Progress: (16/20) | 15.79 s
    [Task  1/25]  Current/Best:   16.70/  19.17 GFLOPS | Progress: (20/20) | 17.84 s Done.
+
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:   17.52/  17.52 GFLOPS | Progress: (4/20) | 3.44 s
    [Task  2/25]  Current/Best:   16.30/  17.52 GFLOPS | Progress: (8/20) | 4.90 s
    [Task  2/25]  Current/Best:   17.98/  18.50 GFLOPS | Progress: (12/20) | 5.88 s
    [Task  2/25]  Current/Best:   17.69/  18.50 GFLOPS | Progress: (16/20) | 7.00 s
    [Task  2/25]  Current/Best:   15.83/  18.50 GFLOPS | Progress: (20/20) | 8.17 s Done.
+
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:   22.26/  22.26 GFLOPS | Progress: (4/20) | 3.95 s
    [Task  3/25]  Current/Best:   11.12/  22.26 GFLOPS | Progress: (8/20) | 5.93 s
    [Task  3/25]  Current/Best:    6.36/  22.26 GFLOPS | Progress: (12/20) | 8.23 s
    [Task  3/25]  Current/Best:   14.59/  22.26 GFLOPS | Progress: (16/20) | 9.98 s
    [Task  3/25]  Current/Best:   10.03/  22.26 GFLOPS | Progress: (20/20) | 11.81 s Done.
+
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:   17.20/  17.20 GFLOPS | Progress: (4/20) | 3.18 s
    [Task  4/25]  Current/Best:    9.94/  17.20 GFLOPS | Progress: (8/20) | 6.37 s
    [Task  4/25]  Current/Best:   19.42/  19.42 GFLOPS | Progress: (12/20) | 12.60 s
    [Task  4/25]  Current/Best:    7.52/  19.42 GFLOPS | Progress: (16/20) | 14.44 s
    [Task  4/25]  Current/Best:    5.06/  19.42 GFLOPS | Progress: (20/20) | 16.01 s Done.
+
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  5/25]  Current/Best:    4.86/  16.18 GFLOPS | Progress: (4/20) | 3.68 s
    [Task  5/25]  Current/Best:    7.50/  16.18 GFLOPS | Progress: (8/20) | 5.53 s
    [Task  5/25]  Current/Best:    8.11/  20.02 GFLOPS | Progress: (12/20) | 7.76 s
    [Task  5/25]  Current/Best:   11.52/  21.16 GFLOPS | Progress: (16/20) | 10.71 s
    [Task  5/25]  Current/Best:   10.93/  21.16 GFLOPS | Progress: (20/20) | 12.32 s Done.
+
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:   12.26/  17.35 GFLOPS | Progress: (4/20) | 4.37 s
    [Task  6/25]  Current/Best:   17.01/  19.19 GFLOPS | Progress: (8/20) | 7.48 s
    [Task  6/25]  Current/Best:    8.50/  19.19 GFLOPS | Progress: (12/20) | 9.90 s
    [Task  6/25]  Current/Best:   12.02/  19.19 GFLOPS | Progress: (16/20) | 12.11 s
    [Task  6/25]  Current/Best:   11.09/  19.19 GFLOPS | Progress: (20/20) | 15.72 s Done.
+
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:   12.52/  12.82 GFLOPS | Progress: (4/20) | 5.10 s
    [Task  7/25]  Current/Best:   15.78/  22.25 GFLOPS | Progress: (8/20) | 6.64 s
    [Task  7/25]  Current/Best:   10.03/  22.25 GFLOPS | Progress: (12/20) | 8.98 s
    [Task  7/25]  Current/Best:   15.84/  22.25 GFLOPS | Progress: (16/20) | 11.21 s
    [Task  7/25]  Current/Best:   17.87/  22.25 GFLOPS | Progress: (20/20) | 13.10 s Done.
+
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:   10.32/  11.78 GFLOPS | Progress: (4/20) | 8.72 s
    [Task  8/25]  Current/Best:    7.71/  14.12 GFLOPS | Progress: (8/20) | 12.76 s
    [Task  8/25]  Current/Best:   10.05/  17.85 GFLOPS | Progress: (12/20) | 15.14 s
    [Task  8/25]  Current/Best:   14.07/  17.85 GFLOPS | Progress: (16/20) | 26.44 s
    [Task  8/25]  Current/Best:    7.20/  17.85 GFLOPS | Progress: (20/20) | 29.50 s
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:   10.04/  22.68 GFLOPS | Progress: (4/20) | 4.78 s
    [Task  9/25]  Current/Best:   23.19/  23.19 GFLOPS | Progress: (8/20) | 6.17 s
    [Task  9/25]  Current/Best:    7.39/  23.19 GFLOPS | Progress: (12/20) | 8.07 s
    [Task  9/25]  Current/Best:   10.02/  23.19 GFLOPS | Progress: (16/20) | 11.53 s
    [Task  9/25]  Current/Best:   10.90/  23.19 GFLOPS | Progress: (20/20
 ) | 16.94 s Done.
+
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:    3.91/  17.79 GFLOPS | Progress: (4/20) | 3.33 s
    [Task 10/25]  Current/Best:   13.07/  17.79 GFLOPS | Progress: (8/20) | 5.12 s
    [Task 10/25]  Current/Best:    9.79/  17.79 GFLOPS | Progress: (12/20) | 8.00 s
    [Task 10/25]  Current/Best:   14.83/  19.36 GFLOPS | Progress: (16/20) | 9.51 s
    [Task 10/25]  Current/Best:   11.34/  19.36 GFLOPS | Progress: (20/20) | 11.25 s Done.
+
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   16.43/  23.70 GFLOPS | Progress: (4/20) | 3.43 s
    [Task 11/25]  Current/Best:   15.65/  23.70 GFLOPS | Progress: (8/20) | 5.49 s
    [Task 11/25]  Current/Best:   18.70/  23.70 GFLOPS | Progress: (12/20) | 7.59 s
    [Task 11/25]  Current/Best:   21.15/  23.70 GFLOPS | Progress: (16/20) | 9.97 s
    [Task 11/25]  Current/Best:   13.25/  23.70 GFLOPS | Progress: (20/20) | 13.60 s Done.
+
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:   13.49/  18.73 GFLOPS | Progress: (4/20) | 4.42 s
    [Task 12/25]  Current/Best:   11.70/  21.59 GFLOPS | Progress: (8/20) | 12.48 s
    [Task 12/25]  Current/Best:    4.83/  21.59 GFLOPS | Progress: (12/20) | 16.05 s
    [Task 12/25]  Current/Best:   20.30/  21.59 GFLOPS | Progress: (16/20) | 18.15 s
    [Task 12/25]  Current/Best:   12.54/  21.59 GFLOPS | Progress: (20/20) | 20.37 s Done.
+
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:    3.08/  13.79 GFLOPS | Progress: (4/20) | 4.62 s
    [Task 13/25]  Current/Best:   12.51/  13.79 GFLOPS | Progress: (8/20) | 8.04 s
    [Task 13/25]  Current/Best:   19.73/  21.56 GFLOPS | Progress: (12/20) | 11.52 s
    [Task 13/25]  Current/Best:   22.43/  22.43 GFLOPS | Progress: (16/20) | 13.84 s
    [Task 13/25]  Current/Best:   16.85/  22.43 GFLOPS | Progress: (20/20) | 15.98 s Done.
+
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:   15.01/  19.62 GFLOPS | Progress: (4/20) | 4.39 s
    [Task 14/25]  Current/Best:    3.87/  19.62 GFLOPS | Progress: (8/20) | 9.19 s
    [Task 14/25]  Current/Best:   17.57/  19.62 GFLOPS | Progress: (12/20) | 11.19 s
    [Task 14/25]  Current/Best:    9.26/  19.62 GFLOPS | Progress: (16/20) | 16.43 s
    [Task 14/25]  Current/Best:    8.89/  19.62 GFLOPS | Progress: (20/20) | 18.41 s
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+     Done.
+
    [Task 15/25]  Current/Best:   12.43/  16.04 GFLOPS | Progress: (4/20) | 4.08 s
    [Task 15/25]  Current/Best:   16.64/  23.36 GFLOPS | Progress: (8/20) | 5.61 s
    [Task 15/25]  Current/Best:   16.33/  23.36 GFLOPS | Progress: (12/20) | 10.54 s
    [Task 15/25]  Current/Best:    7.02/  23.36 GFLOPS | Progress: (16/20) | 13.33 s
    [Task 15/25]  Current/Best:   10.60/  23.36 GFLOPS | Progress: (20/20) | 17.51 s
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:   10.35/  18.03 GFLOPS | Progress: (4/20) | 5.46 s
    [Task 16/25]  Current/Best:   19.51/  19.51 GFLOPS | Progress: (8/20) | 6.67 s
    [Task 16/25]  Current/Best:   17.73/  19.51 GFLOPS | Progress: (12/20) | 8.40 s
    [Task 16/25]  Current/Best:    4.38/  19.51 GFLOPS | Progress: (16/20) | 10.14 s
    [Task 16/25]  Current/Best:   11.29/  19.51 GFLOPS | Progress: (20/20) | 13.70 s Done.
+
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:   16.50/  16.50 GFLOPS | Progress: (4/20) | 3.99 s
    [Task 17/25]  Current/Best:   11.52/  18.13 GFLOPS | Progress: (8/20) | 7.08 s
    [Task 17/25]  Current/Best:   21.52/  21.52 GFLOPS | Progress: (12/20) | 9.35 s
    [Task 17/25]  Current/Best:   13.23/  21.52 GFLOPS | Progress: (16/20) | 11.34 s
    [Task 17/25]  Current/Best:   19.41/  21.52 GFLOPS | Progress: (20/20) | 14.73 s Done.
+
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:    8.40/  18.89 GFLOPS | Progress: (4/20) | 3.15 s
    [Task 18/25]  Current/Best:   13.33/  18.89 GFLOPS | Progress: (8/20) | 5.52 s
    [Task 18/25]  Current/Best:    9.91/  18.89 GFLOPS | Progress: (12/20) | 10.63 s
    [Task 18/25]  Current/Best:   16.55/  18.89 GFLOPS | Progress: (16/20) | 13.64 s
    [Task 18/25]  Current/Best:    9.94/  18.89 GFLOPS | Progress: (20/20) | 18.54 s Done.
+
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:    3.08/  14.24 GFLOPS | Progress: (4/20) | 4.66 s
    [Task 19/25]  Current/Best:   11.19/  19.30 GFLOPS | Progress: (8/20) | 8.09 s
    [Task 19/25]  Current/Best:    7.80/  19.30 GFLOPS | Progress: (12/20) | 11.36 s
    [Task 19/25]  Current/Best:   17.47/  19.30 GFLOPS | Progress: (16/20) | 15.19 s
    [Task 19/25]  Current/Best:   20.52/  20.52 GFLOPS | Progress: (20/20) | 19.27 s Done.
+
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:    7.51/  16.83 GFLOPS | Progress: (4/20) | 5.01 s
    [Task 20/25]  Current/Best:    1.56/  16.83 GFLOPS | Progress: (8/20) | 7.92 s
    [Task 20/25]  Current/Best:   10.78/  16.83 GFLOPS | Progress: (12/20) | 11.28 s
    [Task 20/25]  Current/Best:   10.20/  16.83 GFLOPS | Progress: (16/20) | 13.15 s
    [Task 20/25]  Current/Best:    8.07/  19.83 GFLOPS | Progress: (20/20) | 14.94 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 21/25]  Current/Best:   13.26/  13.26 GFLOPS | Progress: (4/20) | 2.67 s
    [Task 21/25]  Current/Best:   14.47/  14.47 GFLOPS | Progress: (8/20) | 6.34 s
    [Task 21/25]  Current/Best:   10.51/  19.83 GFLOPS | Progress: (12/20) | 7.80 s
    [Task 21/25]  Current/Best:    6.35/  19.83 GFLOPS | Progress: (16/20) | 10.63 s Done.
+     Done.
+
    [Task 21/25]  Current/Best:   18.48/  19.83 GFLOPS | Progress: (20/20) | 13.32 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:   16.29/  16.29 GFLOPS | Progress: (4/20) | 4.10 s
    [Task 22/25]  Current/Best:   11.10/  18.38 GFLOPS | Progress: (8/20) | 5.60 s
    [Task 22/25]  Current/Best:   14.66/  18.38 GFLOPS | Progress: (12/20) | 7.38 s
    [Task 22/25]  Current/Best:    9.37/  18.38 GFLOPS | Progress: (16/20) | 8.89 s
    [Task 22/25]  Current/Best:    5.23/  18.38 GFLOPS | Progress: (20/20) | 13.80 s Done.
+
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:   19.58/  19.58 GFLOPS | Progress: (4/20) | 3.35 s
    [Task 23/25]  Current/Best:   10.64/  19.58 GFLOPS | Progress: (8/20) | 6.91 s
    [Task 23/25]  Current/Best:   20.20/  20.20 GFLOPS | Progress: (12/20) | 10.00 s
    [Task 23/25]  Current/Best:    4.65/  20.20 GFLOPS | Progress: (16/20) | 12.24 s
    [Task 23/25]  Current/Best:   11.56/  20.71 GFLOPS | Progress: (20/20) | 15.19 s Done.
+
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    6.87/   6.87 GFLOPS | Progress: (4/20) | 11.78 s
    [Task 24/25]  Current/Best:    6.97/  10.09 GFLOPS | Progress: (8/20) | 22.50 s
    [Task 24/25]  Current/Best:    8.51/  10.28 GFLOPS | Progress: (12/20) | 24.26 s
    [Task 24/25]  Current/Best:    7.70/  10.28 GFLOPS | Progress: (16/20) | 34.92 s
    [Task 24/25]  Current/Best:    5.40/  10.28 GFLOPS | Progress: (20/20) | 45.38 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
    [Task 25/25]  Current/Best:    7.89/   8.47 GFLOPS | Progress: (4/20) | 6.75 s
    [Task 25/25]  Current/Best:    1.54/   9.17 GFLOPS | Progress: (8/20) | 8.13 s
    [Task 25/25]  Current/Best:    7.52/   9.17 GFLOPS | Progress: (12/20) | 18.84 s
    [Task 25/25]  Current/Best:    5.39/   9.17 GFLOPS | Progress: (16/20) | 29.55 s
    [Task 25/25]  Current/Best:    7.77/   9.17 GFLOPS | Progress: (20/20) | 34.40 s
 
 
 
@@ -673,8 +673,8 @@ 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=0.621104
+    class='n02123159 tiger cat' with probability=0.356378
     class='n02124075 Egyptian cat' with probability=0.019712
     class='n02129604 tiger, Panthera tigris' with probability=0.001215
     class='n04040759 radiator' with probability=0.000262
@@ -731,8 +731,8 @@ improvement in comparing the optimized model to the unoptimized model.
 
  .. code-block:: none
 
-    optimized: {'mean': 425.9876747699991, 'median': 425.46388890000344, 'std': 2.9148418458696193}
-    unoptimized: {'mean': 521.2336288400002, 'median': 521.1876098499943, 'std': 1.7164045226842666}
+    optimized: {'mean': 423.6774043899959, 'median': 422.1668977499917, 'std': 4.600969726904317}
+    unoptimized: {'mean': 515.9857380699988, 'median': 515.575389899999, 'std': 2.4389230765799526}
 
 
 
@@ -755,7 +755,7 @@ profiling/benchmarking.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 11 minutes  12.467 seconds)
+   **Total running time of the script:** ( 10 minutes  49.881 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 a2b2825eb0..47962226a0 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
 
-    1.32e-07 secs/op
+    1.272e-07 secs/op
 
 
 
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 4e4f163389..69a2054d39 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, 0x4530350)), stage(b, placeholder(b, 0x5bf1f30)), 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, 0x57cb480)), stage(b, placeholder(b, 0x70fd140)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min= [...]
 
 
 
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index 4d95670ca7..50fdf10b88 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
 =================
-**14:39.372** total execution time for **tutorial** files:
+**14:20.697** total execution time for **tutorial** files:
 
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 11:12.467 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 10:49.881 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:18.135 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:14.367 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 01:01.990 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 01:00.125 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:36.333 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:38.783 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:28.445 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:35.454 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:01.051 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:01.173 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.763 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.733 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.179 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.174 | 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``)                                             | 00:00.002 | 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_uma.py` (``uma.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 |
++------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.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 cc401f1e4e..c4c86ecc30 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -501,10 +501,10 @@ We can now compare the different schedules
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                   numpy    7.176100000378938e-06                    1.0
-                   naive              6.7905e-06      0.9462660776245347
-                parallel              6.9306e-06      0.9657892169331568
-                  vector    2.4547700000000002e-05    3.4207577930496713
+                   numpy    7.37176000029649e-06                     1.0
+                   naive              6.6637e-06      0.9039496673429396
+                parallel              6.8707e-06      0.9320298001730472
+                  vector             2.46505e-05      3.3439097310558896
 
 
 
@@ -925,7 +925,7 @@ matrix multiplication.
 
  .. code-block:: none
 
-    Numpy running time: 0.018895
+    Numpy running time: 0.018499
 
 
 
@@ -983,7 +983,7 @@ optimizations.
 
  .. code-block:: none
 
-    none: 3.403558
+    none: 3.334699
 
 
 
@@ -1086,7 +1086,7 @@ schedule.
 
  .. code-block:: none
 
-    blocking: 0.336028
+    blocking: 0.306825
 
 
 
@@ -1182,7 +1182,7 @@ already cache friendly from our previous optimizations.
 
  .. code-block:: none
 
-    vectorization: 0.362116
+    vectorization: 0.341911
     @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: 0.135282
+    loop permutation: 0.117094
     @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: 0.109471
+    array packing: 0.107859
     @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: 0.112258
+    block caching: 0.110001
     @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: 0.149092
+    parallelization: 0.146074
     @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            3.4035581496                     1.0
-                blocking     0.33602819669999995     0.09872850174147645
-           vectorization            0.3621163498     0.10639346644997308
-        loop permutation            0.1352822269    0.039747294141543876
-           array packing     0.10947078369999999     0.03216362961592575
-           block caching            0.1122579553     0.03298252897873745
-         parallelization            0.1490921385     0.04380478662235341
+                    none            3.3346987206                     1.0
+                blocking            0.3068253467     0.09200991525998908
+           vectorization            0.3419112393     0.10253137328054666
+        loop permutation     0.11709402359999999    0.035113823889593145
+           array packing     0.10785941219999999     0.03234457479882718
+           block caching     0.11000114000000001     0.03298683006068024
+         parallelization            0.1460740626     0.04380427583986281
 
 
 
@@ -1663,7 +1663,7 @@ the computation for specific platforms.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  1.990 seconds)
+   **Total running time of the script:** ( 1 minutes  0.125 seconds)
 
 
 .. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/_static/documentation_options.js b/docs/_static/documentation_options.js
index 9ce5359a8c..7271870dfa 100644
--- a/docs/_static/documentation_options.js
+++ b/docs/_static/documentation_options.js
@@ -1,6 +1,6 @@
 var DOCUMENTATION_OPTIONS = {
     URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'),
-    VERSION: '0.10.dev0',
+    VERSION: '0.11.dev0',
     LANGUAGE: 'None',
     COLLAPSE_INDEX: false,
     BUILDER: 'html',
diff --git a/docs/arch/benchmark.html b/docs/arch/benchmark.html
index 9e5f8b6d98..ff38c672e5 100644
--- a/docs/arch/benchmark.html
+++ b/docs/arch/benchmark.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Benchmark Performance Log Format &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Benchmark Performance Log Format &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/convert_layout.html b/docs/arch/convert_layout.html
index 4281822ff6..86b4000e38 100644
--- a/docs/arch/convert_layout.html
+++ b/docs/arch/convert_layout.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Convert Layout Pass &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Convert Layout Pass &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/debugger.html b/docs/arch/debugger.html
index 46e7597e7a..6f113429e5 100644
--- a/docs/arch/debugger.html
+++ b/docs/arch/debugger.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Debugger &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Debugger &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/device_target_interactions.html b/docs/arch/device_target_interactions.html
index 829216ef21..ceb7b7bef4 100644
--- a/docs/arch/device_target_interactions.html
+++ b/docs/arch/device_target_interactions.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Device/Target Interactions &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Device/Target Interactions &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/frontend/tensorflow.html b/docs/arch/frontend/tensorflow.html
index 895b8af677..1fa0a8aac2 100644
--- a/docs/arch/frontend/tensorflow.html
+++ b/docs/arch/frontend/tensorflow.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TensorFlow Frontend &mdash; tvm 0.10.dev0 documentation</title>
+  <title>TensorFlow Frontend &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/hybrid_script.html b/docs/arch/hybrid_script.html
index d4b9b0739c..f7e8e85657 100644
--- a/docs/arch/hybrid_script.html
+++ b/docs/arch/hybrid_script.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Hybrid Frontend Developer Guide &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Hybrid Frontend Developer Guide &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/index.html b/docs/arch/index.html
index 71b1dff143..b6a1f47a7a 100644
--- a/docs/arch/index.html
+++ b/docs/arch/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Design and Architecture &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Design and Architecture &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/inferbound.html b/docs/arch/inferbound.html
index 14319f59e2..e48dcb01a6 100644
--- a/docs/arch/inferbound.html
+++ b/docs/arch/inferbound.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>InferBound Pass &mdash; tvm 0.10.dev0 documentation</title>
+  <title>InferBound Pass &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/introduction_to_module_serialization.html b/docs/arch/introduction_to_module_serialization.html
index 2e2e345537..03ae9cc703 100644
--- a/docs/arch/introduction_to_module_serialization.html
+++ b/docs/arch/introduction_to_module_serialization.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Introduction to Module Serialization &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Introduction to Module Serialization &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/microtvm_design.html b/docs/arch/microtvm_design.html
index cee674cf5d..0104543fb1 100644
--- a/docs/arch/microtvm_design.html
+++ b/docs/arch/microtvm_design.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>microTVM Design Document &mdash; tvm 0.10.dev0 documentation</title>
+  <title>microTVM Design Document &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/microtvm_project_api.html b/docs/arch/microtvm_project_api.html
index ab7af6f417..feda4a1f1f 100644
--- a/docs/arch/microtvm_project_api.html
+++ b/docs/arch/microtvm_project_api.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>microTVM Project API &mdash; tvm 0.10.dev0 documentation</title>
+  <title>microTVM Project API &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/model_library_format.html b/docs/arch/model_library_format.html
index d3cc0e2882..a4b9a26431 100644
--- a/docs/arch/model_library_format.html
+++ b/docs/arch/model_library_format.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Model Library Format &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Model Library Format &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/pass_infra.html b/docs/arch/pass_infra.html
index cf0d59b7cb..1524400058 100644
--- a/docs/arch/pass_infra.html
+++ b/docs/arch/pass_infra.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Pass Infrastructure &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Pass Infrastructure &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/relay_intro.html b/docs/arch/relay_intro.html
index bc4701c417..f84f8281cf 100644
--- a/docs/arch/relay_intro.html
+++ b/docs/arch/relay_intro.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Introduction to Relay IR &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Introduction to Relay IR &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/relay_op_strategy.html b/docs/arch/relay_op_strategy.html
index fc45856e77..65674f35ee 100644
--- a/docs/arch/relay_op_strategy.html
+++ b/docs/arch/relay_op_strategy.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Relay Operator Strategy &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Relay Operator Strategy &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/runtime.html b/docs/arch/runtime.html
index 21299442b5..bb1b60395c 100644
--- a/docs/arch/runtime.html
+++ b/docs/arch/runtime.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TVM Runtime System &mdash; tvm 0.10.dev0 documentation</title>
+  <title>TVM Runtime System &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/runtimes/vulkan.html b/docs/arch/runtimes/vulkan.html
index aabb3ffc3f..29144e7c3a 100644
--- a/docs/arch/runtimes/vulkan.html
+++ b/docs/arch/runtimes/vulkan.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Vulkan Runtime &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Vulkan Runtime &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/security.html b/docs/arch/security.html
index cc1c14ebc0..76bc92db5d 100644
--- a/docs/arch/security.html
+++ b/docs/arch/security.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Security Guide &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Security Guide &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/arch/virtual_machine.html b/docs/arch/virtual_machine.html
index fb7503037f..ff2578b64a 100644
--- a/docs/arch/virtual_machine.html
+++ b/docs/arch/virtual_machine.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Putting the VM in TVM: The Relay Virtual Machine &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Putting the VM in TVM: The Relay Virtual Machine &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/commit_hash b/docs/commit_hash
index 52c86e30b6..a22c1bb786 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-afeab6e8a320b3a4aedbbe0b25886501146c1942
+cd077575f365cc2e7fb20b263d7e533b202518e4
diff --git a/docs/contribute/ci.html b/docs/contribute/ci.html
index b8522fa970..b380718c57 100644
--- a/docs/contribute/ci.html
+++ b/docs/contribute/ci.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Using TVM’s CI &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Using TVM’s CI &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/code_guide.html b/docs/contribute/code_guide.html
index 1441a2538a..ed93405104 100644
--- a/docs/contribute/code_guide.html
+++ b/docs/contribute/code_guide.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Code Guide and Tips &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Code Guide and Tips &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/code_review.html b/docs/contribute/code_review.html
index 1d9ed0f02c..f16ad247f4 100644
--- a/docs/contribute/code_review.html
+++ b/docs/contribute/code_review.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Code Reviews &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Code Reviews &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/committer_guide.html b/docs/contribute/committer_guide.html
index e919120f33..2681f22c8d 100644
--- a/docs/contribute/committer_guide.html
+++ b/docs/contribute/committer_guide.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Committer Guide &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Committer Guide &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/community.html b/docs/contribute/community.html
index d6c61e4649..5ec1b663e4 100644
--- a/docs/contribute/community.html
+++ b/docs/contribute/community.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TVM Community Guidelines &mdash; tvm 0.10.dev0 documentation</title>
+  <title>TVM Community Guidelines &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/document.html b/docs/contribute/document.html
index 71d5e11d25..926fb047bf 100644
--- a/docs/contribute/document.html
+++ b/docs/contribute/document.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Documentation &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Documentation &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/error_handling.html b/docs/contribute/error_handling.html
index f57c78190c..1df03f9fda 100644
--- a/docs/contribute/error_handling.html
+++ b/docs/contribute/error_handling.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Error Handling Guide &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Error Handling Guide &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/git_howto.html b/docs/contribute/git_howto.html
index 958c84e5f1..ead9332344 100644
--- a/docs/contribute/git_howto.html
+++ b/docs/contribute/git_howto.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Git Usage Tips &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Git Usage Tips &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/index.html b/docs/contribute/index.html
index a05890bcd2..6c6ea102d5 100644
--- a/docs/contribute/index.html
+++ b/docs/contribute/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Contributor Guide &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Contributor Guide &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/pull_request.html b/docs/contribute/pull_request.html
index fbf002a507..48ef6d463d 100644
--- a/docs/contribute/pull_request.html
+++ b/docs/contribute/pull_request.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Submit a Pull Request &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Submit a Pull Request &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/contribute/release_process.html b/docs/contribute/release_process.html
index 6ac46261f2..b1acfa8520 100644
--- a/docs/contribute/release_process.html
+++ b/docs/contribute/release_process.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Release Process &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Release Process &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/how_to/debugging_tvm.html b/docs/dev/how_to/debugging_tvm.html
index aeec20aea7..75d57cda04 100644
--- a/docs/dev/how_to/debugging_tvm.html
+++ b/docs/dev/how_to/debugging_tvm.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Debugging TVM &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Debugging TVM &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/how_to/how_to.html b/docs/dev/how_to/how_to.html
index 10a6b0b223..57b8525c24 100644
--- a/docs/dev/how_to/how_to.html
+++ b/docs/dev/how_to/how_to.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Developer How-To Guide &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Developer How-To Guide &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/how_to/pytest_target_parametrization.html b/docs/dev/how_to/pytest_target_parametrization.html
index 8d57f2c760..b39bf2344f 100644
--- a/docs/dev/how_to/pytest_target_parametrization.html
+++ b/docs/dev/how_to/pytest_target_parametrization.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Python Target Parametrization &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Python Target Parametrization &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/how_to/relay_add_op.html b/docs/dev/how_to/relay_add_op.html
index 07ff011eb7..a9d0d1ac8f 100644
--- a/docs/dev/how_to/relay_add_op.html
+++ b/docs/dev/how_to/relay_add_op.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Adding an Operator to Relay &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Adding an Operator to Relay &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -175,7 +175,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -185,7 +185,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -197,6 +197,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/how_to/relay_add_pass.html b/docs/dev/how_to/relay_add_pass.html
index 7be2f4b964..ce9f6d781e 100644
--- a/docs/dev/how_to/relay_add_pass.html
+++ b/docs/dev/how_to/relay_add_pass.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Adding a Compiler Pass to Relay &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Adding a Compiler Pass to Relay &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/how_to/relay_bring_your_own_codegen.html b/docs/dev/how_to/relay_bring_your_own_codegen.html
index 149b04c0b6..93200f3a0d 100644
--- a/docs/dev/how_to/relay_bring_your_own_codegen.html
+++ b/docs/dev/how_to/relay_bring_your_own_codegen.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Bring Your Own Codegen To TVM &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Bring Your Own Codegen To TVM &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/tutorial/codebase_walkthrough.html b/docs/dev/tutorial/codebase_walkthrough.html
index a45a06ecb3..7391ab0234 100644
--- a/docs/dev/tutorial/codebase_walkthrough.html
+++ b/docs/dev/tutorial/codebase_walkthrough.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TVM Codebase Walkthrough by Example &mdash; tvm 0.10.dev0 documentation</title>
+  <title>TVM Codebase Walkthrough by Example &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/dev/tutorial/index.html b/docs/dev/tutorial/index.html
index 352ecb4c59..ee33d40ff6 100644
--- a/docs/dev/tutorial/index.html
+++ b/docs/dev/tutorial/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Developer Tutorial &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Developer Tutorial &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/errors.html b/docs/errors.html
index 6b045ac715..57ccc2f2ad 100644
--- a/docs/errors.html
+++ b/docs/errors.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Handle TVM Errors &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Handle TVM Errors &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/faq.html b/docs/faq.html
index 19956ccccd..c2b122b031 100644
--- a/docs/faq.html
+++ b/docs/faq.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Frequently Asked Questions &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Frequently Asked Questions &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/genindex.html b/docs/genindex.html
index 9352807078..87bdc91cfe 100644
--- a/docs/genindex.html
+++ b/docs/genindex.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Index &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Index &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -172,7 +172,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -182,7 +182,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -194,6 +194,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/compile_models/from_coreml.html b/docs/how_to/compile_models/from_coreml.html
index df0aa850ef..0ffb782fcf 100644
--- a/docs/how_to/compile_models/from_coreml.html
+++ b/docs/how_to/compile_models/from_coreml.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile CoreML Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile CoreML Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index e0b4873d73..eb91bdd265 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile YOLO-V2 and YOLO-V3 in DarkNet Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile YOLO-V2 and YOLO-V3 in DarkNet Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -572,7 +577,7 @@ class:[&#39;truck 0.9266&#39;] left:471 top:83 right:689 bottom:169
 class:[&#39;bicycle 0.9984&#39;] 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  11.050 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  8.358 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 7cba98c7da..c39e091c69 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile Keras Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile Keras Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -493,7 +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 1s/step
+1/1 [==============================] - 1s 963ms/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 ef231e2566..458dc401c7 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile MXNet Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile MXNet Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -427,7 +432,7 @@ to download the full example code</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;x&quot;</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.zipa1c06891-93d7-4365-abc9-377bcd6779e3 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.zip360dc28d-d1b8-47e5-843b-e602cdf08c95 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 da042ff735..be1d95b206 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile OneFlow Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile OneFlow Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -435,9 +440,10 @@ 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: &quot;https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip&quot; to /workspace/.oneflow/flowvision_cache/resnet18.zip
 
   0%|          | 0.00/41.5M [00:00&lt;?, ?B/s]
- 39%|###8      | 16.0M/41.5M [00:00&lt;00:00, 136MB/s]
- 70%|######9   | 29.0M/41.5M [00:00&lt;00:00, 132MB/s]
-100%|##########| 41.5M/41.5M [00:00&lt;00:00, 117MB/s]
+ 31%|###1      | 13.0M/41.5M [00:00&lt;00:00, 136MB/s]
+ 63%|######2   | 26.0M/41.5M [00:00&lt;00:00, 100MB/s]
+ 87%|########6 | 36.1M/41.5M [00:00&lt;00:00, 97.2MB/s]
+100%|##########| 41.5M/41.5M [00:00&lt;00:00, 92.6MB/s]
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/compile_models/from_onnx.html b/docs/how_to/compile_models/from_onnx.html
index 2933e9524b..c56eb5d711 100644
--- a/docs/how_to/compile_models/from_onnx.html
+++ b/docs/how_to/compile_models/from_onnx.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile ONNX Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile ONNX Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/compile_models/from_paddle.html b/docs/how_to/compile_models/from_paddle.html
index a5cafe31f6..56652a4900 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile PaddlePaddle Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile PaddlePaddle Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index 4391758547..3fff939b01 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile PyTorch Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile PyTorch Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -414,9 +419,10 @@ be unstable.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/resnet18-f37072fd.pth&quot; to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
 
   0%|          | 0.00/44.7M [00:00&lt;?, ?B/s]
- 38%|###8      | 17.0M/44.7M [00:00&lt;00:00, 177MB/s]
- 76%|#######5  | 33.9M/44.7M [00:00&lt;00:00, 122MB/s]
-100%|##########| 44.7M/44.7M [00:00&lt;00:00, 120MB/s]
+ 28%|##8       | 12.7M/44.7M [00:00&lt;00:00, 133MB/s]
+ 57%|#####6    | 25.4M/44.7M [00:00&lt;00:00, 117MB/s]
+ 82%|########2 | 36.7M/44.7M [00:00&lt;00:00, 100MB/s]
+100%|##########| 44.7M/44.7M [00:00&lt;00:00, 110MB/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 80ff01a9e1..dc8538987a 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile Tensorflow Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile Tensorflow Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -632,7 +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  13.246 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  9.408 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/from_tflite.html b/docs/how_to/compile_models/from_tflite.html
index af977752b3..1d2a23b5db 100644
--- a/docs/how_to/compile_models/from_tflite.html
+++ b/docs/how_to/compile_models/from_tflite.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile TFLite Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile TFLite Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/compile_models/index.html b/docs/how_to/compile_models/index.html
index d0ad9d7e33..52f901f845 100644
--- a/docs/how_to/compile_models/index.html
+++ b/docs/how_to/compile_models/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile Deep Learning Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile Deep Learning Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/compile_models/sg_execution_times.html b/docs/how_to/compile_models/sg_execution_times.html
index 4b2abd258b..785656a12d 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Computation times &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Computation times &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -172,7 +172,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -182,7 +182,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -194,6 +194,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -327,7 +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>05:51.552</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:37.660</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 81%" />
@@ -336,43 +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>01:13.246</p></td>
+<td><p>01:09.408</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>01:11.050</p></td>
+<td><p>01:08.358</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>00:47.116</p></td>
+<td><p>00:45.374</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>00:32.464</p></td>
+<td><p>00:31.466</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>00:30.984</p></td>
+<td><p>00:30.137</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>00:27.484</p></td>
+<td><p>00:26.074</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>00:24.871</p></td>
+<td><p>00:25.004</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>00:22.612</p></td>
+<td><p>00:21.674</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>00:19.266</p></td>
+<td><p>00:17.808</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>00:02.460</p></td>
+<td><p>00:02.358</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/deploy/android.html b/docs/how_to/deploy/android.html
index 7f9c26af2e..752c696cc1 100644
--- a/docs/how_to/deploy/android.html
+++ b/docs/how_to/deploy/android.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy to Android &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy to Android &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/arm_compute_lib.html b/docs/how_to/deploy/arm_compute_lib.html
index 8aa72aba71..a4e3a2cb2c 100644
--- a/docs/how_to/deploy/arm_compute_lib.html
+++ b/docs/how_to/deploy/arm_compute_lib.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Relay Arm® Compute Library Integration &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Relay Arm® Compute Library Integration &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/bnns.html b/docs/how_to/deploy/bnns.html
index 58f131b25e..4b013b16b8 100644
--- a/docs/how_to/deploy/bnns.html
+++ b/docs/how_to/deploy/bnns.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Relay BNNS Integration &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Relay BNNS Integration &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/cpp_deploy.html b/docs/how_to/deploy/cpp_deploy.html
index 6978c00f65..76c422a99f 100644
--- a/docs/how_to/deploy/cpp_deploy.html
+++ b/docs/how_to/deploy/cpp_deploy.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy TVM Module using C++ API &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy TVM Module using C++ API &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/hls.html b/docs/how_to/deploy/hls.html
index 0ac8cadf78..48e2344445 100644
--- a/docs/how_to/deploy/hls.html
+++ b/docs/how_to/deploy/hls.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>HLS Backend Example &mdash; tvm 0.10.dev0 documentation</title>
+  <title>HLS Backend Example &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/index.html b/docs/how_to/deploy/index.html
index 05617e918b..3b6864e79e 100644
--- a/docs/how_to/deploy/index.html
+++ b/docs/how_to/deploy/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy Models and Integrate TVM &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy Models and Integrate TVM &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/integrate.html b/docs/how_to/deploy/integrate.html
index 47fe508cb1..02e2c3237b 100644
--- a/docs/how_to/deploy/integrate.html
+++ b/docs/how_to/deploy/integrate.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Integrate TVM into Your Project &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Integrate TVM into Your Project &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/tensorrt.html b/docs/how_to/deploy/tensorrt.html
index e86b41d358..03661d51cd 100644
--- a/docs/how_to/deploy/tensorrt.html
+++ b/docs/how_to/deploy/tensorrt.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Relay TensorRT Integration &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Relay TensorRT Integration &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy/vitis_ai.html b/docs/how_to/deploy/vitis_ai.html
index d53546cea8..e50028ae18 100644
--- a/docs/how_to/deploy/vitis_ai.html
+++ b/docs/how_to/deploy/vitis_ai.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Vitis AI Integration &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Vitis AI Integration &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 3466317e4b..125e77ac46 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy the Pretrained Model on Android &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy the Pretrained Model on Android &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -649,7 +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)
-  16.3367      16.3350      16.4466      16.2776       0.0429
+  16.4607      16.6100      16.9983      15.9131       0.4308
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_nano.html b/docs/how_to/deploy_models/deploy_model_on_nano.html
index 3f1e5146d5..66097d98b7 100644
--- a/docs/how_to/deploy_models/deploy_model_on_nano.html
+++ b/docs/how_to/deploy_models/deploy_model_on_nano.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy the Pretrained Model on Jetson Nano &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy the Pretrained Model on Jetson Nano &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy_models/deploy_model_on_rasp.html b/docs/how_to/deploy_models/deploy_model_on_rasp.html
index 20723ffadd..36a9a2b140 100644
--- a/docs/how_to/deploy_models/deploy_model_on_rasp.html
+++ b/docs/how_to/deploy_models/deploy_model_on_rasp.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy the Pretrained Model on Raspberry Pi &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy the Pretrained Model on Raspberry Pi &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </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 8a47922d2f..703b95fdba 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compile PyTorch Object Detection Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Compile PyTorch Object Detection Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -436,23 +441,21 @@ be unstable.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth&quot; to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
 
   0%|          | 0.00/170M [00:00&lt;?, ?B/s]
-  6%|5         | 10.1M/170M [00:00&lt;00:01, 94.5MB/s]
- 14%|#3        | 23.0M/170M [00:00&lt;00:01, 118MB/s]
- 20%|##        | 34.4M/170M [00:00&lt;00:01, 109MB/s]
- 26%|##6       | 44.9M/170M [00:00&lt;00:01, 106MB/s]
- 32%|###2      | 55.0M/170M [00:00&lt;00:01, 104MB/s]
- 38%|###8      | 64.9M/170M [00:00&lt;00:01, 103MB/s]
- 44%|####4     | 74.7M/170M [00:00&lt;00:00, 101MB/s]
- 50%|####9     | 84.4M/170M [00:00&lt;00:00, 101MB/s]
- 55%|#####5    | 94.2M/170M [00:00&lt;00:00, 101MB/s]
- 61%|######1   | 104M/170M [00:01&lt;00:00, 99.7MB/s]
- 67%|######6   | 114M/170M [00:01&lt;00:00, 101MB/s]
- 73%|#######2  | 123M/170M [00:01&lt;00:00, 101MB/s]
- 78%|#######8  | 133M/170M [00:01&lt;00:00, 98.0MB/s]
- 84%|########4 | 143M/170M [00:01&lt;00:00, 101MB/s]
- 90%|######### | 153M/170M [00:01&lt;00:00, 101MB/s]
- 96%|#########5| 163M/170M [00:01&lt;00:00, 100MB/s]
-100%|##########| 170M/170M [00:01&lt;00:00, 102MB/s]
+  9%|9         | 15.6M/170M [00:00&lt;00:00, 164MB/s]
+ 18%|#8        | 31.2M/170M [00:00&lt;00:01, 119MB/s]
+ 25%|##5       | 43.2M/170M [00:00&lt;00:01, 111MB/s]
+ 32%|###1      | 54.1M/170M [00:00&lt;00:01, 98.8MB/s]
+ 39%|###8      | 66.1M/170M [00:00&lt;00:01, 107MB/s]
+ 45%|####5     | 76.6M/170M [00:00&lt;00:01, 83.6MB/s]
+ 55%|#####4    | 93.0M/170M [00:00&lt;00:00, 105MB/s]
+ 61%|######1   | 104M/170M [00:01&lt;00:00, 107MB/s]
+ 68%|######7   | 115M/170M [00:01&lt;00:00, 106MB/s]
+ 74%|#######4  | 126M/170M [00:01&lt;00:00, 104MB/s]
+ 80%|########  | 136M/170M [00:01&lt;00:00, 101MB/s]
+ 86%|########6 | 146M/170M [00:01&lt;00:00, 101MB/s]
+ 92%|#########2| 156M/170M [00:01&lt;00:00, 103MB/s]
+ 98%|#########7| 166M/170M [00:01&lt;00:00, 102MB/s]
+100%|##########| 170M/170M [00:01&lt;00:00, 103MB/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 &#39;trunc&#39; function NOT &#39;floor&#39;). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode=&#39;trunc&#39;), or for actual floor division, use torch.div(a, b, rounding_mode=& [...]
@@ -550,7 +553,7 @@ 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  22.709 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  13.999 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 52c8d34968..f863c47b69 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy a Framework-prequantized Model with TVM &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy a Framework-prequantized Model with TVM &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -480,8 +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: &quot;https://download.pytorch.org/models/mobilenet_v2-b0353104.pth&quot; to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
 
   0%|          | 0.00/13.6M [00:00&lt;?, ?B/s]
- 73%|#######3  | 9.95M/13.6M [00:00&lt;00:00, 104MB/s]
-100%|##########| 13.6M/13.6M [00:00&lt;00:00, 106MB/s]
+ 59%|#####8    | 7.99M/13.6M [00:00&lt;00:00, 83.7MB/s]
+100%|##########| 13.6M/13.6M [00:00&lt;00:00, 95.2MB/s]
 </pre></div>
 </div>
 </div>
@@ -572,7 +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)
-  90.6393      90.5738      92.2934      90.3124       0.3266
+  90.3089      90.2385      91.0332      89.9026       0.2178
 </pre></div>
 </div>
 <div class="admonition note">
@@ -611,7 +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  23.780 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  20.538 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 0e49815faf..a16657731a 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite) &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite) &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -569,7 +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)
-  123.3837     123.2487     129.0543     122.0951      0.6967
+  119.4024     119.3994     120.7627     118.5256      0.4341
 </pre></div>
 </div>
 <div class="admonition note">
@@ -597,7 +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> ( 1 minutes  57.380 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  57.169 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 0830b422e0..a7bbbba9a0 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy a Quantized Model on Cuda &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy a Quantized Model on Cuda &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -507,7 +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  17.982 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  18.619 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_sparse.html b/docs/how_to/deploy_models/deploy_sparse.html
index 0583f7d22a..831866af4e 100644
--- a/docs/how_to/deploy_models/deploy_sparse.html
+++ b/docs/how_to/deploy_models/deploy_sparse.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy a Hugging Face Pruned Model on CPU &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy a Hugging Face Pruned Model on CPU &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
index 89dc19b6c7..df0a9cd4b1 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy Single Shot Multibox Detector(SSD) model &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy Single Shot Multibox Detector(SSD) model &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -449,25 +454,24 @@ 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&lt;?, ?KB/s]
-  3%|3         | 4247/132723 [00:00&lt;00:03, 42463.36KB/s]
-  8%|8         | 10914/132723 [00:00&lt;00:02, 56694.36KB/s]
- 14%|#3        | 18483/132723 [00:00&lt;00:01, 65350.78KB/s]
- 20%|#9        | 26049/132723 [00:00&lt;00:01, 69418.39KB/s]
- 25%|##5       | 33535/132723 [00:00&lt;00:01, 71378.07KB/s]
- 31%|###       | 41122/132723 [00:00&lt;00:01, 72900.80KB/s]
- 37%|###6      | 48769/132723 [00:00&lt;00:01, 74066.15KB/s]
- 42%|####2     | 56311/132723 [00:00&lt;00:01, 74495.72KB/s]
- 48%|####8     | 63891/132723 [00:00&lt;00:00, 74900.54KB/s]
- 54%|#####3    | 71573/132723 [00:01&lt;00:00, 75491.70KB/s]
- 60%|#####9    | 79208/132723 [00:01&lt;00:00, 75750.19KB/s]
- 65%|######5   | 86784/132723 [00:01&lt;00:00, 75638.49KB/s]
- 71%|#######1  | 94348/132723 [00:01&lt;00:00, 75430.55KB/s]
- 77%|#######6  | 101927/132723 [00:01&lt;00:00, 75536.51KB/s]
- 82%|########2 | 109481/132723 [00:01&lt;00:00, 75330.47KB/s]
- 88%|########8 | 117046/132723 [00:01&lt;00:00, 75423.93KB/s]
- 94%|#########3| 124589/132723 [00:01&lt;00:00, 75392.61KB/s]
-100%|#########9| 132161/132723 [00:01&lt;00:00, 75489.36KB/s]
-100%|##########| 132723/132723 [00:01&lt;00:00, 73266.12KB/s]
+  3%|3         | 4246/132723 [00:00&lt;00:03, 42454.27KB/s]
+  8%|8         | 11251/132723 [00:00&lt;00:02, 58682.07KB/s]
+ 13%|#3        | 17523/132723 [00:00&lt;00:01, 60523.67KB/s]
+ 19%|#9        | 25482/132723 [00:00&lt;00:01, 68048.67KB/s]
+ 25%|##5       | 33433/132723 [00:00&lt;00:01, 72173.81KB/s]
+ 31%|###1      | 41313/132723 [00:00&lt;00:01, 74423.00KB/s]
+ 37%|###7      | 49360/132723 [00:00&lt;00:01, 76395.80KB/s]
+ 43%|####3     | 57461/132723 [00:00&lt;00:00, 77861.30KB/s]
+ 49%|####9     | 65496/132723 [00:00&lt;00:00, 78637.38KB/s]
+ 55%|#####5    | 73531/132723 [00:01&lt;00:00, 79160.48KB/s]
+ 61%|######1   | 81579/132723 [00:01&lt;00:00, 79563.14KB/s]
+ 68%|######7   | 89680/132723 [00:01&lt;00:00, 80002.17KB/s]
+ 74%|#######3  | 97681/132723 [00:01&lt;00:00, 79978.00KB/s]
+ 80%|#######9  | 105679/132723 [00:01&lt;00:00, 79769.93KB/s]
+ 86%|########5 | 113657/132723 [00:01&lt;00:00, 79676.52KB/s]
+ 92%|#########1| 121713/132723 [00:01&lt;00:00, 79937.75KB/s]
+ 98%|#########7| 129720/132723 [00:01&lt;00:00, 79974.61KB/s]
+100%|##########| 132723/132723 [00:01&lt;00:00, 76338.18KB/s]
 </pre></div>
 </div>
 <p>Create TVM runtime and do inference
@@ -506,7 +510,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
 <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  3.620 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  55.549 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/index.html b/docs/how_to/deploy_models/index.html
index 20deae7fca..586ddf6078 100644
--- a/docs/how_to/deploy_models/index.html
+++ b/docs/how_to/deploy_models/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Deploy Deep Learning Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Deploy Deep Learning Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/deploy_models/sg_execution_times.html b/docs/how_to/deploy_models/sg_execution_times.html
index c5439a0831..38c454907c 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Computation times &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Computation times &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -172,7 +172,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -182,7 +182,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -194,6 +194,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -327,7 +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>12:33.619</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>12:11.026</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 86%" />
@@ -336,35 +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>03:22.709</p></td>
+<td><p>03:13.999</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>03:03.620</p></td>
+<td><p>02:55.549</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></td>
-<td><p>01:57.380</p></td>
+<td><p>01:57.169</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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>01:23.780</p></td>
+<td><p>01:20.538</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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>01:17.982</p></td>
+<td><p>01:18.619</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>00:37.144</p></td>
+<td><p>00:35.487</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>00:25.669</p></td>
+<td><p>00:25.094</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>00:25.329</p></td>
+<td><p>00:24.564</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 c7209f6236..d46ecde339 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Bring Your Own Datatypes to TVM &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Bring Your Own Datatypes to TVM &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -608,7 +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.zipf45eae96-4cc9-400c-8240-b8a923568747 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.zip3503d0bd-f202-4194-8f6a-5d7cb2e726e8 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/index.html b/docs/how_to/extend_tvm/index.html
index a8aa337fb2..c0b07e32cd 100644
--- a/docs/how_to/extend_tvm/index.html
+++ b/docs/how_to/extend_tvm/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Extend TVM &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Extend TVM &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/extend_tvm/low_level_custom_pass.html b/docs/how_to/extend_tvm/low_level_custom_pass.html
index cbf8e8cc87..af43e44a9d 100644
--- a/docs/how_to/extend_tvm/low_level_custom_pass.html
+++ b/docs/how_to/extend_tvm/low_level_custom_pass.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Writing a Customized Pass &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Writing a Customized Pass &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index 71d74142cb..3cfcad814b 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Computation times &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Computation times &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -172,7 +172,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -182,7 +182,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -194,6 +194,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -327,7 +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>00:48.829</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:46.929</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -336,15 +341,15 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:45.210</p></td>
+<td><p>00:43.432</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>00:02.518</p></td>
+<td><p>00:02.439</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>00:01.093</p></td>
+<td><p>00:01.050</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/use_pass_infra.html b/docs/how_to/extend_tvm/use_pass_infra.html
index 69c50119d1..a5929148db 100644
--- a/docs/how_to/extend_tvm/use_pass_infra.html
+++ b/docs/how_to/extend_tvm/use_pass_infra.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to Use TVM Pass Infra &mdash; tvm 0.10.dev0 documentation</title>
+  <title>How to Use TVM Pass Infra &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index 9ebdaa8bea..32ad7bd9a5 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to Use TVM Pass Instrument &mdash; tvm 0.10.dev0 documentation</title>
+  <title>How to Use TVM Pass Instrument &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -512,10 +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: 7310us [7310us] (46.99%; 46.99%)
-FoldScaleAxis: 8245us [7us] (53.01%; 53.01%)
-        FoldConstant: 8238us [1682us] (52.96%; 99.91%)
-                InferType: 6556us [6556us] (42.15%; 79.58%)
+InferType: 6841us [6841us] (46.30%; 46.30%)
+FoldScaleAxis: 7935us [6us] (53.70%; 53.70%)
+        FoldConstant: 7929us [1650us] (53.66%; 99.93%)
+                InferType: 6280us [6280us] (42.50%; 79.20%)
 </pre></div>
 </div>
 </div>
@@ -537,10 +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: 6729us [6729us] (44.76%; 44.76%)
-FoldScaleAxis: 8304us [8us] (55.24%; 55.24%)
-        FoldConstant: 8297us [1723us] (55.19%; 99.91%)
-                InferType: 6574us [6574us] (43.73%; 79.23%)
+InferType: 6329us [6329us] (42.30%; 42.30%)
+FoldScaleAxis: 8632us [5us] (57.70%; 57.70%)
+        FoldConstant: 8628us [1690us] (57.67%; 99.95%)
+                InferType: 6938us [6938us] (46.37%; 80.41%)
 </pre></div>
 </div>
 <p>Register empty list to clear existing instruments.</p>
diff --git a/docs/how_to/index.html b/docs/how_to/index.html
index 194a2b4f1d..e8d6ef1d3c 100644
--- a/docs/how_to/index.html
+++ b/docs/how_to/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How To Guides &mdash; tvm 0.10.dev0 documentation</title>
+  <title>How To Guides &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/optimize_operators/index.html b/docs/how_to/optimize_operators/index.html
index 0cc0dcd0f6..99136ec686 100644
--- a/docs/how_to/optimize_operators/index.html
+++ b/docs/how_to/optimize_operators/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Optimize Tensor Operators &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Optimize Tensor Operators &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/optimize_operators/opt_conv_cuda.html b/docs/how_to/optimize_operators/opt_conv_cuda.html
index 82da4de7c4..996116bf8d 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to optimize convolution on GPU &mdash; tvm 0.10.dev0 documentation</title>
+  <title>How to optimize convolution on GPU &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -564,7 +569,7 @@ latency of convolution.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Convolution: </span><span class="si">%f</span><span class="s2"> ms&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 48.016288 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 34.771232 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 511e42eb96..4579d26960 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to optimize convolution using TensorCores &mdash; tvm 0.10.dev0 documentation</title>
+  <title>How to optimize convolution using TensorCores &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -906,7 +911,7 @@ be able to run on our build server</p>
     <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;conv2d with tensor core: </span><span class="si">%f</span><span class="s2"> ms&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">* [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 10.185052 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 13.361171 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 48a11fb2d3..2fba419a81 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to optimize GEMM on CPU &mdash; tvm 0.10.dev0 documentation</title>
+  <title>How to optimize GEMM on CPU &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -461,8 +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">&quot;Baseline: </span><span class="si">%f</span><span class="s2">&quot;</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: 0.020022
-Baseline: 3.553984
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018603
+Baseline: 3.452391
 </pre></div>
 </div>
 <p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -522,7 +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">&quot;Opt1: </span><span class="si">%f</span><span class="s2">&quot;</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: 0.333014
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.309239
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -589,7 +594,7 @@ vastly.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt2: </span><span class="si">%f</span><span class="s2">&quot;</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: 0.354947
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.342060
 </pre></div>
 </div>
 <p>Here is the generated IR after vectorization.</p>
@@ -650,7 +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">&quot;Opt3: </span><span class="si">%f</span><span class="s2">&quot;</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: 0.136089
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.115616
 </pre></div>
 </div>
 <p>Here is the generated IR after loop permutation.</p>
@@ -733,7 +738,7 @@ flattening.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt4: </span><span class="si">%f</span><span class="s2">&quot;</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: 0.111404
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.110118
 </pre></div>
 </div>
 <p>Here is the generated IR after array packing.</p>
@@ -819,7 +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">&quot;Opt5: </span><span class="si">%f</span><span class="s2">&quot;</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: 0.113366
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.112093
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -909,7 +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">&quot;Opt6: </span><span class="si">%f</span><span class="s2">&quot;</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: 0.148662
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147396
 </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 2eddae5e18..86c4f6314c 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Computation times &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Computation times &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -172,7 +172,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -182,7 +182,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -194,6 +194,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -327,7 +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>00:36.294</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.052</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -336,15 +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>00:33.906</p></td>
+<td><p>00:32.687</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>00:01.331</p></td>
+<td><p>00:01.354</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>00:01.058</p></td>
+<td><p>00:01.012</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/profile/index.html b/docs/how_to/profile/index.html
index bd697ef0b7..97d072d8a0 100644
--- a/docs/how_to/profile/index.html
+++ b/docs/how_to/profile/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Profile Models &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Profile Models &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/profile/papi.html b/docs/how_to/profile/papi.html
index 6729deea8a..0726c194c7 100644
--- a/docs/how_to/profile/papi.html
+++ b/docs/how_to/profile/papi.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Getting Started With PAPI &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Getting Started With PAPI &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
diff --git a/docs/how_to/tune_with_autoscheduler/index.html b/docs/how_to/tune_with_autoscheduler/index.html
index b4d62f9141..c336c4387e 100644
--- a/docs/how_to/tune_with_autoscheduler/index.html
+++ b/docs/how_to/tune_with_autoscheduler/index.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Use AutoScheduler for Template-Free Scheduling &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Use AutoScheduler for Template-Free Scheduling &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
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 d5172d7b64..7ada4d51b3 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Computation times &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Computation times &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -172,7 +172,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -182,7 +182,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -194,6 +194,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -327,7 +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>08:52.975</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>08:32.196</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 85%" />
@@ -336,27 +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>05:22.781</p></td>
+<td><p>05:08.147</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>01:34.218</p></td>
+<td><p>01:32.684</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>01:04.203</p></td>
+<td><p>01:02.782</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>00:29.074</p></td>
+<td><p>00:27.068</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>00:11.443</p></td>
+<td><p>00:10.772</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>00:11.256</p></td>
+<td><p>00:10.743</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 24a99464c8..e10410e12f 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
@@ -11,7 +11,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Auto-scheduling a Convolution Layer for GPU &mdash; tvm 0.10.dev0 documentation</title>
+  <title>Auto-scheduling a Convolution Layer for GPU &mdash; tvm 0.11.dev0 documentation</title>
   
 
   
@@ -174,7 +174,7 @@
               <input type="checkbox" class="version-toggle-box" hidden id="version-toggle">
               <label for="version-toggle" class="version-toggle-label">
                   <div tabindex="0" class="version version-selector version-selector-show">
-                    0.10.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
+                    0.11.dev0 <span class="chevron versions-hidden"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m8 4 8 8-8 8" stroke="#000" stroke-linecap="round" stroke-linejoin="round" stroke-width="2"/></svg></span><span class="chevron versions-shown"><svg fill="none" height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path d="m4 8 8 8 8-8" stroke="#000" stroke-linecap="round" stroke-linejoin="round [...]
                   </div>
                 </label>
                 <div class="version-details wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
@@ -184,7 +184,7 @@
                     
                     
                     
-                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.10.dev0 (main)</a></div></li>
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="/">0.11.dev0 (main)</a></div></li>
                     
                     
                     
@@ -196,6 +196,11 @@
                     
                       <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.9.0/">v0.9.0</a></div></li>
                     
+                    
+                    
+                    
+                      <li><div class="version"><a style="font-size: 0.8em; padding: 4px" href="v0.10.0/">v0.10.0</a></div></li>
+                    
                   </ol>
                 </div>
             
@@ -491,11 +496,11 @@ 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), &quot;ThreadIndex&quot;, &quot;blockIdx.x&quot;)] &quot;thread_extent&quot; = 16;
+  attr [IterVar(blockIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;blockIdx.x&quot;)] &quot;thread_extent&quot; = 64;
   allocate(conv2d_nchw: Pointer(local float32), float32, [7]), 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), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224 {
+  allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
+  allocate(kernel.shared: Pointer(shared float32), float32, [192]), storage_scope = shared;
+  attr [IterVar(threadIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56 {
     conv2d_nchw_1: Buffer(conv2d_nchw, float32, [7], [], scope=&quot;local&quot;, align=16)[0] = 0f32
     conv2d_nchw_1[1] = 0f32
     conv2d_nchw_1[2] = 0f32
@@ -503,673 +508,602 @@ cooperative fetching, unrolling and operator fusion.</p>
     conv2d_nchw_1[4] = 0f32
     conv2d_nchw_1[5] = 0f32
     conv2d_nchw_1[6] = 0f32
-    for (rc.outer.outer: int32, 0, 32) {
-      let cse_var_1: int32 = (rc.outer.outer*144)
+    for (rc.outer.outer: int32, 0, 64) {
+      let cse_var_2: int32 = (rc.outer.outer*392)
+      let cse_var_1: int32 = (rc.outer.outer*72)
        {
-        attr [IterVar(threadIdx.x_1: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224 {
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1296], [], scope=&quot;shared&quot;)[(threadIdx.x_1*36)] = 0f32
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 1)] = @tir.if_then_else(((1 &lt;= floormod((threadIdx.x_1*4), 9)) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 7)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 2)] = @tir.if_then_else(((1 &lt;= floormod((threadIdx.x_1*4), 9)) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 6)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 3)] = @tir.if_then_else(((1 &lt;= floormod((threadIdx.x_1*4), 9)) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 5)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 4)] = @tir.if_then_else(((1 &lt;= floormod((threadIdx.x_1*4), 9)) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 4)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 5)] = @tir.if_then_else(((1 &lt;= floormod((threadIdx.x_1*4), 9)) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 3)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 6)] = @tir.if_then_else(((1 &lt;= floormod((threadIdx.x_1*4), 9)) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 2)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 7)] = @tir.if_then_else(((1 &lt;= floormod((threadIdx.x_1*4), 9)) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv((threadIdx.x_1*4), 9)*49)) + (floormod((threadIdx.x_1*4), 9)*7)) - 1)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 8)] = 0f32
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 9)] = 0f32
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 10)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 7)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 11)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 6)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 12)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 5)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 13)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 4)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 14)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 3)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 15)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 2)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 16)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (floormod(((threadIdx.x_1*4) + 1), 9)*7)) - 1)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 17)] = 0f32
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 18)] = 0f32
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 19)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 7)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 20)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 6)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 21)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 5)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 22)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 4)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 23)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 3)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 24)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 2)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 25)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (floormod(((threadIdx.x_1*4) + 2), 9)*7)) - 1)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 26)] = 0f32
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 27)] = 0f32
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 28)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 7)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 29)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 6)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 30)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 5)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 31)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 4)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 32)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 3)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 33)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 2)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 34)] = @tir.if_then_else(((1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9)) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((rc.outer.outer*784) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (floormod(((threadIdx.x_1*4) + 3), 9)*7)) - 1)], 0f32, dtype=float32)
-          }
-          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
-            pad_temp.shared_1[((threadIdx.x_1*36) + 35)] = 0f32
-          }
+        attr [IterVar(threadIdx.x_1: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope=&quot;shared&quot;)[threadIdx.x_1] = @tir.if_then_else(((7 &lt;= threadIdx.x_1) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((cse_var_2 + threadIdx.x_1) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 56)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 8), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 8), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 8), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 4), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 4), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 4), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 336), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else(((threadIdx.x_1 &lt; 49) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 448), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        kernel.shared_1: Buffer(kernel.shared, float32, [192], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 56), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 112), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        if @tir.likely((threadIdx.x_2 &lt; 24), dtype=bool) {
+          kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[((((blockIdx.x*36864) + cse_var_1) + (threadIdx.x_2*3)) + 32256)]
         }
-        attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        kernel.shared_1: Buffer(kernel.shared, float32, [4608], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel[((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 144)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 144))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 8), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 7), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 672), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 96), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 3)*3)) + floormod(threadIdx.x_2, 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 5), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 4), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1344), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 48), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 1), 3)*3)) + floormod(threadIdx.x_2, 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 144)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 144)) + 64512)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 8), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 7), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2688), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 96), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 2), 3)*3)) + floormod(threadIdx.x_2, 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 5), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 4), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        kernel.shared_1[(threadIdx.x_2 + 3360)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3360), 144)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 48), 144), 9)*9)) + (floormod((floordiv(threadIdx.x_2, 3) + 1), 3)*3)) + floormod(threadIdx.x_2, 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + floormod((threadIdx.x_2 + 2), 9))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + floormod((threadIdx.x_2 + 1), 9))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 144)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 144)) + 129024)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 8), 9), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-        if @tir.likely((threadIdx.x_2 &lt; 128), dtype=bool) {
-          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), 9)*9)) + (floordiv(floormod((threadIdx.x_2 + 7), 9), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 14)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 21)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 42)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 56)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 77)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 84)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 91)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 105)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 112)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 119)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 140)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 154)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 161)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 168)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 175)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 182)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 203)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 210)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 217)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 224)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 231)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 238)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 245)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 266)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 273)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 280)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 287)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 294)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 301)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 308)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 329)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 336)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 343)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 350)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 357)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 364)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 371)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 392)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 399)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 406)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 413)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 420)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 427)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 434)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 455)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 462)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 469)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 476)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 483)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 490)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(floormod(threadIdx.x, 7) + 497)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((7 &lt;= threadIdx.x_1), data[((cse_var_2 + threadIdx.x_1) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 56)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 8), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 8), 9) &lt; 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 56), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 8), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) &lt; 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 112), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 6), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 6), 9) &lt; 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 168), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 6), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) &lt; 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 224), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 4), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 4), 9) &lt; 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 280), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 4), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) &lt; 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 336), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 2), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 2), 9) &lt; 8)), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 392), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 2), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else((threadIdx.x_1 &lt; 49), data[((((cse_var_2 + (floordiv((threadIdx.x_1 + 448), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + 1)]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[((((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 56), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + (floormod((threadIdx.x_2 + 2), 3)*3)) + 1)]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 112), 24)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + (floormod((threadIdx.x_2 + 1), 3)*3)) + 1)]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+        if @tir.likely((threadIdx.x_2 &lt; 24), dtype=bool) {
+          kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[((((blockIdx.x*36864) + cse_var_1) + (threadIdx.x_2*3)) + 32257)]
         }
-        for (rc.outer.inner: int32, 0, 2) {
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9))]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*648) + (floormod(threadIdx.x, 7)*9)) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*144) + (rc.outer.inner*72))]))
... 11020 lines suppressed ...