[tvm-site] branch asf-site updated: deploying docs (apache/tvm@daa068979a464bd96f6f48e244a8917c4b353503)

[tq...@apache.org]

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 0800ce2  deploying docs (apache/tvm@daa068979a464bd96f6f48e244a8917c4b353503)
0800ce2 is described below

commit 0800ce2512c250d7ae31adb0b57e8aed92cea5e8
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Sat Apr 2 22:52:04 2022 +0000

    deploying docs (apache/tvm@daa068979a464bd96f6f48e244a8917c4b353503)
---
 .../how_to/compile_models/from_mxnet.rst.txt       |    2 +-
 .../how_to/compile_models/from_paddle.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      |   20 +-
 .../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    |    4 +-
 .../how_to/extend_tvm/sg_execution_times.rst.txt   |   10 +-
 .../how_to/extend_tvm/use_pass_instrument.rst.txt  |   16 +-
 .../optimize_operators/opt_conv_cuda.rst.txt       |    2 +-
 .../optimize_operators/opt_conv_tensorcore.rst.txt |    2 +-
 .../how_to/optimize_operators/opt_gemm.rst.txt     |   16 +-
 .../optimize_operators/sg_execution_times.rst.txt  |    8 +-
 .../sg_execution_times.rst.txt                     |   16 +-
 .../tune_conv2d_layer_cuda.rst.txt                 | 3390 +++++---------------
 .../tune_network_cuda.rst.txt                      |    2 +-
 .../tune_network_x86.rst.txt                       |    4 +-
 .../tune_sparse_x86.rst.txt                        |   32 +-
 .../tune_with_autotvm/sg_execution_times.rst.txt   |   12 +-
 .../tune_with_autotvm/tune_conv2d_cuda.rst.txt     |   34 +-
 .../work_with_microtvm/micro_autotune.rst.txt      |   16 +-
 .../work_with_microtvm/sg_execution_times.rst.txt  |   12 +-
 .../work_with_relay/sg_execution_times.rst.txt     |    8 +-
 .../work_with_schedules/sg_execution_times.rst.txt |   18 +-
 .../how_to/work_with_schedules/tensorize.rst.txt   |    4 +-
 .../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     |    9 +-
 docs/_sources/tutorial/autotvm_relay_x86.rst.txt   |   55 +-
 .../tutorial/cross_compilation_and_rpc.rst.txt     |    2 +-
 docs/_sources/tutorial/intro_topi.rst.txt          |    2 +-
 docs/_sources/tutorial/sg_execution_times.rst.txt  |   26 +-
 .../tutorial/tensor_expr_get_started.rst.txt       |   49 +-
 docs/commit_hash                                   |    2 +-
 docs/how_to/compile_models/from_mxnet.html         |    2 +-
 docs/how_to/compile_models/from_paddle.html        |    2 +-
 docs/how_to/compile_models/from_pytorch.html       |   33 +-
 docs/how_to/compile_models/from_tensorflow.html    |    2 +-
 docs/how_to/compile_models/sg_execution_times.html |   20 +-
 .../deploy_models/deploy_model_on_android.html     |    2 +-
 .../deploy_object_detection_pytorch.html           |  118 +-
 docs/how_to/deploy_models/deploy_prequantized.html |   10 +-
 .../deploy_models/deploy_prequantized_tflite.html  |    4 +-
 docs/how_to/deploy_models/deploy_quantized.html    |    2 +-
 docs/how_to/deploy_models/deploy_ssd_gluoncv.html  |   38 +-
 docs/how_to/deploy_models/sg_execution_times.html  |   18 +-
 .../extend_tvm/bring_your_own_datatypes.html       |    4 +-
 docs/how_to/extend_tvm/sg_execution_times.html     |   10 +-
 docs/how_to/extend_tvm/use_pass_instrument.html    |   16 +-
 docs/how_to/optimize_operators/opt_conv_cuda.html  |    2 +-
 .../optimize_operators/opt_conv_tensorcore.html    |    2 +-
 docs/how_to/optimize_operators/opt_gemm.html       |   16 +-
 .../optimize_operators/sg_execution_times.html     |    8 +-
 .../sg_execution_times.html                        |   14 +-
 .../tune_conv2d_layer_cuda.html                    | 3390 +++++---------------
 .../tune_with_autoscheduler/tune_network_cuda.html |    2 +-
 .../tune_with_autoscheduler/tune_network_x86.html  |    4 +-
 .../tune_with_autoscheduler/tune_sparse_x86.html   |   32 +-
 .../tune_with_autotvm/sg_execution_times.html      |   12 +-
 .../how_to/tune_with_autotvm/tune_conv2d_cuda.html |   34 +-
 docs/how_to/work_with_microtvm/micro_autotune.html |   16 +-
 .../work_with_microtvm/sg_execution_times.html     |   12 +-
 .../how_to/work_with_relay/sg_execution_times.html |    8 +-
 .../work_with_schedules/sg_execution_times.html    |   18 +-
 docs/how_to/work_with_schedules/tensorize.html     |    4 +-
 docs/reference/api/python/auto_scheduler.html      |    4 +-
 .../api/typedoc/classes/bytestreamreader.html      |   12 +-
 .../api/typedoc/classes/cachedcallstack.html       |   34 +-
 docs/reference/api/typedoc/classes/dldatatype.html |   12 +-
 docs/reference/api/typedoc/classes/dldevice.html   |   10 +-
 .../reference/api/typedoc/classes/environment.html |   12 +-
 docs/reference/api/typedoc/classes/ffilibrary.html |   20 +-
 .../api/typedoc/classes/graphexecutor.html         |   16 +-
 docs/reference/api/typedoc/classes/instance.html   |   40 +-
 docs/reference/api/typedoc/classes/memory.html     |   34 +-
 docs/reference/api/typedoc/classes/module.html     |   10 +-
 docs/reference/api/typedoc/classes/ndarray.html    |   22 +-
 .../api/typedoc/classes/packedfunccell.html        |    6 +-
 docs/reference/api/typedoc/classes/rpcserver.html  |   14 +-
 docs/reference/api/typedoc/classes/scalar.html     |    6 +-
 .../api/typedoc/classes/webgpucontext.html         |   12 +-
 docs/reference/api/typedoc/enums/argtypecode.html  |   30 +-
 .../api/typedoc/enums/aynccallbackcode.html        |    4 +-
 .../api/typedoc/enums/dldatatypecode.html          |    8 +-
 .../api/typedoc/enums/rpcserverstate.html          |   12 +-
 docs/reference/api/typedoc/enums/sizeof.html       |   18 +-
 docs/reference/api/typedoc/index.html              |  112 +-
 .../api/typedoc/interfaces/disposable.html         |    2 +-
 .../api/typedoc/interfaces/functioninfo.html       |    6 +-
 .../api/typedoc/interfaces/libraryprovider.html    |    4 +-
 docs/searchindex.js                                |    2 +-
 .../vta/tutorials/autotvm/sg_execution_times.html  |    6 +-
 .../tutorials/frontend/deploy_classification.html  |    2 +-
 .../vta/tutorials/frontend/deploy_detection.html   |    2 +-
 .../vta/tutorials/frontend/sg_execution_times.html |    6 +-
 .../vta/tutorials/optimize/sg_execution_times.html |    6 +-
 docs/topic/vta/tutorials/sg_execution_times.html   |    6 +-
 docs/tutorial/auto_scheduler_matmul_x86.html       |    5 +-
 docs/tutorial/autotvm_relay_x86.html               |  168 +-
 docs/tutorial/cross_compilation_and_rpc.html       |    2 +-
 docs/tutorial/intro_topi.html                      |    2 +-
 docs/tutorial/sg_execution_times.html              |   26 +-
 docs/tutorial/tensor_expr_get_started.html         |   45 +-
 113 files changed, 2618 insertions(+), 5822 deletions(-)

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 392c7b2..0fdde29 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -98,7 +98,7 @@ In this section, we download a pretrained imagenet model and classify an image.
 
  .. code-block:: none
 
-    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipb5406133-5280-46f8-94d4-fac0f9dec188 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipcb8bcc43-51ed-43ad-a91c-4e6a7e0c1b40 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_paddle.rst.txt b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
index b72dd0b..ce980f5 100644
--- a/docs/_sources/how_to/compile_models/from_paddle.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
@@ -182,7 +182,7 @@ Look up prediction top 1 index in 1000 class synset.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  4.026 seconds)
+   **Total running time of the script:** ( 1 minutes  3.921 seconds)
 
 
 .. _sphx_glr_download_how_to_compile_models_from_paddle.py:
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 c49268c..30032cc 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -79,7 +79,7 @@ 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]
      6%|6         | 2.75M/44.7M [00:00<00:01, 26.0MB/s]
     14%|#4        | 6.30M/44.7M [00:00<00:01, 32.3MB/s]
     22%|##1       | 9.67M/44.7M [00:00<00:01, 31.1MB/s]
     30%|###       | 13.4M/44.7M [00:00<00:00, 33.4MB/s]
     37%|###7      | 16.6M/44.7M [00:00<00:01, 25.0MB/s]
     43%|####3     | 19.3M/44.7M [00:00<00:01, 23.8MB/s]
     50%|####9     | 22.1M/44.7M [00:00<00:00, 24.9MB/s]
     56%|#####6    | 25.2M/44.7M [00:00<00:00, 26.0MB/s]
     67%|######6   | 29.8M/44.7M [00:01<00:00, 32.0MB/s]
     74%|#######3  | 33.0M/44.7M [00:01<00:00, 31.1MB/s]
     81%|########  | 36.1M/44.7M [00:01<00:00, 30.4MB/s]
     88%|########7 | 39.1M/44.7M [00:01<00:00, 30.5MB/s]
     94%|#########4| 42.1M/44.7M [00:01<00:00, 18.4MB/s]
    100%|#########9| 44.6M/44.7M [00:01<00:00, 20.0MB/s]
    100%|##########| 44.7M/44.7M [00:01<00:00, 25.3MB/s]
+
      0%|          | 0.00/44.7M [00:00<?, ?B/s]
      4%|3         | 1.62M/44.7M [00:00<00:03, 15.0MB/s]
      7%|7         | 3.31M/44.7M [00:00<00:02, 16.0MB/s]
     12%|#1        | 5.19M/44.7M [00:00<00:02, 17.5MB/s]
     16%|#5        | 7.06M/44.7M [00:00<00:02, 17.9MB/s]
     20%|#9        | 8.78M/44.7M [00:00<00:02, 16.6MB/s]
     27%|##6       | 12.0M/44.7M [00:00<00:01, 22.0MB/s]
     32%|###1      | 14.1M/44.7M [00:00<00:01, 21.1MB/s]
     36%|###6      | 16.2M/44.7M [00:00<00:01, 18.8MB/s]
     40%|####      | 18.1M/44.7M [00:01<00:01, 14.1MB/s]
     44%|####4     | 19.7M/44.7M [00:01<00:01, 14.5MB/s]
     52%|#####1    | 23.1M/44.7M [00:01<00:01, 19.7MB/s]
     58%|#####8    | 26.1M/44.7M [00:01<00:00, 22.6MB/s]
     64%|######3   | 28.4M/44.7M [00:01<00:00, 21.7MB/s]
     74%|#######3  | 32.9M/44.7M [00:01<00:00, 27.8MB/s]
     81%|########  | 36.0M/44.7M [00:01<00:00, 29.0MB/s]
     88%|########7 | 39.2M/44.7M [00:01<00:00, 26.8MB/s]
     94%|#########3| 41.9M/44.7M [00
 :02<00:00, 24.4MB/s]
    100%|##########| 44.7M/44.7M [00:02<00:00, 21.8MB/s]
 
 
 
diff --git a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
index 4efdb0f..1c41551 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -372,7 +372,7 @@ Run the corresponding model on tensorflow
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  3.686 seconds)
+   **Total running time of the script:** ( 1 minutes  3.065 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 b2a13d3..2dc6bd7 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
@@ -5,14 +5,14 @@
 
 Computation times
 =================
-**04:46.638** total execution time for **how_to_compile_models** files:
+**04:46.054** total execution time for **how_to_compile_models** files:
 
-- **01:04.026**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
-- **01:03.686**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
-- **00:55.234**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
-- **00:25.504**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
-- **00:22.062**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
-- **00:20.919**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
-- **00:20.697**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
-- **00:12.018**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
-- **00:02.490**: :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)
+- **01:03.921**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
+- **01:03.065**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
+- **00:55.658**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
+- **00:25.455**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
+- **00:21.090**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
+- **00:20.918**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
+- **00:20.577**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
+- **00:12.441**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
+- **00:02.928**: :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)
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 b742a71..5d0ed67 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
@@ -393,7 +393,7 @@ Execute on TVM
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      15.8919      15.8639      16.1597      15.7815       0.1118   
+      15.6399      15.6042      15.8487      15.5657       0.0846   
                
 
 
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 208be8d..bd7e06f 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
@@ -108,7 +108,7 @@ 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]
      2%|2         | 3.56M/170M [00:00<00:04, 36.9MB/s]
      4%|4         | 7.09M/170M [00:00<00:06, 28.4MB/s]
      6%|5         | 9.91M/170M [00:00<00:06, 25.7MB/s]
      8%|7         | 13.0M/170M [00:00<00:05, 27.9MB/s]
      9%|9         | 15.8M/170M [00:00<00:06, 26.5MB/s]
     11%|#1        | 19.2M/170M [00:00<00:05, 27.0MB/s]
     13%|#2        | 21.8M/170M [00:00<00:06, 25.8MB/s]
     15%|#4        | 25.1M/170M [00:00<00:05, 28.3MB/s]
     16%|#6        | 27.9M/170M [00:01<00:05, 25.6MB/s]
     18%|#7        | 30.4M/170M [00:01<00:08, 18.2MB/s]
     19%|#9        | 32.4M/170M [00:01<00:08, 17.5MB/s]
     20%|##        | 34.3M/170M [00:01<00:08, 17.7MB/s]
     21%|##1       | 36.5M/170M [00:01<00:07, 19.1MB/s]
     23%|##2       | 38.5M/170M [00:01<00:07, 17.9MB/s]
     24%|##3       | 40.7M/170M [00:01<00:07, 19.2MB/s]
     26%|##5       | 43.8M/170M [00:02<00:05, 22.5MB/s]
     27%|##7       | 46.0M/170M [00:02<00:06, 20.9MB/
 s]
     29%|##8       | 48.8M/170M [00:02<00:05, 22.7MB/s]
     30%|###       | 51.6M/170M [00:02<00:05, 24.6MB/s]
     32%|###2      | 55.1M/170M [00:02<00:04, 25.3MB/s]
     34%|###3      | 57.5M/170M [00:02<00:05, 22.9MB/s]
     36%|###6      | 61.6M/170M [00:02<00:04, 27.6MB/s]
     38%|###8      | 65.4M/170M [00:02<00:03, 29.8MB/s]
     41%|####      | 68.9M/170M [00:02<00:03, 31.7MB/s]
     42%|####2     | 72.0M/170M [00:03<00:03, 31.1MB/s]
     44%|####4     | 75.1M/170M [00:03<00:03, 28.8MB/s]
     46%|####5     | 77.9M/170M [00:03<00:03, 28.4MB/s]
     47%|####7     | 80.6M/170M [00:03<00:03, 26.1MB/s]
     49%|####8     | 83.2M/170M [00:03<00:03, 26.2MB/s]
     50%|#####     | 85.7M/170M [00:03<00:04, 18.9MB/s]
     52%|#####2    | 88.6M/170M [00:03<00:03, 21.4MB/s]
     54%|#####3    | 90.9M/170M [00:04<00:03, 21.4MB/s]
     55%|#####4    | 93.2M/170M [00:04<00:03, 21.4MB/s]
     56%|#####6    | 95.4M/170M [00:04<00:03, 20.5MB/s]
     57%|#####7    | 97.4M/170M [00:04<00:
 03, 20.5MB/s]
     59%|#####9    | 101M/170M [00:04<00:02, 24.4MB/s] 
     61%|######1   | 104M/170M [00:04<00:02, 26.9MB/s]
     63%|######3   | 107M/170M [00:04<00:02, 28.8MB/s]
     65%|######4   | 110M/170M [00:04<00:02, 28.3MB/s]
     67%|######6   | 113M/170M [00:04<00:02, 26.5MB/s]
     69%|######9   | 117M/170M [00:04<00:01, 31.6MB/s]
     71%|#######   | 120M/170M [00:05<00:01, 29.7MB/s]
     73%|#######2  | 124M/170M [00:05<00:01, 31.0MB/s]
     75%|#######4  | 127M/170M [00:05<00:01, 30.7MB/s]
     77%|#######6  | 130M/170M [00:05<00:01, 32.2MB/s]
     78%|#######8  | 133M/170M [00:05<00:01, 29.6MB/s]
     80%|########  | 137M/170M [00:05<00:01, 30.8MB/s]
     82%|########2 | 140M/170M [00:05<00:00, 32.0MB/s]
     84%|########4 | 143M/170M [00:05<00:01, 27.2MB/s]
     86%|########6 | 146M/170M [00:06<00:00, 29.2MB/s]
     89%|########8 | 151M/170M [00:06<00:00, 33.7MB/s]
     91%|######### | 154M/170M [00:06<00:00, 35.1MB/s]
     93%|#########2| 158M/170M [00:06<00:00, 29
 .2MB/s]
     95%|#########4| 161M/170M [00:06<00:00, 29.9MB/s]
     97%|#########7| 165M/170M [00:06<00:00, 32.9MB/s]
     99%|#########9| 168M/170M [00:06<00:00, 32.2MB/s]
    100%|##########| 170M/170M [00:06<00:00, 26.3MB/s]
+
      0%|          | 0.00/170M [00:00<?, ?B/s]
      1%|1         | 1.81M/170M [00:00<00:09, 18.9MB/s]
      2%|2         | 4.24M/170M [00:00<00:07, 22.7MB/s]
      4%|3         | 6.41M/170M [00:00<00:07, 22.7MB/s]
      5%|5         | 8.57M/170M [00:00<00:07, 22.2MB/s]
      7%|6         | 11.4M/170M [00:00<00:06, 24.9MB/s]
      8%|8         | 13.8M/170M [00:00<00:07, 20.5MB/s]
     10%|#         | 17.2M/170M [00:00<00:06, 24.7MB/s]
     12%|#1        | 19.7M/170M [00:00<00:07, 20.9MB/s]
     13%|#3        | 22.7M/170M [00:01<00:06, 23.2MB/s]
     15%|#5        | 25.6M/170M [00:01<00:06, 25.0MB/s]
     17%|#6        | 28.1M/170M [00:01<00:05, 24.9MB/s]
     18%|#8        | 30.6M/170M [00:01<00:06, 23.1MB/s]
     19%|#9        | 32.9M/170M [00:01<00:06, 22.4MB/s]
     21%|##        | 35.1M/170M [00:01<00:06, 22.4MB/s]
     22%|##1       | 37.2M/170M [00:01<00:06, 21.2MB/s]
     23%|##3       | 39.3M/170M [00:01<00:06, 20.4MB/s]
     24%|##4       | 41.3M/170M [00:02<00:08, 16.2MB/
 s]
     25%|##5       | 43.0M/170M [00:02<00:08, 14.8MB/s]
     27%|##6       | 45.3M/170M [00:02<00:07, 16.9MB/s]
     28%|##8       | 48.0M/170M [00:02<00:06, 19.8MB/s]
     30%|##9       | 50.8M/170M [00:02<00:05, 22.4MB/s]
     31%|###1      | 53.1M/170M [00:02<00:05, 22.1MB/s]
     33%|###2      | 55.4M/170M [00:02<00:05, 22.4MB/s]
     35%|###5      | 59.6M/170M [00:02<00:04, 27.3MB/s]
     38%|###7      | 64.2M/170M [00:02<00:03, 33.0MB/s]
     40%|####      | 68.4M/170M [00:03<00:02, 36.1MB/s]
     42%|####2     | 71.9M/170M [00:03<00:03, 33.8MB/s]
     45%|####4     | 76.1M/170M [00:03<00:02, 36.1MB/s]
     47%|####6     | 79.6M/170M [00:03<00:03, 30.6MB/s]
     50%|####9     | 84.1M/170M [00:03<00:02, 34.4MB/s]
     52%|#####1    | 87.6M/170M [00:03<00:02, 30.3MB/s]
     53%|#####3    | 90.7M/170M [00:04<00:04, 17.1MB/s]
     55%|#####4    | 93.0M/170M [00:04<00:05, 15.3MB/s]
     56%|#####6    | 95.5M/170M [00:04<00:04, 16.8MB/s]
     57%|#####7    | 97.5M/170M [00:04<00:
 04, 17.2MB/s]
     59%|#####8    | 99.5M/170M [00:04<00:04, 17.3MB/s]
     60%|######    | 102M/170M [00:04<00:03, 18.7MB/s] 
     62%|######1   | 105M/170M [00:04<00:03, 21.3MB/s]
     63%|######3   | 107M/170M [00:05<00:03, 18.6MB/s]
     65%|######4   | 110M/170M [00:05<00:03, 20.0MB/s]
     66%|######6   | 113M/170M [00:05<00:02, 22.8MB/s]
     68%|######7   | 115M/170M [00:05<00:02, 22.8MB/s]
     70%|#######   | 119M/170M [00:05<00:01, 28.5MB/s]
     73%|#######2  | 123M/170M [00:05<00:01, 31.9MB/s]
     74%|#######4  | 126M/170M [00:05<00:01, 30.5MB/s]
     76%|#######6  | 129M/170M [00:05<00:01, 28.9MB/s]
     78%|#######7  | 132M/170M [00:05<00:01, 27.7MB/s]
     80%|#######9  | 135M/170M [00:06<00:01, 28.1MB/s]
     82%|########2 | 139M/170M [00:06<00:00, 32.3MB/s]
     84%|########3 | 142M/170M [00:06<00:00, 31.5MB/s]
     86%|########5 | 146M/170M [00:06<00:01, 24.9MB/s]
     88%|########7 | 149M/170M [00:06<00:00, 27.7MB/s]
     90%|######### | 153M/170M [00:06<00:00, 3
 1.2MB/s]
     92%|#########2| 156M/170M [00:06<00:00, 31.7MB/s]
     94%|#########3| 160M/170M [00:06<00:00, 29.6MB/s]
     96%|#########5| 162M/170M [00:07<00:00, 27.3MB/s]
     98%|#########7| 166M/170M [00:07<00:00, 29.3MB/s]
     99%|#########9| 169M/170M [00:07<00:00, 28.1MB/s]
    100%|##########| 170M/170M [00:07<00:00, 24.4MB/s]
     /usr/local/lib/python3.7/dist-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)
     /usr/local/lib/python3.7/dist-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').
@@ -253,7 +253,7 @@ Get boxes with score larger than 0.9
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  7.507 seconds)
+   **Total running time of the script:** ( 3 minutes  5.573 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 776779e..1847503 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -187,7 +187,7 @@ 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]
     51%|#####1    | 6.95M/13.6M [00:00<00:00, 72.9MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 79.7MB/s]
+
      0%|          | 0.00/13.6M [00:00<?, ?B/s]
     27%|##7       | 3.69M/13.6M [00:00<00:00, 38.6MB/s]
     54%|#####4    | 7.38M/13.6M [00:00<00:00, 37.8MB/s]
     81%|########1 | 11.0M/13.6M [00:00<00:00, 30.2MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 29.5MB/s]
 
 
 
@@ -344,7 +344,7 @@ 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)  
-      90.0946      90.0127      92.1835      89.8798       0.3329   
+      90.1165      90.0963      90.5290      89.9079       0.1071   
                
 
 
@@ -384,7 +384,7 @@ TODO
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  3.698 seconds)
+   **Total running time of the script:** ( 1 minutes  3.679 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 cb66a3a..8885df9 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
@@ -351,7 +351,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      118.8002     118.7311     125.5149     117.6018      0.8094   
+      120.1621     120.0953     121.5890     119.3815      0.3826   
                
 
 
@@ -385,7 +385,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  2.262 seconds)
+   **Total running time of the script:** ( 2 minutes  0.330 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 f634b0b..ece0182 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -221,7 +221,7 @@ We create a Relay VM to build and execute the model.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  12.617 seconds)
+   **Total running time of the script:** ( 1 minutes  9.260 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 30052eb..e6a80cc 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
@@ -137,7 +137,7 @@ 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]
      2%|1         | 2355/132723 [00:00<00:05, 23547.16KB/s]
      5%|5         | 6897/132723 [00:00<00:03, 36411.79KB/s]
     11%|#         | 14426/132723 [00:00<00:02, 54159.47KB/s]
     17%|#6        | 22218/132723 [00:00<00:01, 63532.80KB/s]
     23%|##2       | 29969/132723 [00:00<00:01, 68570.90KB/s]
     28%|##8       | 37784/132723 [00:00<00:01, 71824.80KB/s]
     34%|###4      | 45553/132723 [00:00<00:01, 73739.37KB/s]
     40%|####      | 53409/132723 [00:00<00:01, 75270.44KB/s]
     46%|####6     | 61325/132723 [00:00<00:00, 76483.50KB/s]
     52%|#####2    | 69215/132723 [00:01<00:00, 77228.17KB/s]
     58%|#####8    | 77139/132723 [00:01<00:00, 77842.06KB/s]
     64%|######4   | 85089/132723 [00:01<00:00, 78341.01KB/s]
     70%|#######   | 93065/132723 [00:01<00:00, 78743.56KB/s]
     76%|#######6  | 100940/132723 [00:01<00:00, 78694.31KB/s]
     82%|########1 | 108810/132723 [00:01<00:00, 78686.50KB/s]
     88%|########7 
 | 116707/132723 [00:01<00:00, 78769.49KB/s]
     94%|#########3| 124688/132723 [00:01<00:00, 79080.17KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 79053.09KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 73609.57KB/s]
+
      0%|          | 0/132723 [00:00<?, ?KB/s]
      3%|3         | 4049/132723 [00:00<00:03, 40486.18KB/s]
      8%|7         | 10020/132723 [00:00<00:02, 50142.97KB/s]
     13%|#3        | 17656/132723 [00:00<00:01, 61901.21KB/s]
     20%|#9        | 26112/132723 [00:00<00:01, 70748.49KB/s]
     26%|##6       | 34522/132723 [00:00<00:01, 75527.74KB/s]
     32%|###2      | 43021/132723 [00:00<00:01, 78726.98KB/s]
     39%|###8      | 51464/132723 [00:00<00:01, 80582.82KB/s]
     45%|####5     | 59957/132723 [00:00<00:00, 81962.62KB/s]
     52%|#####1    | 68507/132723 [00:00<00:00, 83065.29KB/s]
     58%|#####8    | 77012/132723 [00:01<00:00, 83676.10KB/s]
     64%|######4   | 85589/132723 [00:01<00:00, 84309.43KB/s]
     71%|#######   | 94109/132723 [00:01<00:00, 84578.26KB/s]
     77%|#######7  | 102629/132723 [00:01<00:00, 84759.21KB/s]
     84%|########3 | 111139/132723 [00:01<00:00, 84859.87KB/s]
     90%|######### | 119626/132723 [00:01<00:00, 84771.34KB/s]
     97%|########
 #6| 128134/132723 [00:01<00:00, 84862.91KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 79867.17KB/s]
 
 
 
@@ -202,7 +202,7 @@ Display result
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  21.077 seconds)
+   **Total running time of the script:** ( 2 minutes  19.503 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 935a54b..02bb7c0 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
@@ -5,13 +5,13 @@
 
 Computation times
 =================
-**10:36.641** total execution time for **how_to_deploy_models** files:
+**10:26.863** total execution time for **how_to_deploy_models** files:
 
-- **03:07.507**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
-- **02:21.077**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
-- **02:02.262**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
-- **01:12.617**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
-- **01:03.698**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
-- **00:27.408**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
-- **00:21.882**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
-- **00:00.189**: :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)
+- **03:05.573**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
+- **02:19.503**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
+- **02:00.330**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
+- **01:09.260**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
+- **01:03.679**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
+- **00:26.693**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
+- **00:21.642**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
+- **00:00.183**: :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)
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 68e7b57..3e3bb01 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
@@ -423,7 +423,7 @@ 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.zip61a6064e-a681-4b2c-8249-0be39dd15fe4 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+    Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip9bcc8657-c468-43c6-8edf-4fc5c507204b from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
 
 
 
@@ -525,7 +525,7 @@ Now, to actually convert the entire network, we have written `a pass in Relay <h
 
  .. code-block:: none
 
-      Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
+      Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
 
 
 
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 b7477c7..a15fc3d 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
@@ -5,9 +5,9 @@
 
 Computation times
 =================
-**00:38.763** total execution time for **how_to_extend_tvm** files:
+**00:37.626** total execution time for **how_to_extend_tvm** files:
 
-- **00:35.196**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
-- **00:02.284**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
-- **00:01.080**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
-- **00:00.203**: :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)
+- **00:34.113**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
+- **00:02.239**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
+- **00:01.074**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
+- **00:00.199**: :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)
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 0541e1d..549bb19 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
@@ -199,10 +199,10 @@ profile the execution time of each passes.
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 6728us [6728us] (47.31%; 47.31%)
-    FoldScaleAxis: 7493us [3us] (52.69%; 52.69%)
-            FoldConstant: 7490us [1591us] (52.67%; 99.97%)
-                    InferType: 5900us [5900us] (41.49%; 78.76%)
+    InferType: 6184us [6184us] (45.49%; 45.49%)
+    FoldScaleAxis: 7411us [2us] (54.51%; 54.51%)
+            FoldConstant: 7408us [1532us] (54.49%; 99.97%)
+                    InferType: 5876us [5876us] (43.23%; 79.32%)
 
 
 
@@ -239,10 +239,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 5981us [5981us] (44.90%; 44.90%)
-    FoldScaleAxis: 7338us [2us] (55.10%; 55.10%)
-            FoldConstant: 7336us [1520us] (55.08%; 99.97%)
-                    InferType: 5816us [5816us] (43.67%; 79.28%)
+    InferType: 5964us [5964us] (44.54%; 44.54%)
+    FoldScaleAxis: 7425us [2us] (55.46%; 55.46%)
+            FoldConstant: 7424us [1539us] (55.44%; 99.97%)
+                    InferType: 5885us [5885us] (43.95%; 79.27%)
 
 
 
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 e6be97b..a4503e6 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
@@ -295,7 +295,7 @@ latency of convolution.
 
  .. code-block:: none
 
-    Convolution: 54.113171 ms
+    Convolution: 41.541017 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 ff0496d..305d45f 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
@@ -626,7 +626,7 @@ be able to run on our build server
 
  .. code-block:: none
 
-    conv2d with tensor core: 6.615613 ms
+    conv2d with tensor core: 10.331691 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 87760f4..6d9b575 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -118,8 +118,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
 
  .. code-block:: none
 
-    Numpy running time: 0.018181
-    Baseline: 3.428213
+    Numpy running time: 0.019122
+    Baseline: 3.457420
 
 
 
@@ -209,7 +209,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
 
  .. code-block:: none
 
-    Opt1: 0.288643
+    Opt1: 0.296049
 
 
 
@@ -307,7 +307,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
 
  .. code-block:: none
 
-    Opt2: 0.330478
+    Opt2: 0.326334
 
 
 
@@ -398,7 +398,7 @@ the access pattern for A matrix is more cache friendly.
 
  .. code-block:: none
 
-    Opt3: 0.118830
+    Opt3: 0.115367
 
 
 
@@ -516,7 +516,7 @@ flattening.
 
  .. code-block:: none
 
-    Opt4: 0.110923
+    Opt4: 0.111165
 
 
 
@@ -633,7 +633,7 @@ write to C when all the block results are ready.
 
  .. code-block:: none
 
-    Opt5: 0.111604
+    Opt5: 0.111855
 
 
 
@@ -753,7 +753,7 @@ Futhermore, we can also utilize multi-core processors to do the thread-level par
 
  .. code-block:: none
 
-    Opt6: 0.144703
+    Opt6: 0.144806
 
 
 
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 fbe7840..7b84ebe 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
@@ -5,8 +5,8 @@
 
 Computation times
 =================
-**00:34.798** total execution time for **how_to_optimize_operators** files:
+**00:35.060** total execution time for **how_to_optimize_operators** files:
 
-- **00:32.189**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
-- **00:01.378**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
-- **00:01.232**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)
+- **00:32.395**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
+- **00:01.459**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
+- **00:01.207**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)
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 e01526d..dcfc7a7 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
@@ -5,11 +5,11 @@
 
 Computation times
 =================
-**04:58.318** total execution time for **how_to_tune_with_autoscheduler** files:
-
-- **02:24.018**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
-- **01:20.658**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
-- **00:40.577**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
-- **00:16.002**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
-- **00:08.604**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
-- **00:08.459**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)
+**05:04.467** total execution time for **how_to_tune_with_autoscheduler** files:
+
+- **02:29.882**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
+- **01:19.112**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
+- **00:40.010**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
+- **00:18.765**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
+- **00:08.566**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
+- **00:08.132**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)
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 941f8cd..6648475 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
@@ -223,1338 +223,481 @@ cooperative fetching, unrolling and operator fusion.
       buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute} {
       attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 28;
       allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
-      allocate(pad_temp.shared: Pointer(shared float32), float32, [216]), storage_scope = shared;
-      allocate(kernel.shared: Pointer(shared float32), float32, [9216]), storage_scope = shared;
+      allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
+      allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
       attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
-        conv2d_nchw_1: Buffer(conv2d_nchw, float32, [49], [], scope="local", align=16)[0] = 0f32
-        conv2d_nchw_1[7] = 0f32
+        conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
         conv2d_nchw_1[1] = 0f32
-        conv2d_nchw_1[8] = 0f32
         conv2d_nchw_1[2] = 0f32
-        conv2d_nchw_1[9] = 0f32
         conv2d_nchw_1[3] = 0f32
-        conv2d_nchw_1[10] = 0f32
         conv2d_nchw_1[4] = 0f32
-        conv2d_nchw_1[11] = 0f32
         conv2d_nchw_1[5] = 0f32
-        conv2d_nchw_1[12] = 0f32
         conv2d_nchw_1[6] = 0f32
+        conv2d_nchw_1[7] = 0f32
+        conv2d_nchw_1[8] = 0f32
+        conv2d_nchw_1[9] = 0f32
+        conv2d_nchw_1[10] = 0f32
+        conv2d_nchw_1[11] = 0f32
+        conv2d_nchw_1[12] = 0f32
         conv2d_nchw_1[13] = 0f32
         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" = 64;
-            pad_temp.shared_1: Buffer(pad_temp.shared, float32, [216], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 27), 9) + floormod(blockIdx.x, 7))) && ((floordiv(floormod(threadIdx.x_1, 27), 9) + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[(((((cse_var_2 + (floordiv(threadIdx.x_1, 27)*49)) + (floordiv(floormod(threadIdx.x_1, 27), 9)*7)) + (floormod(blockIdx.x, 7)*7 [...]
-            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            pad_temp.shared_1[(threadIdx.x_1 + 64)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 64), 27), 9) + floormod(blockIdx.x, 7))) && ((floordiv(floormod((threadIdx.x_1 + 64), 27), 9) + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 64), 27)*49)) + (floordiv(floormod((threadIdx.x_1 + 64), 27), 9)*7)) + (floormod(blockIdx.x, 7)*7)) + floormod( [...]
-            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            pad_temp.shared_1[(threadIdx.x_1 + 128)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 128), 27), 9) + floormod(blockIdx.x, 7))) && ((floordiv(floormod((threadIdx.x_1 + 128), 27), 9) + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 128), 27)*49)) + (floordiv(floormod((threadIdx.x_1 + 128), 27), 9)*7)) + (floormod(blockIdx.x, 7)*7)) + floo [...]
-            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            if @tir.likely((threadIdx.x_1 < 24), dtype=bool) {
-              pad_temp.shared_1[(threadIdx.x_1 + 192)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 192), 27), 9) + floormod(blockIdx.x, 7))) && ((floordiv(floormod((threadIdx.x_1 + 192), 27), 9) + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 192), 27)*49)) + (floordiv(floormod((threadIdx.x_1 + 192), 27), 9)*7)) + (floormod(blockIdx.x, 7)*7)) + fl [...]
+          for (ry.outer.outer: int32, 0, 3) {
+            let cse_var_2: int32 = (rc.outer.outer*72)
+            let cse_var_1: int32 = (ry.outer.outer*3)
+             {
+              attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
+                if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+                  pad_temp.shared_1: Buffer(pad_temp.shared, float32, [72], [], scope="shared")[(threadIdx.x_1*4)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod((threadIdx.x_1*4), 9))) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*4), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod((threadIdx.x_1*4), 9)) - 8)], 0f3 [...]
+                }
+                if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+                  pad_temp.shared_1[((threadIdx.x_1*4) + 1)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 1), 9))) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 1), 9)) - 8)], 0f32, dtype=float32)
+                }
+                if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+                  pad_temp.shared_1[((threadIdx.x_1*4) + 2)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 2), 9))) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 2), 9)) - 8)], 0f32, dtype=float32)
+                }
+                if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+                  pad_temp.shared_1[((threadIdx.x_1*4) + 3)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 3), 9))) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 3), 9)) - 8)], 0f32, dtype=float32)
+                }
+              }
+              attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 8), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 64), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 16), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 128), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 32), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 256), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 40), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 320), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 56), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 448), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 64), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 512), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 80), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 640), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 88), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 704), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 147456)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 104), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 832), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 112), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 896), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 184320)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 128), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1024), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 136), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1088), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 221184)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 152), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1216), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 160), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1280), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 176), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1408), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 184), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1472), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 294912)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 200), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1600), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 208), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1664), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 331776)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 224), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1792), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 232), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1856), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 368640)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 248), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1984), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 256), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2048), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 405504)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 272), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2176), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 280), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2240), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 442368)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 296), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2368), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 304), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2432), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 479232)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 320), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2560), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 328), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2624), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 516096)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 344), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2752), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 352), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2816), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 552960)]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 368), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2944), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
+              kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 376), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 3008), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*48)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*48)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*48)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*48)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*48)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*48)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*48)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 47)]))
             }
-            attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1: Buffer(kernel.shared, float32, [9216], [], scope="shared")[threadIdx.x_2] = kernel[(((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 8), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 16), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 24), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 32), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 40), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 48), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 56), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 64), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 36864)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 80), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 88), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 96), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 104), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 120), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 128), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 136), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 73728)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 152), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 160), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 168), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 176), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 184), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 192), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 200), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 208), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 110592)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 224), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 232), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 240), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 248), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 256), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 264), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 272), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 280), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 147456)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 296), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 304), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 312), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 320), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 328), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 336), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 344), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 352), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 184320)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 368), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 376), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3072)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 384), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 392), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3200)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 400), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3264)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 408), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3328)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 416), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3392)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 424), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3456)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 221184)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3520)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 440), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3584)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 448), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3648)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 456), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3712)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 464), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3776)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 472), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3840)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 480), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3904)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 488), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 3968)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 496), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 258048)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4096)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 512), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4160)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 520), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4224)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 528), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4288)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 536), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4352)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 544), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4416)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 552), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4480)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 560), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4544)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 568), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4608)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 294912)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4672)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 584), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4736)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 592), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4800)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 600), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4864)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 608), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4928)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 616), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 4992)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 624), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5056)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 632), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5120)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 640), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5184)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 331776)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5248)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 656), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5312)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 664), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5376)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 672), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5440)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 680), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5504)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 688), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5568)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 696), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5632)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 704), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5696)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 712), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5760)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 368640)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5824)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 728), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5888)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 736), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 5952)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 744), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6016)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 752), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6080)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 760), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6144)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 768), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6208)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 776), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6272)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 784), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6336)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 405504)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6400)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 800), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6464)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 808), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6528)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 816), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6592)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 824), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6656)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 832), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6720)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 840), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6784)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 848), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6848)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 856), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6912)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 442368)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 6976)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 872), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7040)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 880), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7104)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 888), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7168)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 896), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7232)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 904), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7296)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 912), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7360)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 920), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7424)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 928), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7488)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 479232)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7552)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 944), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7616)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 952), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7680)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 960), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7744)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 968), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7808)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 976), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7872)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 984), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 7936)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 992), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8000)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1000), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8064)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 516096)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8128)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1016), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8192)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1024), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8256)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1032), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8320)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1040), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8384)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1048), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8448)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1056), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8512)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1064), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8576)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1072), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8640)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 552960)]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8704)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1088), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8768)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1096), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8832)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1104), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8896)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 8960)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1120), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 9024)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1128), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 9088)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1136), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
-            kernel.shared_1[(threadIdx.x_2 + 9152)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1144), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*72)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[81]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[81]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*72)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*72)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*72)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*72)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*72)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*72)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[89]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[89]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[90]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[90]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[98]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[98]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[72]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[72]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[99]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[99]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[80]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[80]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[107]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[107]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[108]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[108]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[135]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[135]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[162]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[162]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[189]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[189]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[116]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[116]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[143]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[143]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[170]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[170]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[197]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[197]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[117]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[117]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[144]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[144]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[171]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[171]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[198]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[198]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[125]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[125]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[152]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[152]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[179]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[179]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[206]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[206]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[126]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[126]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[153]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[153]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[180]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[180]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[207]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[207]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-            conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-            conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-            conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-            conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-            conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[134]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[134]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[161]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[161]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[188]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[188]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[215]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-            conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[215]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
           }
         }
-        for (i3.inner: int32, 0, 7) {
-          compute[((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[i3.inner] + bias[((floordiv(blockIdx.x, 7)*128) + threadIdx.x)]), 0f32)
-          compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner) + 3136)] = max((conv2d_nchw_1[(i3.inner + 7)] + bias[(((floordiv(blockIdx.x, 7)*128) + threadIdx.x) + 64)]), 0f32)
+        for (i1.inner: int32, 0, 2) {
+          for (i3.inner: int32, 0, 7) {
+            compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*98)) + (i1.inner*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[((i1.inner*7) + i3.inner)] + bias[(((floordiv(blockIdx.x, 7)*128) + (threadIdx.x*2)) + i1.inner)]), 0f32)
+          }
         }
       }
     }
@@ -1607,7 +750,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 0.255 ms
+    Execution time of this operator: 0.371 ms
 
 
 
@@ -1652,9 +795,9 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
     conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
     conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
-    conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
+    conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
     conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
-    conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
+    conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
     conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
     conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
     conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
@@ -1663,19 +806,19 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
     conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
     conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-    conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
-    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+    conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
     conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-    conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
-    conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
-    conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+    conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+    conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
+    conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
     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_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
     compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
-    compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
+    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=1)
     compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
@@ -1703,11 +846,11 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
     s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
     pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
-    pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
+    pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=4)
     s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
     pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
     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", 1024)
+    s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
     s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
 
     CUDA source code:
@@ -1727,1187 +870,428 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     #endif
     extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
       float conv2d_nchw[14];
-      __shared__ float pad_temp_shared[216];
-      __shared__ float kernel_shared[9216];
+      __shared__ float pad_temp_shared[72];
+      __shared__ float kernel_shared[3072];
       conv2d_nchw[0] = 0.000000e+00f;
-      conv2d_nchw[7] = 0.000000e+00f;
       conv2d_nchw[1] = 0.000000e+00f;
-      conv2d_nchw[8] = 0.000000e+00f;
       conv2d_nchw[2] = 0.000000e+00f;
-      conv2d_nchw[9] = 0.000000e+00f;
       conv2d_nchw[3] = 0.000000e+00f;
-      conv2d_nchw[10] = 0.000000e+00f;
       conv2d_nchw[4] = 0.000000e+00f;
-      conv2d_nchw[11] = 0.000000e+00f;
       conv2d_nchw[5] = 0.000000e+00f;
-      conv2d_nchw[12] = 0.000000e+00f;
       conv2d_nchw[6] = 0.000000e+00f;
+      conv2d_nchw[7] = 0.000000e+00f;
+      conv2d_nchw[8] = 0.000000e+00f;
+      conv2d_nchw[9] = 0.000000e+00f;
+      conv2d_nchw[10] = 0.000000e+00f;
+      conv2d_nchw[11] = 0.000000e+00f;
+      conv2d_nchw[12] = 0.000000e+00f;
       conv2d_nchw[13] = 0.000000e+00f;
       for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 27) / 9) + (((int)blockIdx.x) % 7))) && ((((((int)threadIdx.x) % 27) / 9) + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
-        pad_temp_shared[(((int)threadIdx.x) + 64)] = (((((1 <= ((((((int)threadIdx.x) + 10) % 27) / 9) + (((int)blockIdx.x) % 7))) && (((((((int)threadIdx.x) + 10) % 27) / 9) + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 64) / 27) * 49)) + ((((((int)threadIdx.x) + 10) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] :  [...]
-        pad_temp_shared[(((int)threadIdx.x) + 128)] = (((((1 <= ((((((int)threadIdx.x) + 20) % 27) / 9) + (((int)blockIdx.x) % 7))) && (((((((int)threadIdx.x) + 20) % 27) / 9) + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 128) / 27) * 49)) + ((((((int)threadIdx.x) + 20) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)]  [...]
-        if (((int)threadIdx.x) < 24) {
-          pad_temp_shared[(((int)threadIdx.x) + 192)] = (((((1 <= ((((((int)threadIdx.x) + 3) % 27) / 9) + (((int)blockIdx.x) % 7))) && (((((((int)threadIdx.x) + 3) % 27) / 9) + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 192) / 27) * 49)) + ((((((int)threadIdx.x) + 3) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : [...]
+        for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
+          __syncthreads();
+          if (((int)threadIdx.x) < 18) {
+            pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= ((((int)threadIdx.x) * 4) % 9))) && (((((int)threadIdx.x) * 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
+          }
+          if (((int)threadIdx.x) < 18) {
+            pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 1) % 9))) && ((((((int)threadIdx.x) * 4) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
+          }
+          if (((int)threadIdx.x) < 18) {
+            pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 2) % 9))) && ((((((int)threadIdx.x) * 4) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
+          }
+          if (((int)threadIdx.x) < 18) {
+            pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 <= (ry_outer_outer + (((int)blockIdx.x) % 7))) && ((ry_outer_outer + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 4) + 3) % 9))) && ((((((int)threadIdx.x) * 4) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
+          }
+          kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 64)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
+          kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
+          kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
+          kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
+          kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
+          kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
+          kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
+          kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
+          kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
+          kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
+          kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
+          kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
+          kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
+          kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
+          kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
+          kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          __syncthreads();
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
         }
-        kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x))];
-        kernel_shared[(((int)threadIdx.x) + 64)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 192)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 192) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 384)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 384) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 576)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 36864)];
-        kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 768)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 768) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 960)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 960) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 73728)];
-        kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1344) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1536) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 110592)];
-        kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1920) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2112) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 147456)];
-        kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2496) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2688) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 184320)];
-        kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3072)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3072) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3136) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3200)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3200) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3264)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3264) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3328)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3328) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3392)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3392) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 3456)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 221184)];
-        kernel_shared[(((int)threadIdx.x) + 3520)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3520) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3584) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3648)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3648) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3712)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3712) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3776)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3776) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3840)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3840) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3904)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3904) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 3968)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3968) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 258048)];
-        kernel_shared[(((int)threadIdx.x) + 4096)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4096) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4160)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4160) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4224)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4224) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4288)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4288) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4352)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4352) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4416)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4416) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4480) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4544)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4544) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 4608)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 294912)];
-        kernel_shared[(((int)threadIdx.x) + 4672)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4672) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4736)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4736) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4800)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4800) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4864)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4864) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4928)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4928) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 4992)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4992) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5056)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5056) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5120)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5120) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 5184)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 331776)];
-        kernel_shared[(((int)threadIdx.x) + 5248)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5248) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5312)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5312) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5376)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5376) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5440)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5440) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5504)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5504) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5568)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5568) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5632)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5632) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5696)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5696) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 5760)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 368640)];
-        kernel_shared[(((int)threadIdx.x) + 5824)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5824) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5888)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5888) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 5952)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5952) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6016)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6016) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6080)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6080) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6144)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6144) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6208)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6208) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6272)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6272) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 6336)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 405504)];
-        kernel_shared[(((int)threadIdx.x) + 6400)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6400) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6464)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6464) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6528)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6528) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6592)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6592) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6656)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6656) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6720)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6720) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6784)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6784) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 6848)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6848) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 6912)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 442368)];
-        kernel_shared[(((int)threadIdx.x) + 6976)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6976) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7040)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7040) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7104)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7104) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7168)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7168) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7232)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7232) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7296)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7296) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7360)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7360) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7424)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7424) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 7488)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 479232)];
-        kernel_shared[(((int)threadIdx.x) + 7552)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7552) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7616)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7616) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7680)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7680) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7744)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7744) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7808)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7808) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7872)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7872) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 7936)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7936) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8000)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8000) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 8064)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 516096)];
-        kernel_shared[(((int)threadIdx.x) + 8128)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8128) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8192)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8192) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8256)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8256) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8320)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8320) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8384)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8384) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8448)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8448) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8512)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8512) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8576)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8576) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        kernel_shared[(((int)threadIdx.x) + 8640)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 552960)];
-        kernel_shared[(((int)threadIdx.x) + 8704)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8704) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8768)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8768) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8832)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8832) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8896)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8896) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 8960)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8960) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 9024)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 9024) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 9088)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 9088) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-        kernel_shared[(((int)threadIdx.x) + 9152)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 9152) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-        __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 72)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 72)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 72)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 72)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 72)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 72)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 72)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-        conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-        conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-        conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-        conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-        conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-        conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
       }
-      for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
-        compute[(((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[i3_inner] + bias[(((((int)blockIdx.x) / 7) * 128) + ((int)threadIdx.x))]), 0.000000e+00f);
-        compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner) + 3136)] = max((conv2d_nchw[(i3_inner + 7)] + bias[((((((int)blockIdx.x) / 7) * 128) + ((int)threadIdx.x)) + 64)]), 0.000000e+00f);
+      for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+        for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
+          compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
+        }
       }
     }
 
@@ -2966,7 +1350,7 @@ In the example below we resume the status and do more 5 trials.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  24.018 seconds)
+   **Total running time of the script:** ( 2 minutes  29.882 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 45523f3..fb99aca 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
@@ -614,7 +614,7 @@ 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)  
-       9.8818       9.8835       9.8960       9.8659       0.0123   
+      10.0844      10.0845      10.1364      10.0324       0.0425   
                
 
 
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 fbffd65..8e26ba2 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
@@ -633,7 +633,7 @@ 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)  
-      764.9774     762.1082     772.5468     760.2772      5.4043   
+      765.4736     767.2281     767.7041     761.4887      2.8245   
                
 
 
@@ -658,7 +658,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  20.658 seconds)
+   **Total running time of the script:** ( 1 minutes  19.112 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 ecf5430..09da688 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
@@ -362,23 +362,27 @@ 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} {
-      for (i0.outer.i1.outer.fused: int32, 0, 64) "parallel" {
-        allocate(compute_3: Pointer(global float32), float32, [1024]), storage_scope = global {
-          for (i.outer.inner: int32, 0, 64) {
-            for (j.init: int32, 0, 16) {
-              compute_4: Buffer(compute_3, float32, [1024], [])[((i.outer.inner*16) + j.init)] = 0f32
+      for (i0.outer.i1.outer.fused: int32, 0, 128) "parallel" {
+        allocate(compute_3: Pointer(global float32), float32, [512]), storage_scope = global {
+          for (nb_j.inner: int32, 0, 2) {
+            for (i.inner.init: int32, 0, 16) {
+              for (j.init: int32, 0, 16) {
+                compute_4: Buffer(compute_3, float32, [512], [])[(((i.inner.init*32) + (nb_j.inner*16)) + j.init)] = 0f32
+              }
             }
-            for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
-              for (j: int32, 0, 16) {
-                let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
-                let cse_var_2: int32 = ((i.outer.inner*16) + j)
-                compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i.outer.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+            for (elem_idx: int32, 0, let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+              for (i.inner: int32, 0, 16) {
+                for (j: int32, 0, 16) {
+                  let cse_var_3: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
+                  let cse_var_2: int32 = (((i.inner*32) + (nb_j.inner*16)) + j)
+                  compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 16)*4096) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+                }
               }
             }
           }
-          for (i0.inner: int32, 0, 64) {
-            let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*32768) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
-            compute[ramp(cse_var_4, 1, 16)] = max((compute_4[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_4, 1, 16)]), broadcast(0f32, 16))
+          for (i0.inner: int32, 0, 16) {
+            let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
+            compute[ramp(cse_var_4, 1, 32)] = max((compute_4[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
           }
         }
       }
@@ -432,7 +436,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 2.262 ms
+    Execution time of this operator: 1.491 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 424fadd..5dc2682 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
@@ -5,10 +5,10 @@
 
 Computation times
 =================
-**00:43.629** total execution time for **how_to_tune_with_autotvm** files:
+**00:44.335** total execution time for **how_to_tune_with_autotvm** files:
 
-- **00:42.830**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
-- **00:00.210**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
-- **00:00.199**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
-- **00:00.196**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
-- **00:00.194**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)
+- **00:43.478**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
+- **00:00.224**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
+- **00:00.215**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
+- **00:00.212**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
+- **00:00.207**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)
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 3771bbd..d89a887 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
@@ -859,8 +859,8 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2885496
-    No: 6   GFLOPS: 103.84/103.84   result: MeasureResult(costs=(0.0022294135833333334,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5974090099334717, timestamp=1648924642.0972478)      [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
-    No: 7   GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 6   GFLOPS: 109.34/109.34   result: MeasureResult(costs=(0.0021172386666666663,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6623878479003906, timestamp=1648932944.1037817)      [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
+    No: 7   GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -983,7 +983,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6225319
-    No: 8   GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 8   GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1106,7 +1106,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,943546
-    No: 9   GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 9   GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1229,7 +1229,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2868708
-    No: 10  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 10  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
         res = future.result()
       File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
@@ -1247,7 +1247,7 @@ for this template
     TimeoutError
 
             [('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4691833
-    No: 11  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 11  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1370,7 +1370,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1042124
-    No: 12  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 12  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1493,7 +1493,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, 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, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10013405
-    No: 13  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 13  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1616,7 +1616,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, 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, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6732082
-    No: 14  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 14  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1739,7 +1739,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7536735
-    No: 15  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 15  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1862,7 +1862,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,482121
-    No: 16  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 16  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1985,7 +1985,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2824525
-    No: 17  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 17  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2108,7 +2108,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559286
-    No: 18  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 18  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2231,7 +2231,7 @@ for this template
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
     tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9677544
-    No: 19  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+    No: 19  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 721, in __call__
         yield remote, remote.load_module(os.path.split(build_result.filename)[1])
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 685, in run_through_rpc
@@ -2319,7 +2319,7 @@ for this template
       15: _PyEval_EvalFrameDefault
       14: 0x0000000000537c30
       13: _PyObject_FastCallKeywords
-      12: 0x00007f4b0f68efa2
+      12: 0x00007f4585a94fa2
       11: _ctypes_callproc
       10: ffi_call
       9: ffi_call_unix64
@@ -2384,7 +2384,7 @@ for this template
       21: _PyFunction_FastCallKeywords
       20: _PyEval_EvalFrameDefault
       19: _PyFunction_FastCall      [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6390073
-    No: 20  GFLOPS: 144.80/144.80   result: MeasureResult(costs=(0.0015987917100000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3352305889129639, timestamp=1648924667.7228255)      [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
+    No: 20  GFLOPS: 141.51/141.51   result: MeasureResult(costs=(0.00163588642,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.414628505706787, timestamp=1648932970.4464226)       [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
 
 
 
@@ -2437,7 +2437,7 @@ and measure running time.
 
     Best config:
     [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
-    Time cost of this operator: 0.001954
+    Time cost of this operator: 0.002028
 
 
 
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 86cf6bc..f80e2ba 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
@@ -292,10 +292,10 @@ Timing the untuned program
     ########## Build without Autotuning ##########
     Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs  
     ---------                                     ---                                           --------  -------  -----              ------  -------  
-    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  313.2     98.712   (1, 2, 10, 10, 3)  2       1        
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.144     0.991    (1, 6, 10, 10)     1       1        
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.944     0.298    (1, 1, 10, 10, 3)  1       1        
-    Total_time                                    -                                             317.288   -        -                  -       -        
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  312.5     98.719   (1, 2, 10, 10, 3)  2       1        
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.153     0.996    (1, 6, 10, 10)     1       1        
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.901     0.285    (1, 1, 10, 10, 3)  1       1        
+    Total_time                                    -                                             316.554   -        -                  -       -        
 
 
 
@@ -357,10 +357,10 @@ Timing the tuned program
     ########## Build with Autotuning ##########
     Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs  
     ---------                                     ---                                           --------  -------  -----              ------  -------  
-    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  95.25     97.289   (1, 6, 10, 10, 1)  2       1        
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.736     1.773    (1, 6, 10, 10)     1       1        
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.918     0.938    (1, 1, 10, 10, 3)  1       1        
-    Total_time                                    -                                             97.905    -        -                  -       -        
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  123.6     97.871   (1, 6, 10, 10, 1)  2       1        
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.741     1.379    (1, 6, 10, 10)     1       1        
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.947     0.75     (1, 1, 10, 10, 3)  1       1        
+    Total_time                                    -                                             126.289   -        -                  -       -        
 
 
 
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 1af6423..8acb44c 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
@@ -5,10 +5,10 @@
 
 Computation times
 =================
-**00:44.863** total execution time for **how_to_work_with_microtvm** files:
+**00:44.217** total execution time for **how_to_work_with_microtvm** files:
 
-- **00:40.719**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
-- **00:03.546**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
-- **00:00.200**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
-- **00:00.200**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
-- **00:00.197**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)
+- **00:40.189**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
+- **00:03.475**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
+- **00:00.190**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)
+- **00:00.183**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
+- **00:00.179**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
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 44068c9..0236f9a 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
@@ -5,8 +5,8 @@
 
 Computation times
 =================
-**00:08.912** total execution time for **how_to_work_with_relay** files:
+**00:10.469** total execution time for **how_to_work_with_relay** files:
 
-- **00:06.761**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
-- **00:01.947**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
-- **00:00.205**: :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)
+- **00:08.368**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
+- **00:01.899**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
+- **00:00.202**: :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)
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 172be59..47af497 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
@@ -5,13 +5,13 @@
 
 Computation times
 =================
-**00:05.449** total execution time for **how_to_work_with_schedules** files:
+**00:05.564** total execution time for **how_to_work_with_schedules** files:
 
-- **00:02.002**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
-- **00:01.130**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
-- **00:00.699**: :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)
-- **00:00.681**: :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)
-- **00:00.295**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
-- **00:00.227**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
-- **00:00.211**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
-- **00:00.204**: :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)
+- **00:02.021**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
+- **00:01.203**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
+- **00:00.707**: :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)
+- **00:00.698**: :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)
+- **00:00.284**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
+- **00:00.223**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
+- **00:00.216**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
+- **00:00.211**: :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)
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 5f7c342..d0664be 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -314,8 +314,8 @@ The importing needs to happen before the tensorized GEMV being executed.
                  B: Buffer(B_2: Pointer(float32), float32, [32768], []),
                  C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
       buffer_map = {A_1: A, B_1: B, C_1: C} {
-      attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpjiuprxtx/input0.cc'
-    source_filename = "/tmp/tmpjiuprxtx/input0.cc"
+      attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp5a5wc43t/input0.cc'
+    source_filename = "/tmp/tmp5a5wc43t/input0.cc"
     target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
     target triple = "x86_64-pc-linux-gnu"
 
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 0af3300..586cb90 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
@@ -5,7 +5,7 @@
 
 Computation times
 =================
-**00:20.787** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:19.478** total execution time for **topic_vta_tutorials_autotvm** files:
 
-- **00:20.575**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
-- **00:00.212**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)
+- **00:19.295**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
+- **00:00.184**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)
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 f08ef1c..67a1b00 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -265,7 +265,7 @@ 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 21.52s!
+    resnet18_v1 inference graph built in 20.57s!
 
 
 
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 2011880..99daee9 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -301,7 +301,7 @@ The compilation steps are:
 
     /workspace/python/tvm/relay/build_module.py:439: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
       DeprecationWarning,
-    yolov3-tiny inference graph built in 14.76s!
+    yolov3-tiny inference graph built in 14.24s!
 
 
 
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 2de67ca..4e2202c 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
@@ -5,7 +5,7 @@
 
 Computation times
 =================
-**01:28.715** total execution time for **topic_vta_tutorials_frontend** files:
+**01:27.350** total execution time for **topic_vta_tutorials_frontend** files:
 
-- **00:46.968**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
-- **00:41.747**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``)
+- **00:46.114**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
+- **00:41.236**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``)
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 c6d902c..f3ff26d 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
@@ -5,7 +5,7 @@
 
 Computation times
 =================
-**00:03.534** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.539** total execution time for **topic_vta_tutorials_optimize** files:
 
-- **00:02.992**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
-- **00:00.543**: :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``)
+- **00:03.015**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
+- **00:00.524**: :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``)
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 91a3ba7..7ef1d02 100644
--- a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
 
 Computation times
 =================
-**00:00.989** total execution time for **topic_vta_tutorials** files:
+**00:00.950** total execution time for **topic_vta_tutorials** files:
 
-- **00:00.503**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
-- **00:00.485**: :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``)
+- **00:00.481**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
+- **00:00.468**: :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``)
diff --git a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
index e624d70..c059102 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -184,7 +184,7 @@ trials, we can load the best schedule from the log file and apply it.
 
  .. code-block:: none
 
-
+    *E*E
 
 
 
@@ -305,7 +305,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 93.320 ms
+    Execution time of this operator: 93.315 ms
 
 
 
@@ -414,6 +414,11 @@ Expression (TE) language that demonstrates how TVM can optimize computational
 operations.
 
 
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** ( 1 minutes  15.749 seconds)
+
+
 .. _sphx_glr_download_tutorial_auto_scheduler_matmul_x86.py:
 
 
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 916254e..0dbb6d5 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -268,7 +268,7 @@ standard deviation.
 
  .. code-block:: none
 
-    {'mean': 494.12892748000104, 'median': 493.87077314999885, 'std': 1.1549446178682683}
+    {'mean': 492.8096377599991, 'median': 492.41929994999794, 'std': 1.6378357186589703}
 
 
 
@@ -482,30 +482,31 @@ the tuning data to.
 
  .. code-block:: none
 
-
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  1/25]  Current/Best:   22.88/  22.88 GFLOPS | Progress: (4/10) | 6.02 s
    [Task  1/25]  Current/Best:   15.62/  24.00 GFLOPS | Progress: (8/10) | 10.13 s
    [Task  1/25]  Current/Best:   15.48/  24.00 GFLOPS | Progress: (10/10) | 11.93 s Done.
-
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  2/25]  Current/Best:   23.08/  23.08 GFLOPS | Progress: (4/10) | 2.34 s
    [Task  2/25]  Current/Best:   12.55/  23.08 GFLOPS | Progress: (8/10) | 3.70 s
    [Task  2/25]  Current/Best:    5.20/  23.08 GFLOPS | Progress: (10/10) | 4.69 s Done.
-
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  3/25]  Current/Best:    9.64/  15.37 GFLOPS | Progress: (4/10) | 3.50 s
    [Task  3/25]  Current/Best:    1.61/  23.46 GFLOPS | Progress: (8/10) | 6.66 s
    [Task  3/25]  Current/Best:   17.13/  23.46 GFLOPS | Progress: (10/10) | 7.71 s Done.
-
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  4/25]  Current/Best:   14.37/  14.37 GFLOPS | Progress: (4/10) | 2.38 s
    [Task  4/25]  Current/Best:    9.57/  16.89 GFLOPS | Progress: (8/10) | 3.86 s
    [Task  4/25]  Current/Best:   12.42/  16.89 GFLOPS | Progress: (10/10) | 4.53 s Done.
-
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  5/25]  Current/Best:    3.15/  14.03 GFLOPS | Progress: (4/10) | 4.51 s
    [Task  5/25]  Current/Best:   20.49/  20.49 GFLOPS | Progress: (8/10) | 7.32 s
    [Task  5/25]  Current/Best:   19.22/  20.49 GFLOPS | Progress: (10/10) | 7.85 s Done.
-
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  6/25]  Current/Best:    6.43/  15.14 GFLOPS | Progress: (4/10) | 4.47 s
    [Task  6/25]  Current/Best:   14.19/  18.03 GFLOPS | Progress: (8/10) | 7.02 s
    [Task  6/25]  Current/Best:    8.87/  18.03 GFLOPS | Progress: (10/10) | 7.99 s Done.
-
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  7/25]  Current/Best:   15.15/  15.15 GFLOPS | Progress: (4/10) | 3.55 s
    [Task  7/25]  Current/Best:   15.81/  23.56 GFLOPS | Progress: (8/10) | 5.07 s
    [Task  7/25]  Current/Best:   17.11/  23.56 GFLOPS | Progress: (10/10) | 5.90 s Done.
-
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  8/25]  Current/Best:   18.47/  18.47 GFLOPS | Progress: (4/10) | 3.70 s
    [Task  8/25]  Current/Best:   17.76/  18.47 GFLOPS | Progress: (8/10) | 6.04 s
    [Task  8/25]  Current/Best:   13.13/  18.47 GFLOPS | Progress: (10/10) | 8.00 s Done.
-
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  9/25]  Current/Best:   19.67/  19.67 GFLOPS | Progress: (4/10) | 2.39 s
    [Task  9/25]  Current/Best:    7.17/  19.67 GFLOPS | Progress: (8/10) | 4.12 s
    [Task  9/25]  Current/Best:   18.16/  19.67 GFLOPS | Progress: (10/10) | 4.77 s Done.
-
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 10/25]  Current/Best:    6.16/  18.56 GFLOPS | Progress: (4/10) | 2.28 s
    [Task 10/25]  Current/Best:    4.18/  18.56 GFLOPS | Progress: (8/10) | 5.68 s
    [Task 10/25]  Current/Best:    4.85/  18.56 GFLOPS | Progress: (10/10) | 6.76 s Done.
-
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 11/25]  Current/Best:   18.63/  24.22 GFLOPS | Progress: (4/10) | 2.78 s
    [Task 11/25]  Current/Best:   15.28/  24.22 GFLOPS | Progress: (8/10) | 4.61 s
    [Task 11/25]  Current/Best:   11.86/  24.22 GFLOPS | Progress: (10/10) | 5.78 s Done.
-
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 12/25]  Current/Best:   22.02/  22.02 GFLOPS | Progress: (4/10) | 2.91 s
    [Task 12/25]  Current/Best:   15.48/  22.02 GFLOPS | Progress: (8/10) | 7.47 s
    [Task 12/25]  Current/Best:    6.20/  22.02 GFLOPS | Progress: (10/10) | 8.72 s Done.
-
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 13/25]  Current/Best:    3.11/  18.92 GFLOPS | Progress: (4/10) | 4.44 s
    [Task 13/25]  Current/Best:   13.79/  18.92 GFLOPS | Progress: (8/10) | 7.24 s
    [Task 13/25]  Current/Best:   19.58/  22.45 GFLOPS | Progress: (10/10) | 8.43 s Done.
-
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 14/25]  Current/Best:   17.55/  17.55 GFLOPS | Progress: (4/10) | 3.26 s
    [Task 14/25]  Current/Best:    2.98/  17.55 GFLOPS | Progress: (8/10) | 5.44 s
    [Task 14/25]  Current/Best:    6.31/  17.55 GFLOPS | Progress: (10/10) | 7.38 s
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 15/25]  Current/Best:   18.32/  20.95 GFLOPS | Progress: (4/10) | 3.11 s Done.
-
    [Task 15/25]  Current/Best:    6.41/  20.95 GFLOPS | Progress: (8/10) | 5.41 s
    [Task 15/25]  Current/Best:   23.27/  23.27 GFLOPS | Progress: (10/10) | 6.25 s
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 16/25]  Current/Best:   21.47/  21.47 GFLOPS | Progress: (4/10) | 4.32 s
    [Task 16/25]  Current/Best:   15.12/  21.47 GFLOPS | Progress: (8/10) | 7.00 s
    [Task 16/25]  Current/Best:   15.63/  21.47 GFLOPS | Progress: (10/10) | 7.57 s Done.
-
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 17/25]  Current/Best:   17.85/  17.85 GFLOPS | Progress: (4/10) | 3.88 s
    [Task 17/25]  Current/Best:   21.74/  21.74 GFLOPS | Progress: (8/10) | 5.80 s
    [Task 17/25]  Current/Best:    3.10/  22.32 GFLOPS | Progress: (10/10) | 7.21 s Done.
-
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 18/25]  Current/Best:   16.36/  19.54 GFLOPS | Progress: (4/10) | 3.73 s
    [Task 18/25]  Current/Best:   18.22/  20.55 GFLOPS | Progress: (8/10) | 5.97 s
    [Task 18/25]  Current/Best:   19.83/  20.55 GFLOPS | Progress: (10/10) | 7.04 s Done.
-
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 19/25]  Current/Best:    9.87/  15.77 GFLOPS | Progress: (4/10) | 6.53 s
    [Task 19/25]  Current/Best:    8.46/  15.77 GFLOPS | Progress: (8/10) | 10.08 s
    [Task 19/25]  Current/Best:   14.15/  15.77 GFLOPS | Progress: (10/10) | 11.91 s Done.
-
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 20/25]  Current/Best:   17.95/  17.95 GFLOPS | Progress: (4/10) | 2.40 s
    [Task 20/25]  Current/Best:   10.63/  17.95 GFLOPS | Progress: (8/10) | 6.61 s
    [Task 20/25]  Current/Best:    9.17/  17.95 GFLOPS | Progress: (10/10) | 9.36 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 21/25]  Current/Best:   16.35/  18.59 GFLOPS | Progress: (4/10) | 2.16 s
    [Task 21/25]  Current/Best:   15.09/  18.59 GFLOPS | Progress: (8/10) | 6.41 s
    [Task 21/25]  Current/Best:   15.52/  18.59 GFLOPS | Progress: (10/10) | 7.51 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 22/25]  Current/Best:   10.27/  18.01 GFLOPS | Progress: (4/10) | 2.71 s Done.
+
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  1/25]  Current/Best:   24.15/  24.15 GFLOPS | Progress: (4/10) | 8.53 s
    [Task  1/25]  Current/Best:   16.55/  24.15 GFLOPS | Progress: (8/10) | 10.70 s
    [Task  1/25]  Current/Best:    9.11/  24.15 GFLOPS | Progress: (10/10) | 12.27 s Done.
+
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  2/25]  Current/Best:    7.06/  17.07 GFLOPS | Progress: (4/10) | 2.38 s
    [Task  2/25]  Current/Best:    6.89/  19.34 GFLOPS | Progress: (8/10) | 4.28 s
    [Task  2/25]  Current/Best:   22.90/  22.90 GFLOPS | Progress: (10/10) | 4.70 s Done.
+
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  3/25]  Current/Best:   14.14/  22.71 GFLOPS | Progress: (4/10) | 3.24 s
    [Task  3/25]  Current/Best:   15.62/  22.71 GFLOPS | Progress: (8/10) | 5.09 s
    [Task  3/25]  Current/Best:   14.87/  22.71 GFLOPS | Progress: (10/10) | 5.85 s Done.
+
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  4/25]  Current/Best:    3.50/  18.91 GFLOPS | Progress: (4/10) | 4.20 s
    [Task  4/25]  Current/Best:   12.95/  18.91 GFLOPS | Progress: (8/10) | 5.70 s
    [Task  4/25]  Current/Best:   12.08/  18.91 GFLOPS | Progress: (10/10) | 8.16 s Done.
+
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  5/25]  Current/Best:    8.09/  20.29 GFLOPS | Progress: (4/10) | 3.33 s
    [Task  5/25]  Current/Best:    4.66/  20.29 GFLOPS | Progress: (8/10) | 5.28 s
    [Task  5/25]  Current/Best:   14.79/  20.29 GFLOPS | Progress: (10/10) | 6.44 s Done.
+
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  6/25]  Current/Best:   22.79/  22.79 GFLOPS | Progress: (4/10) | 4.18 s
    [Task  6/25]  Current/Best:    5.22/  22.79 GFLOPS | Progress: (8/10) | 6.63 s
    [Task  6/25]  Current/Best:   15.01/  22.79 GFLOPS | Progress: (10/10) | 8.85 s Done.
+
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  7/25]  Current/Best:   18.40/  20.67 GFLOPS | Progress: (4/10) | 2.92 s
    [Task  7/25]  Current/Best:    5.33/  20.67 GFLOPS | Progress: (8/10) | 5.48 s
    [Task  7/25]  Current/Best:    5.59/  20.67 GFLOPS | Progress: (10/10) | 6.59 s Done.
+
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  8/25]  Current/Best:   10.48/  13.28 GFLOPS | Progress: (4/10) | 8.02 s
    [Task  8/25]  Current/Best:   12.59/  17.02 GFLOPS | Progress: (8/10) | 10.59 s
    [Task  8/25]  Current/Best:   17.62/  17.62 GFLOPS | Progress: (10/10) | 11.55 s Done.
+
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task  9/25]  Current/Best:   13.78/  17.22 GFLOPS | Progress: (4/10) | 2.68 s
    [Task  9/25]  Current/Best:    6.23/  17.22 GFLOPS | Progress: (8/10) | 4.29 s
    [Task  9/25]  Current/Best:    6.04/  17.22 GFLOPS | Progress: (10/10) | 7.44 s Done.
+
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 10/25]  Current/Best:    6.05/  12.50 GFLOPS | Progress: (4/10) | 3.69 s
    [Task 10/25]  Current/Best:    9.56/  14.97 GFLOPS | Progress: (8/10) | 5.95 s
    [Task 10/25]  Current/Best:    9.91/  14.97 GFLOPS | Progress: (10/10) | 6.72 s Done.
+
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 11/25]  Current/Best:   19.69/  19.69 GFLOPS | Progress: (4/10) | 3.97 s
    [Task 11/25]  Current/Best:    5.49/  19.69 GFLOPS | Progress: (8/10) | 7.16 s
    [Task 11/25]  Current/Best:   14.58/  19.69 GFLOPS | Progress: (10/10) | 8.12 s Done.
+
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 12/25]  Current/Best:   11.13/  11.13 GFLOPS | Progress: (4/10) | 3.51 s
    [Task 12/25]  Current/Best:   17.42/  22.71 GFLOPS | Progress: (8/10) | 5.10 s
    [Task 12/25]  Current/Best:    7.49/  22.71 GFLOPS | Progress: (10/10) | 7.82 s Done.
+
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 13/25]  Current/Best:   11.28/  23.14 GFLOPS | Progress: (4/10) | 6.61 s
    [Task 13/25]  Current/Best:   11.13/  23.14 GFLOPS | Progress: (8/10) | 9.88 s
    [Task 13/25]  Current/Best:   16.80/  23.14 GFLOPS | Progress: (10/10) | 11.26 s Done.
+
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 14/25]  Current/Best:   19.56/  19.56 GFLOPS | Progress: (4/10) | 4.73 s
    [Task 14/25]  Current/Best:   13.83/  19.56 GFLOPS | Progress: (8/10) | 11.25 s
    [Task 14/25]  Current/Best:   13.62/  19.56 GFLOPS | Progress: (10/10) | 12.64 s
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 15/25]  Current/Best:   12.07/  21.24 GFLOPS | Progress: (4/10) | 6.00 s
    [Task 15/25]  Current/Best:    6.73/  21.24 GFLOPS | Progress: (8/10) | 7.40 s
    [Task 15/25]  Current/Best:    9.92/  21.24 GFLOPS | Progress: (10/10) | 11.23 s
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
      Done.
-
    [Task 22/25]  Current/Best:    2.50/  18.73 GFLOPS | Progress: (8/10) | 4.98 s
    [Task 22/25]  Current/Best:   15.01/  18.73 GFLOPS | Progress: (10/10) | 5.84 s Done.
-
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 23/25]  Current/Best:   14.23/  17.50 GFLOPS | Progress: (4/10) | 4.59 s
    [Task 23/25]  Current/Best:    9.75/  17.50 GFLOPS | Progress: (8/10) | 6.82 s
    [Task 23/25]  Current/Best:   13.47/  20.37 GFLOPS | Progress: (10/10) | 8.25 s Done.
-
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 24/25]  Current/Best:    8.93/   8.93 GFLOPS | Progress: (4/10) | 56.84 s
    [Task 24/25]  Current/Best:    2.04/   8.93 GFLOPS | Progress: (8/10) | 140.67 s
    [Task 24/25]  Current/Best:    4.05/   8.93 GFLOPS | Progress: (10/10) | 142.95 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 25/25]  Current/Best:    5.63/   9.14 GFLOPS | Progress: (4/10) | 2.50 s
    [Task 25/25]  Current/Best:    1.55/   9.14 GFLOPS | Progress: (8/10) | 12.46 s
    [Task 25/25]  Current/Best:    3.60/   9.81 GFLOPS | Progress: (10/10) | 13.58 s Done.
-
+
    [Task 16/25]  Current/Best:   16.96/  17.18 GFLOPS | Progress: (4/10) | 2.76 s
    [Task 16/25]  Current/Best:   16.05/  17.18 GFLOPS | Progress: (8/10) | 6.09 s
    [Task 16/25]  Current/Best:   10.11/  17.18 GFLOPS | Progress: (10/10) | 8.12 s Done.
+
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 17/25]  Current/Best:    3.11/  13.35 GFLOPS | Progress: (4/10) | 4.98 s
    [Task 17/25]  Current/Best:   15.05/  15.86 GFLOPS | Progress: (8/10) | 8.58 s
    [Task 17/25]  Current/Best:   11.26/  15.86 GFLOPS | Progress: (10/10) | 11.29 s Done.
+
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 18/25]  Current/Best:   16.78/  16.78 GFLOPS | Progress: (4/10) | 8.62 s
    [Task 18/25]  Current/Best:   16.87/  18.75 GFLOPS | Progress: (8/10) | 10.04 s
    [Task 18/25]  Current/Best:   14.95/  18.75 GFLOPS | Progress: (10/10) | 11.02 s Done.
+
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 19/25]  Current/Best:    7.10/  12.47 GFLOPS | Progress: (4/10) | 5.94 s
    [Task 19/25]  Current/Best:   11.02/  18.94 GFLOPS | Progress: (8/10) | 7.97 s
    [Task 19/25]  Current/Best:   11.91/  18.94 GFLOPS | Progress: (10/10) | 9.18 s Done.
+
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 20/25]  Current/Best:   11.53/  21.10 GFLOPS | Progress: (4/10) | 3.06 s
    [Task 20/25]  Current/Best:   17.21/  21.10 GFLOPS | Progress: (8/10) | 4.97 s
    [Task 20/25]  Current/Best:    2.55/  21.10 GFLOPS | Progress: (10/10) | 6.61 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 21/25]  Current/Best:   10.00/  18.19 GFLOPS | Progress: (4/10) | 3.36 s
    [Task 21/25]  Current/Best:   17.95/  19.59 GFLOPS | Progress: (8/10) | 4.71 s
    [Task 21/25]  Current/Best:   19.94/  19.94 GFLOPS | Progress: (10/10) | 5.25 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 22/25]  Current/Best:    5.19/  18.09 GFLOPS | Progress: (4/10) | 3.14 s
    [Task 22/25]  Current/Best:    5.86/  18.09 GFLOPS | Progress: (8/10) | 6.08 s
    [Task 22/25]  Current/Best:    9.61/  19.77 GFLOPS | Progress: (10/10) | 6.77
  s Done.
+
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 23/25]  Current/Best:   12.01/  21.60 GFLOPS | Progress: (4/10) | 3.70 s
    [Task 23/25]  Current/Best:   10.64/  21.60 GFLOPS | Progress: (8/10) | 8.78 s
    [Task 23/25]  Current/Best:    5.28/  21.60 GFLOPS | Progress: (10/10) | 10.30 s Done.
+
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s
    [Task 24/25]  Current/Best:    3.28/   8.97 GFLOPS | Progress: (4/10) | 11.67 s
    [Task 24/25]  Current/Best:    8.68/  10.37 GFLOPS | Progress: (8/10) | 13.25 s
    [Task 24/25]  Current/Best:    6.13/  10.37 GFLOPS | Progress: (10/10) | 13.86 s Done.
+
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
+     Done.
+
    [Task 25/25]  Current/Best:    4.69/   7.68 GFLOPS | Progress: (4/10) | 29.19 s
    [Task 25/25]  Current/Best:    1.55/   7.68 GFLOPS | Progress: (8/10) | 31.97 s
    [Task 25/25]  Current/Best:    1.55/   9.00 GFLOPS | Progress: (10/10) | 32.71 s
 
 
 The output from this tuning process will look something like this:
@@ -647,8 +648,8 @@ improvement in comparing the optimized model to the unoptimized model.
 
  .. code-block:: none
 
-    optimized: {'mean': 428.284769710001, 'median': 428.43152789999976, 'std': 0.7073097362466341}
-    unoptimized: {'mean': 494.12892748000104, 'median': 493.87077314999885, 'std': 1.1549446178682683}
+    optimized: {'mean': 476.3876658900005, 'median': 476.6023151500008, 'std': 1.0064242546277575}
+    unoptimized: {'mean': 492.8096377599991, 'median': 492.41929994999794, 'std': 1.6378357186589703}
 
 
 
@@ -668,7 +669,7 @@ profiling/benchmarking.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 8 minutes  35.861 seconds)
+   **Total running time of the script:** ( 7 minutes  17.836 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 07e6d55..ad204a3 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -235,7 +235,7 @@ device and returns the measured cost. Network overhead is excluded.
 
  .. code-block:: none
 
-    1.271e-07 secs/op
+    1.269e-07 secs/op
 
 
 
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index a4f97ef..fe739f4 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -230,7 +230,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
 
  .. code-block:: none
 
-    [stage(a, placeholder(a, 0x19ceaae0)), stage(b, placeholder(b, 0x203edc20)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(mi [...]
+    [stage(a, placeholder(a, 0x52588a0)), stage(b, placeholder(b, 0x26161ca0)), 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 a18af3e..de74f6b 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,17 +5,17 @@
 
 Computation times
 =================
-**11:28.668** total execution time for **tutorial** files:
+**10:20.438** total execution time for **tutorial** files:
 
-- **08:35.861**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
-- **00:59.529**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
-- **00:56.483**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
-- **00:28.998**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
-- **00:25.613**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
-- **00:01.106**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
-- **00:00.712**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
-- **00:00.198**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
-- **00:00.049**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
-- **00:00.047**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
-- **00:00.038**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
-- **00:00.032**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
+- **07:17.836**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
+- **01:15.749**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
+- **01:01.535**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
+- **00:25.766**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
+- **00:17.329**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
+- **00:01.218**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
+- **00:00.687**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
+- **00:00.195**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
+- **00:00.032**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
+- **00:00.031**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
+- **00:00.030**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
+- **00:00.029**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index c1314e6..4cdd1e4 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -243,7 +243,7 @@ helper function to run a profile of the TVM generated code.
 
  .. code-block:: none
 
-    Numpy running time: 0.000007
+    Numpy running time: 0.000012
     naive: 0.000006
 
 
@@ -334,7 +334,7 @@ compile and run this new schedule with the parallel operation applied:
 
  .. code-block:: none
 
-    parallel: 0.000006
+    parallel: 0.000013
 
 
 
@@ -387,7 +387,7 @@ factor to be the number of threads on your CPU.
 
  .. code-block:: none
 
-    vector: 0.000028
+    vector: 0.000025
     @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, [(stride: int32*n: int32)], [], type="auto"),
@@ -436,10 +436,10 @@ We can now compare the different schedules
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                   numpy    6.715190000932125e-06                    1.0
-                   naive              5.8025e-06      0.8640857517351801
-                parallel              5.9872e-06      0.8915905579989436
-                  vector    2.7863900000000003e-05     4.149383710085978
+                   numpy    1.1630689999719835e-05                   1.0
+                   naive              5.8541e-06       0.503332132499535
+                parallel             1.30861e-05      1.1251353101419799
+                  vector             2.45878e-05       2.114044824562625
 
 
 
@@ -828,7 +828,7 @@ matrix multiplication.
 
  .. code-block:: none
 
-    Numpy running time: 0.018637
+    Numpy running time: 0.017957
 
 
 
@@ -884,7 +884,7 @@ optimizations.
 
  .. code-block:: none
 
-    none: 3.312094
+    none: 3.467262
 
 
 
@@ -982,7 +982,7 @@ schedule.
 
  .. code-block:: none
 
-    blocking: 0.295724
+    blocking: 0.299623
 
 
 
@@ -1073,7 +1073,7 @@ already cache friendly from our previous optimizations.
 
  .. code-block:: none
 
-    vectorization: 0.333825
+    vectorization: 0.334804
     @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], []),
@@ -1144,7 +1144,7 @@ more cache friendly.
 
  .. code-block:: none
 
-    loop permutation: 0.117561
+    loop permutation: 0.113482
     @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], []),
@@ -1240,7 +1240,7 @@ optimized schedule.
 
  .. code-block:: none
 
-    array packing: 0.110131
+    array packing: 0.108217
     @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], []),
@@ -1330,7 +1330,7 @@ to `C` when all the block results are ready.
 
  .. code-block:: none
 
-    block caching: 0.111748
+    block caching: 0.111193
     @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], []),
@@ -1413,7 +1413,7 @@ of thread-level parallelization.
 
  .. code-block:: none
 
-    parallelization: 0.144636
+    parallelization: 0.144173
     @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], []),
@@ -1491,13 +1491,13 @@ working, we can compare the results.
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                    none            3.3120943476                     1.0
-                blocking             0.295724329     0.08928620321890494
-           vectorization     0.33382496009999996     0.10078968926168883
-        loop permutation            0.1175610745    0.035494482391537775
-           array packing            0.1101311084     0.03325119904262476
-           block caching            0.1117476551    0.033739272910801664
-         parallelization            0.1446355741    0.043668917283351365
+                    none      3.4672621562000003                     1.0
+                blocking            0.2996231929     0.08641492318780322
+           vectorization            0.3348042092       0.096561550329074
+        loop permutation            0.1134821842     0.03272962328420317
+           array packing     0.10821661600000002    0.031210970248238077
+           block caching     0.11119294939999999     0.03206938050564473
+         parallelization            0.1441725125    0.041581082134847305
 
 
 
@@ -1532,6 +1532,11 @@ operations with tunable parameters that allows you to automatically optimize
 the computation for specific platforms.
 
 
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** ( 1 minutes  1.535 seconds)
+
+
 .. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
 
 
diff --git a/docs/commit_hash b/docs/commit_hash
index d3335f5..1907b17 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-264ee08aef985eddebc633bae356f3931060b6d9
+daa068979a464bd96f6f48e244a8917c4b353503
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index f0e7dc8..387efca 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -400,7 +400,7 @@
 </div>
 <img alt="../../_images/sphx_glr_from_mxnet_001.png" class="sphx-glr-single-img" src="../../_images/sphx_glr_from_mxnet_001.png" />
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipb5406133-5280-46f8-94d4-fac0f9dec188 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipcb8bcc43-51ed-43ad-a91c-4e6a7e0c1b40 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_paddle.html b/docs/how_to/compile_models/from_paddle.html
index 102a47f..11b63f6 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -448,7 +448,7 @@ A quick solution is</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>TVM prediction top-1 id: 282, class name:  282: &#39;tiger cat&#39;,
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  4.026 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  3.921 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-paddle-py">
 <div class="sphx-glr-download docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/16269b77359771348d507395692524cf/from_paddle.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_paddle.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index 35487ea..0454553 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -386,21 +386,24 @@ 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]
-  6%|6         | 2.75M/44.7M [00:00&lt;00:01, 26.0MB/s]
- 14%|#4        | 6.30M/44.7M [00:00&lt;00:01, 32.3MB/s]
- 22%|##1       | 9.67M/44.7M [00:00&lt;00:01, 31.1MB/s]
- 30%|###       | 13.4M/44.7M [00:00&lt;00:00, 33.4MB/s]
- 37%|###7      | 16.6M/44.7M [00:00&lt;00:01, 25.0MB/s]
- 43%|####3     | 19.3M/44.7M [00:00&lt;00:01, 23.8MB/s]
- 50%|####9     | 22.1M/44.7M [00:00&lt;00:00, 24.9MB/s]
- 56%|#####6    | 25.2M/44.7M [00:00&lt;00:00, 26.0MB/s]
- 67%|######6   | 29.8M/44.7M [00:01&lt;00:00, 32.0MB/s]
- 74%|#######3  | 33.0M/44.7M [00:01&lt;00:00, 31.1MB/s]
- 81%|########  | 36.1M/44.7M [00:01&lt;00:00, 30.4MB/s]
- 88%|########7 | 39.1M/44.7M [00:01&lt;00:00, 30.5MB/s]
- 94%|#########4| 42.1M/44.7M [00:01&lt;00:00, 18.4MB/s]
-100%|#########9| 44.6M/44.7M [00:01&lt;00:00, 20.0MB/s]
-100%|##########| 44.7M/44.7M [00:01&lt;00:00, 25.3MB/s]
+  4%|3         | 1.62M/44.7M [00:00&lt;00:03, 15.0MB/s]
+  7%|7         | 3.31M/44.7M [00:00&lt;00:02, 16.0MB/s]
+ 12%|#1        | 5.19M/44.7M [00:00&lt;00:02, 17.5MB/s]
+ 16%|#5        | 7.06M/44.7M [00:00&lt;00:02, 17.9MB/s]
+ 20%|#9        | 8.78M/44.7M [00:00&lt;00:02, 16.6MB/s]
+ 27%|##6       | 12.0M/44.7M [00:00&lt;00:01, 22.0MB/s]
+ 32%|###1      | 14.1M/44.7M [00:00&lt;00:01, 21.1MB/s]
+ 36%|###6      | 16.2M/44.7M [00:00&lt;00:01, 18.8MB/s]
+ 40%|####      | 18.1M/44.7M [00:01&lt;00:01, 14.1MB/s]
+ 44%|####4     | 19.7M/44.7M [00:01&lt;00:01, 14.5MB/s]
+ 52%|#####1    | 23.1M/44.7M [00:01&lt;00:01, 19.7MB/s]
+ 58%|#####8    | 26.1M/44.7M [00:01&lt;00:00, 22.6MB/s]
+ 64%|######3   | 28.4M/44.7M [00:01&lt;00:00, 21.7MB/s]
+ 74%|#######3  | 32.9M/44.7M [00:01&lt;00:00, 27.8MB/s]
+ 81%|########  | 36.0M/44.7M [00:01&lt;00:00, 29.0MB/s]
+ 88%|########7 | 39.2M/44.7M [00:01&lt;00:00, 26.8MB/s]
+ 94%|#########3| 41.9M/44.7M [00:02&lt;00:00, 24.4MB/s]
+100%|##########| 44.7M/44.7M [00:02&lt;00:00, 21.8MB/s]
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/compile_models/from_tensorflow.html b/docs/how_to/compile_models/from_tensorflow.html
index 64ec46b..992a579 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -606,7 +606,7 @@ 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  3.686 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  3.065 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-tensorflow-py">
 <div class="sphx-glr-download docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/7f1d3d1b878694c201c614c807cdebc8/from_tensorflow.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_tensorflow.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/sg_execution_times.html b/docs/how_to/compile_models/sg_execution_times.html
index fedba1c..1a11852 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -300,17 +300,17 @@
             
   <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>04:46.638</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>04:46.054</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
 <ul class="simple">
-<li><p><strong>01:04.026</strong>: <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></li>
-<li><p><strong>01:03.686</strong>: <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></li>
-<li><p><strong>00:55.234</strong>: <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></li>
-<li><p><strong>00:25.504</strong>: <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></li>
-<li><p><strong>00:22.062</strong>: <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></li>
-<li><p><strong>00:20.919</strong>: <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></li>
-<li><p><strong>00:20.697</strong>: <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></li>
-<li><p><strong>00:12.018</strong>: <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></li>
-<li><p><strong>00:02.490</strong>: <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></li>
+<li><p><strong>01:03.921</strong>: <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></li>
+<li><p><strong>01:03.065</strong>: <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></li>
+<li><p><strong>00:55.658</strong>: <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></li>
+<li><p><strong>00:25.455</strong>: <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></li>
+<li><p><strong>00:21.090</strong>: <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></li>
+<li><p><strong>00:20.918</strong>: <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></li>
+<li><p><strong>00:20.577</strong>: <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></li>
+<li><p><strong>00:12.441</strong>: <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></li>
+<li><p><strong>00:02.928</strong>: <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></li>
 </ul>
 </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 21d4a32..c534f16 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -622,7 +622,7 @@ to the remote android device.</p>
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  15.8919      15.8639      16.1597      15.7815       0.1118
+  15.6399      15.6042      15.8487      15.5657       0.0846
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
index 7d87053..efca858 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -409,63 +409,65 @@ 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]
-  2%|2         | 3.56M/170M [00:00&lt;00:04, 36.9MB/s]
-  4%|4         | 7.09M/170M [00:00&lt;00:06, 28.4MB/s]
-  6%|5         | 9.91M/170M [00:00&lt;00:06, 25.7MB/s]
-  8%|7         | 13.0M/170M [00:00&lt;00:05, 27.9MB/s]
-  9%|9         | 15.8M/170M [00:00&lt;00:06, 26.5MB/s]
- 11%|#1        | 19.2M/170M [00:00&lt;00:05, 27.0MB/s]
- 13%|#2        | 21.8M/170M [00:00&lt;00:06, 25.8MB/s]
- 15%|#4        | 25.1M/170M [00:00&lt;00:05, 28.3MB/s]
- 16%|#6        | 27.9M/170M [00:01&lt;00:05, 25.6MB/s]
- 18%|#7        | 30.4M/170M [00:01&lt;00:08, 18.2MB/s]
- 19%|#9        | 32.4M/170M [00:01&lt;00:08, 17.5MB/s]
- 20%|##        | 34.3M/170M [00:01&lt;00:08, 17.7MB/s]
- 21%|##1       | 36.5M/170M [00:01&lt;00:07, 19.1MB/s]
- 23%|##2       | 38.5M/170M [00:01&lt;00:07, 17.9MB/s]
- 24%|##3       | 40.7M/170M [00:01&lt;00:07, 19.2MB/s]
- 26%|##5       | 43.8M/170M [00:02&lt;00:05, 22.5MB/s]
- 27%|##7       | 46.0M/170M [00:02&lt;00:06, 20.9MB/s]
- 29%|##8       | 48.8M/170M [00:02&lt;00:05, 22.7MB/s]
- 30%|###       | 51.6M/170M [00:02&lt;00:05, 24.6MB/s]
- 32%|###2      | 55.1M/170M [00:02&lt;00:04, 25.3MB/s]
- 34%|###3      | 57.5M/170M [00:02&lt;00:05, 22.9MB/s]
- 36%|###6      | 61.6M/170M [00:02&lt;00:04, 27.6MB/s]
- 38%|###8      | 65.4M/170M [00:02&lt;00:03, 29.8MB/s]
- 41%|####      | 68.9M/170M [00:02&lt;00:03, 31.7MB/s]
- 42%|####2     | 72.0M/170M [00:03&lt;00:03, 31.1MB/s]
- 44%|####4     | 75.1M/170M [00:03&lt;00:03, 28.8MB/s]
- 46%|####5     | 77.9M/170M [00:03&lt;00:03, 28.4MB/s]
- 47%|####7     | 80.6M/170M [00:03&lt;00:03, 26.1MB/s]
- 49%|####8     | 83.2M/170M [00:03&lt;00:03, 26.2MB/s]
- 50%|#####     | 85.7M/170M [00:03&lt;00:04, 18.9MB/s]
- 52%|#####2    | 88.6M/170M [00:03&lt;00:03, 21.4MB/s]
- 54%|#####3    | 90.9M/170M [00:04&lt;00:03, 21.4MB/s]
- 55%|#####4    | 93.2M/170M [00:04&lt;00:03, 21.4MB/s]
- 56%|#####6    | 95.4M/170M [00:04&lt;00:03, 20.5MB/s]
- 57%|#####7    | 97.4M/170M [00:04&lt;00:03, 20.5MB/s]
- 59%|#####9    | 101M/170M [00:04&lt;00:02, 24.4MB/s]
- 61%|######1   | 104M/170M [00:04&lt;00:02, 26.9MB/s]
- 63%|######3   | 107M/170M [00:04&lt;00:02, 28.8MB/s]
- 65%|######4   | 110M/170M [00:04&lt;00:02, 28.3MB/s]
- 67%|######6   | 113M/170M [00:04&lt;00:02, 26.5MB/s]
- 69%|######9   | 117M/170M [00:04&lt;00:01, 31.6MB/s]
- 71%|#######   | 120M/170M [00:05&lt;00:01, 29.7MB/s]
- 73%|#######2  | 124M/170M [00:05&lt;00:01, 31.0MB/s]
- 75%|#######4  | 127M/170M [00:05&lt;00:01, 30.7MB/s]
- 77%|#######6  | 130M/170M [00:05&lt;00:01, 32.2MB/s]
- 78%|#######8  | 133M/170M [00:05&lt;00:01, 29.6MB/s]
- 80%|########  | 137M/170M [00:05&lt;00:01, 30.8MB/s]
- 82%|########2 | 140M/170M [00:05&lt;00:00, 32.0MB/s]
- 84%|########4 | 143M/170M [00:05&lt;00:01, 27.2MB/s]
- 86%|########6 | 146M/170M [00:06&lt;00:00, 29.2MB/s]
- 89%|########8 | 151M/170M [00:06&lt;00:00, 33.7MB/s]
- 91%|######### | 154M/170M [00:06&lt;00:00, 35.1MB/s]
- 93%|#########2| 158M/170M [00:06&lt;00:00, 29.2MB/s]
- 95%|#########4| 161M/170M [00:06&lt;00:00, 29.9MB/s]
- 97%|#########7| 165M/170M [00:06&lt;00:00, 32.9MB/s]
- 99%|#########9| 168M/170M [00:06&lt;00:00, 32.2MB/s]
-100%|##########| 170M/170M [00:06&lt;00:00, 26.3MB/s]
+  1%|1         | 1.81M/170M [00:00&lt;00:09, 18.9MB/s]
+  2%|2         | 4.24M/170M [00:00&lt;00:07, 22.7MB/s]
+  4%|3         | 6.41M/170M [00:00&lt;00:07, 22.7MB/s]
+  5%|5         | 8.57M/170M [00:00&lt;00:07, 22.2MB/s]
+  7%|6         | 11.4M/170M [00:00&lt;00:06, 24.9MB/s]
+  8%|8         | 13.8M/170M [00:00&lt;00:07, 20.5MB/s]
+ 10%|#         | 17.2M/170M [00:00&lt;00:06, 24.7MB/s]
+ 12%|#1        | 19.7M/170M [00:00&lt;00:07, 20.9MB/s]
+ 13%|#3        | 22.7M/170M [00:01&lt;00:06, 23.2MB/s]
+ 15%|#5        | 25.6M/170M [00:01&lt;00:06, 25.0MB/s]
+ 17%|#6        | 28.1M/170M [00:01&lt;00:05, 24.9MB/s]
+ 18%|#8        | 30.6M/170M [00:01&lt;00:06, 23.1MB/s]
+ 19%|#9        | 32.9M/170M [00:01&lt;00:06, 22.4MB/s]
+ 21%|##        | 35.1M/170M [00:01&lt;00:06, 22.4MB/s]
+ 22%|##1       | 37.2M/170M [00:01&lt;00:06, 21.2MB/s]
+ 23%|##3       | 39.3M/170M [00:01&lt;00:06, 20.4MB/s]
+ 24%|##4       | 41.3M/170M [00:02&lt;00:08, 16.2MB/s]
+ 25%|##5       | 43.0M/170M [00:02&lt;00:08, 14.8MB/s]
+ 27%|##6       | 45.3M/170M [00:02&lt;00:07, 16.9MB/s]
+ 28%|##8       | 48.0M/170M [00:02&lt;00:06, 19.8MB/s]
+ 30%|##9       | 50.8M/170M [00:02&lt;00:05, 22.4MB/s]
+ 31%|###1      | 53.1M/170M [00:02&lt;00:05, 22.1MB/s]
+ 33%|###2      | 55.4M/170M [00:02&lt;00:05, 22.4MB/s]
+ 35%|###5      | 59.6M/170M [00:02&lt;00:04, 27.3MB/s]
+ 38%|###7      | 64.2M/170M [00:02&lt;00:03, 33.0MB/s]
+ 40%|####      | 68.4M/170M [00:03&lt;00:02, 36.1MB/s]
+ 42%|####2     | 71.9M/170M [00:03&lt;00:03, 33.8MB/s]
+ 45%|####4     | 76.1M/170M [00:03&lt;00:02, 36.1MB/s]
+ 47%|####6     | 79.6M/170M [00:03&lt;00:03, 30.6MB/s]
+ 50%|####9     | 84.1M/170M [00:03&lt;00:02, 34.4MB/s]
+ 52%|#####1    | 87.6M/170M [00:03&lt;00:02, 30.3MB/s]
+ 53%|#####3    | 90.7M/170M [00:04&lt;00:04, 17.1MB/s]
+ 55%|#####4    | 93.0M/170M [00:04&lt;00:05, 15.3MB/s]
+ 56%|#####6    | 95.5M/170M [00:04&lt;00:04, 16.8MB/s]
+ 57%|#####7    | 97.5M/170M [00:04&lt;00:04, 17.2MB/s]
+ 59%|#####8    | 99.5M/170M [00:04&lt;00:04, 17.3MB/s]
+ 60%|######    | 102M/170M [00:04&lt;00:03, 18.7MB/s]
+ 62%|######1   | 105M/170M [00:04&lt;00:03, 21.3MB/s]
+ 63%|######3   | 107M/170M [00:05&lt;00:03, 18.6MB/s]
+ 65%|######4   | 110M/170M [00:05&lt;00:03, 20.0MB/s]
+ 66%|######6   | 113M/170M [00:05&lt;00:02, 22.8MB/s]
+ 68%|######7   | 115M/170M [00:05&lt;00:02, 22.8MB/s]
+ 70%|#######   | 119M/170M [00:05&lt;00:01, 28.5MB/s]
+ 73%|#######2  | 123M/170M [00:05&lt;00:01, 31.9MB/s]
+ 74%|#######4  | 126M/170M [00:05&lt;00:01, 30.5MB/s]
+ 76%|#######6  | 129M/170M [00:05&lt;00:01, 28.9MB/s]
+ 78%|#######7  | 132M/170M [00:05&lt;00:01, 27.7MB/s]
+ 80%|#######9  | 135M/170M [00:06&lt;00:01, 28.1MB/s]
+ 82%|########2 | 139M/170M [00:06&lt;00:00, 32.3MB/s]
+ 84%|########3 | 142M/170M [00:06&lt;00:00, 31.5MB/s]
+ 86%|########5 | 146M/170M [00:06&lt;00:01, 24.9MB/s]
+ 88%|########7 | 149M/170M [00:06&lt;00:00, 27.7MB/s]
+ 90%|######### | 153M/170M [00:06&lt;00:00, 31.2MB/s]
+ 92%|#########2| 156M/170M [00:06&lt;00:00, 31.7MB/s]
+ 94%|#########3| 160M/170M [00:06&lt;00:00, 29.6MB/s]
+ 96%|#########5| 162M/170M [00:07&lt;00:00, 27.3MB/s]
+ 98%|#########7| 166M/170M [00:07&lt;00:00, 29.3MB/s]
+ 99%|#########9| 169M/170M [00:07&lt;00:00, 28.1MB/s]
+100%|##########| 170M/170M [00:07&lt;00:00, 24.4MB/s]
 /usr/local/lib/python3.7/dist-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)
 /usr/local/lib/python3.7/dist-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=&#39;floor&#39;).
@@ -558,7 +560,7 @@ torchvision rcnn models.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Get 9 valid boxes
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  7.507 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  5.573 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-object-detection-pytorch-py">
 <div class="sphx-glr-download 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 9b313a6..192c014 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -450,8 +450,10 @@ 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]
- 51%|#####1    | 6.95M/13.6M [00:00&lt;00:00, 72.9MB/s]
-100%|##########| 13.6M/13.6M [00:00&lt;00:00, 79.7MB/s]
+ 27%|##7       | 3.69M/13.6M [00:00&lt;00:00, 38.6MB/s]
+ 54%|#####4    | 7.38M/13.6M [00:00&lt;00:00, 37.8MB/s]
+ 81%|########1 | 11.0M/13.6M [00:00&lt;00:00, 30.2MB/s]
+100%|##########| 13.6M/13.6M [00:00&lt;00:00, 29.5MB/s]
 </pre></div>
 </div>
 </div>
@@ -540,7 +542,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
 <p class="sphx-glr-script-out">Out:</p>
 <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)
-  90.0946      90.0127      92.1835      89.8798       0.3329
+  90.1165      90.0963      90.5290      89.9079       0.1071
 </pre></div>
 </div>
 <div class="admonition note">
@@ -579,7 +581,7 @@ 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  3.698 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  3.679 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-py">
 <div class="sphx-glr-download 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 4ca9825..de7ca4b 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -540,7 +540,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  118.8002     118.7311     125.5149     117.6018      0.8094
+  120.1621     120.0953     121.5890     119.3815      0.3826
 </pre></div>
 </div>
 <div class="admonition note">
@@ -568,7 +568,7 @@ network for ARM CPU</span></a>.</p></li>
 </ul>
 </div></blockquote>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  2.262 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  0.330 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-tflite-py">
 <div class="sphx-glr-download 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 987d7ed..5ebdaf4 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -480,7 +480,7 @@ 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  12.617 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  9.260 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-quantized-py">
 <div class="sphx-glr-download docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/7810ecf51bfc05f7d5e8a400ac3e815d/deploy_quantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_quantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
index fbecb53..572eeea 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -415,25 +415,23 @@ to your device.</p>
 Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
 
   0%|          | 0/132723 [00:00&lt;?, ?KB/s]
-  2%|1         | 2355/132723 [00:00&lt;00:05, 23547.16KB/s]
-  5%|5         | 6897/132723 [00:00&lt;00:03, 36411.79KB/s]
- 11%|#         | 14426/132723 [00:00&lt;00:02, 54159.47KB/s]
- 17%|#6        | 22218/132723 [00:00&lt;00:01, 63532.80KB/s]
- 23%|##2       | 29969/132723 [00:00&lt;00:01, 68570.90KB/s]
- 28%|##8       | 37784/132723 [00:00&lt;00:01, 71824.80KB/s]
- 34%|###4      | 45553/132723 [00:00&lt;00:01, 73739.37KB/s]
- 40%|####      | 53409/132723 [00:00&lt;00:01, 75270.44KB/s]
- 46%|####6     | 61325/132723 [00:00&lt;00:00, 76483.50KB/s]
- 52%|#####2    | 69215/132723 [00:01&lt;00:00, 77228.17KB/s]
- 58%|#####8    | 77139/132723 [00:01&lt;00:00, 77842.06KB/s]
- 64%|######4   | 85089/132723 [00:01&lt;00:00, 78341.01KB/s]
- 70%|#######   | 93065/132723 [00:01&lt;00:00, 78743.56KB/s]
- 76%|#######6  | 100940/132723 [00:01&lt;00:00, 78694.31KB/s]
- 82%|########1 | 108810/132723 [00:01&lt;00:00, 78686.50KB/s]
- 88%|########7 | 116707/132723 [00:01&lt;00:00, 78769.49KB/s]
- 94%|#########3| 124688/132723 [00:01&lt;00:00, 79080.17KB/s]
-100%|##########| 132723/132723 [00:01&lt;00:00, 79053.09KB/s]
-100%|##########| 132723/132723 [00:01&lt;00:00, 73609.57KB/s]
+  3%|3         | 4049/132723 [00:00&lt;00:03, 40486.18KB/s]
+  8%|7         | 10020/132723 [00:00&lt;00:02, 50142.97KB/s]
+ 13%|#3        | 17656/132723 [00:00&lt;00:01, 61901.21KB/s]
+ 20%|#9        | 26112/132723 [00:00&lt;00:01, 70748.49KB/s]
+ 26%|##6       | 34522/132723 [00:00&lt;00:01, 75527.74KB/s]
+ 32%|###2      | 43021/132723 [00:00&lt;00:01, 78726.98KB/s]
+ 39%|###8      | 51464/132723 [00:00&lt;00:01, 80582.82KB/s]
+ 45%|####5     | 59957/132723 [00:00&lt;00:00, 81962.62KB/s]
+ 52%|#####1    | 68507/132723 [00:00&lt;00:00, 83065.29KB/s]
+ 58%|#####8    | 77012/132723 [00:01&lt;00:00, 83676.10KB/s]
+ 64%|######4   | 85589/132723 [00:01&lt;00:00, 84309.43KB/s]
+ 71%|#######   | 94109/132723 [00:01&lt;00:00, 84578.26KB/s]
+ 77%|#######7  | 102629/132723 [00:01&lt;00:00, 84759.21KB/s]
+ 84%|########3 | 111139/132723 [00:01&lt;00:00, 84859.87KB/s]
+ 90%|######### | 119626/132723 [00:01&lt;00:00, 84771.34KB/s]
+ 97%|#########6| 128134/132723 [00:01&lt;00:00, 84862.91KB/s]
+100%|##########| 132723/132723 [00:01&lt;00:00, 79867.17KB/s]
 </pre></div>
 </div>
 <p>Create TVM runtime and do inference
@@ -473,7 +471,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
 </pre></div>
 </div>
 <img alt="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" class="sphx-glr-single-img" src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" />
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  21.077 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  19.503 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-ssd-gluoncv-py">
 <div class="sphx-glr-download docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/cccb17d28e5e8b2e94ea8cd5ec59f6ed/deploy_ssd_gluoncv.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_ssd_gluoncv.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/sg_execution_times.html b/docs/how_to/deploy_models/sg_execution_times.html
index c6f53f5..a3ccb4c 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -300,16 +300,16 @@
             
   <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>10:36.641</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>10:26.863</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
 <ul class="simple">
-<li><p><strong>03:07.507</strong>: <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></li>
-<li><p><strong>02:21.077</strong>: <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></li>
-<li><p><strong>02:02.262</strong>: <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></li>
-<li><p><strong>01:12.617</strong>: <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></li>
-<li><p><strong>01:03.698</strong>: <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></li>
-<li><p><strong>00:27.408</strong>: <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></li>
-<li><p><strong>00:21.882</strong>: <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></li>
-<li><p><strong>00:00.189</strong>: <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></li>
+<li><p><strong>03:05.573</strong>: <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></li>
+<li><p><strong>02:19.503</strong>: <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></li>
+<li><p><strong>02:00.330</strong>: <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></li>
+<li><p><strong>01:09.260</strong>: <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></li>
+<li><p><strong>01:03.679</strong>: <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></li>
+<li><p><strong>00:26.693</strong>: <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></li>
+<li><p><strong>00:21.642</strong>: <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></li>
+<li><p><strong>00:00.183</strong>: <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></li>
 </ul>
 </div>
 
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 59183e5..f4f33e6 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -588,7 +588,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip61a6064e-a681-4b2c-8249-0be39dd15fe4 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.zip9bcc8657-c468-43c6-8edf-4fc5c507204b 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>
@@ -650,7 +650,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
 </pre></div>
 </div>
 <p>When we attempt to run the model, we get a familiar error telling us that more functions need to be registerd for myfloat.</p>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index 122fd19..c6e148c 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -300,12 +300,12 @@
             
   <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>00:38.763</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:37.626</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
 <ul class="simple">
-<li><p><strong>00:35.196</strong>: <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></li>
-<li><p><strong>00:02.284</strong>: <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></li>
-<li><p><strong>00:01.080</strong>: <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></li>
-<li><p><strong>00:00.203</strong>: <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></li>
+<li><p><strong>00:34.113</strong>: <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></li>
+<li><p><strong>00:02.239</strong>: <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></li>
+<li><p><strong>00:01.074</strong>: <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></li>
+<li><p><strong>00:00.199</strong>: <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></li>
 </ul>
 </div>
 
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index c175276..a658a05 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -486,10 +486,10 @@ profile the execution time of each passes.</p>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 6728us [6728us] (47.31%; 47.31%)
-FoldScaleAxis: 7493us [3us] (52.69%; 52.69%)
-        FoldConstant: 7490us [1591us] (52.67%; 99.97%)
-                InferType: 5900us [5900us] (41.49%; 78.76%)
+InferType: 6184us [6184us] (45.49%; 45.49%)
+FoldScaleAxis: 7411us [2us] (54.51%; 54.51%)
+        FoldConstant: 7408us [1532us] (54.49%; 99.97%)
+                InferType: 5876us [5876us] (43.23%; 79.32%)
 </pre></div>
 </div>
 </div>
@@ -512,10 +512,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 5981us [5981us] (44.90%; 44.90%)
-FoldScaleAxis: 7338us [2us] (55.10%; 55.10%)
-        FoldConstant: 7336us [1520us] (55.08%; 99.97%)
-                InferType: 5816us [5816us] (43.67%; 79.28%)
+InferType: 5964us [5964us] (44.54%; 44.54%)
+FoldScaleAxis: 7425us [2us] (55.46%; 55.46%)
+        FoldConstant: 7424us [1539us] (55.44%; 99.97%)
+                InferType: 5885us [5885us] (43.95%; 79.27%)
 </pre></div>
 </div>
 <p>Register empty list to clear existing instruments.</p>
diff --git a/docs/how_to/optimize_operators/opt_conv_cuda.html b/docs/how_to/optimize_operators/opt_conv_cuda.html
index 365c71b..18b6f11 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -534,7 +534,7 @@ latency of convolution.</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 54.113171 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 41.541017 ms
 </pre></div>
 </div>
 <div class="sphx-glr-footer class 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 639450c..085ca0e 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -876,7 +876,7 @@ be able to run on our build server</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 6.615613 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 10.331691 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 993d30d..3721c81 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -431,8 +431,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018181
-Baseline: 3.428213
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019122
+Baseline: 3.457420
 </pre></div>
 </div>
 <p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -493,7 +493,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.288643
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.296049
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -561,7 +561,7 @@ vastly.</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.330478
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.326334
 </pre></div>
 </div>
 <p>Here is the generated IR after vectorization.</p>
@@ -623,7 +623,7 @@ the access pattern for A matrix is more cache friendly.</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.118830
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.115367
 </pre></div>
 </div>
 <p>Here is the generated IR after loop permutation.</p>
@@ -707,7 +707,7 @@ flattening.</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.110923
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.111165
 </pre></div>
 </div>
 <p>Here is the generated IR after array packing.</p>
@@ -794,7 +794,7 @@ write to C when all the block results are ready.</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111604
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111855
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -885,7 +885,7 @@ write to C when all the block results are ready.</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.144703
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.144806
 </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 2d00bbe..207e929 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -300,11 +300,11 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:34.798</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.060</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
 <ul class="simple">
-<li><p><strong>00:32.189</strong>: <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></li>
-<li><p><strong>00:01.378</strong>: <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></li>
-<li><p><strong>00:01.232</strong>: <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></li>
+<li><p><strong>00:32.395</strong>: <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></li>
+<li><p><strong>00:01.459</strong>: <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></li>
+<li><p><strong>00:01.207</strong>: <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></li>
 </ul>
 </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 14c7c2c..9153eaa 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -300,14 +300,14 @@
             
   <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>04:58.318</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>05:04.467</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
 <ul class="simple">
-<li><p><strong>02:24.018</strong>: <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></li>
-<li><p><strong>01:20.658</strong>: <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></li>
-<li><p><strong>00:40.577</strong>: <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></li>
-<li><p><strong>00:16.002</strong>: <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></li>
-<li><p><strong>00:08.604</strong>: <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></li>
-<li><p><strong>00:08.459</strong>: <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></li>
+<li><p><strong>02:29.882</strong>: <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></li>
+<li><p><strong>01:19.112</strong>: <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></li>
+<li><p><strong>00:40.010</strong>: <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></li>
+<li><p><strong>00:18.765</strong>: <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></li>
+<li><p><strong>00:08.566</strong>: <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></li>
+<li><p><strong>00:08.132</strong>: <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></li>
 </ul>
 </div>
 
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 3c2e63c..cea8bc9 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
@@ -471,1338 +471,481 @@ cooperative fetching, unrolling and operator fusion.</p>
   buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute} {
   attr [IterVar(blockIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;blockIdx.x&quot;)] &quot;thread_extent&quot; = 28;
   allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
-  allocate(pad_temp.shared: Pointer(shared float32), float32, [216]), storage_scope = shared;
-  allocate(kernel.shared: Pointer(shared float32), float32, [9216]), storage_scope = shared;
+  allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
+  allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
   attr [IterVar(threadIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64 {
-    conv2d_nchw_1: Buffer(conv2d_nchw, float32, [49], [], scope=&quot;local&quot;, align=16)[0] = 0f32
-    conv2d_nchw_1[7] = 0f32
+    conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope=&quot;local&quot;, align=32)[0] = 0f32
     conv2d_nchw_1[1] = 0f32
-    conv2d_nchw_1[8] = 0f32
     conv2d_nchw_1[2] = 0f32
-    conv2d_nchw_1[9] = 0f32
     conv2d_nchw_1[3] = 0f32
-    conv2d_nchw_1[10] = 0f32
     conv2d_nchw_1[4] = 0f32
-    conv2d_nchw_1[11] = 0f32
     conv2d_nchw_1[5] = 0f32
-    conv2d_nchw_1[12] = 0f32
     conv2d_nchw_1[6] = 0f32
+    conv2d_nchw_1[7] = 0f32
+    conv2d_nchw_1[8] = 0f32
+    conv2d_nchw_1[9] = 0f32
+    conv2d_nchw_1[10] = 0f32
+    conv2d_nchw_1[11] = 0f32
+    conv2d_nchw_1[12] = 0f32
     conv2d_nchw_1[13] = 0f32
     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; = 64;
-        pad_temp.shared_1: Buffer(pad_temp.shared, float32, [216], [], scope=&quot;shared&quot;)[threadIdx.x_1] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod(threadIdx.x_1, 27), 9) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod(threadIdx.x_1, 27), 9) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 9))) &amp;&amp; (floormod(threadIdx.x_1, 9) &lt; 8)), data[(((((cse_var_2 + (floordiv(threadIdx.x_1, 27)*49)) + (floordiv(floormod(threadIdx.x_1 [...]
-        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        pad_temp.shared_1[(threadIdx.x_1 + 64)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1 + 64), 27), 9) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod((threadIdx.x_1 + 64), 27), 9) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod((threadIdx.x_1 + 1), 9))) &amp;&amp; (floormod((threadIdx.x_1 + 1), 9) &lt; 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 64), 27)*49)) + (floordiv(floormod((threadIdx.x_1 + 64), 27), 9)*7)) + (floormo [...]
-        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        pad_temp.shared_1[(threadIdx.x_1 + 128)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1 + 128), 27), 9) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod((threadIdx.x_1 + 128), 27), 9) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod((threadIdx.x_1 + 2), 9))) &amp;&amp; (floormod((threadIdx.x_1 + 2), 9) &lt; 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 128), 27)*49)) + (floordiv(floormod((threadIdx.x_1 + 128), 27), 9)*7)) + (fl [...]
-        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        if @tir.likely((threadIdx.x_1 &lt; 24), dtype=bool) {
-          pad_temp.shared_1[(threadIdx.x_1 + 192)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1 + 192), 27), 9) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod((threadIdx.x_1 + 192), 27), 9) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod((threadIdx.x_1 + 3), 9))) &amp;&amp; (floormod((threadIdx.x_1 + 3), 9) &lt; 8)), data[(((((cse_var_2 + (floordiv((threadIdx.x_1 + 192), 27)*49)) + (floordiv(floormod((threadIdx.x_1 + 192), 27), 9)*7)) + ( [...]
+      for (ry.outer.outer: int32, 0, 3) {
+        let cse_var_2: int32 = (rc.outer.outer*72)
+        let cse_var_1: int32 = (ry.outer.outer*3)
+         {
+          attr [IterVar(threadIdx.x_1: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64 {
+            if @tir.likely((threadIdx.x_1 &lt; 18), dtype=bool) {
+              pad_temp.shared_1: Buffer(pad_temp.shared, float32, [72], [], scope=&quot;shared&quot;)[(threadIdx.x_1*4)] = @tir.if_then_else(((((1 &lt;= (ry.outer.outer + floormod(blockIdx.x, 7))) &amp;&amp; ((ry.outer.outer + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod((threadIdx.x_1*4), 9))) &amp;&amp; (floormod((threadIdx.x_1*4), 9) &lt; 8)), data[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*4), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) +  [...]
+            }
+            if @tir.likely((threadIdx.x_1 &lt; 18), dtype=bool) {
+              pad_temp.shared_1[((threadIdx.x_1*4) + 1)] = @tir.if_then_else(((((1 &lt;= (ry.outer.outer + floormod(blockIdx.x, 7))) &amp;&amp; ((ry.outer.outer + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod(((threadIdx.x_1*4) + 1), 9))) &amp;&amp; (floormod(((threadIdx.x_1*4) + 1), 9) &lt; 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 1), 9)) - 8)], 0 [...]
+            }
+            if @tir.likely((threadIdx.x_1 &lt; 18), dtype=bool) {
+              pad_temp.shared_1[((threadIdx.x_1*4) + 2)] = @tir.if_then_else(((((1 &lt;= (ry.outer.outer + floormod(blockIdx.x, 7))) &amp;&amp; ((ry.outer.outer + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod(((threadIdx.x_1*4) + 2), 9))) &amp;&amp; (floormod(((threadIdx.x_1*4) + 2), 9) &lt; 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 2), 9)) - 8)], 0 [...]
+            }
+            if @tir.likely((threadIdx.x_1 &lt; 18), dtype=bool) {
+              pad_temp.shared_1[((threadIdx.x_1*4) + 3)] = @tir.if_then_else(((((1 &lt;= (ry.outer.outer + floormod(blockIdx.x, 7))) &amp;&amp; ((ry.outer.outer + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod(((threadIdx.x_1*4) + 3), 9))) &amp;&amp; (floormod(((threadIdx.x_1*4) + 3), 9) &lt; 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 3), 9)) - 8)], 0 [...]
+            }
+          }
+          attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 8), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 64), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 16), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 128), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 32), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 256), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 40), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 320), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 56), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 448), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 64), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 512), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 80), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 640), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 88), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 704), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 147456)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 104), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 832), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 112), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 896), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 184320)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 128), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1024), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 136), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1088), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 221184)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 152), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1216), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 160), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1280), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 176), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1408), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 184), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1472), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 294912)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 200), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1600), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 208), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1664), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 331776)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 224), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1792), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 232), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1856), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 368640)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 248), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 1984), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 256), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2048), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 405504)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 272), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2176), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 280), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2240), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 442368)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 296), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2368), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 304), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2432), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 479232)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 320), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2560), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 328), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2624), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 516096)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 344), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2752), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 352), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2816), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 552960)]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 368), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 2944), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
+          kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 376), 3)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 3008), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*48)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*48)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*48)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*48)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*48)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*48)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*48)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 3)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 24)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 27)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 1)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 4)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 25)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 28)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 2)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 5)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 26)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 29)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 6)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 9)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 30)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 33)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 7)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 10)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 31)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 34)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 8)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 11)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 32)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 35)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 12)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 15)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 36)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 39)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 13)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 16)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 37)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 40)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 14)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 17)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 38)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 41)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 18)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 21)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 42)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 45)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 19)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 22)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 43)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 46)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 20)]))
+          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 23)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 47)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 44)]))
+          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 47)]))
         }
-        attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1: Buffer(kernel.shared, float32, [9216], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel[(((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 8), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 16), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 24), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 32), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 40), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 48), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 56), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 64), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 36864)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 80), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 88), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 96), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 104), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 120), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 128), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 136), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 73728)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 152), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 160), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 168), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 176), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 184), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 192), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 200), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 208), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 110592)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 224), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 232), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 240), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 248), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 256), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 264), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 272), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 280), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 147456)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 296), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 304), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 312), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 320), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 328), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 336), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 344), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 352), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 184320)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 368), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 376), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3072)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 384), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 392), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3200)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 400), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3264)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 408), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3328)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 416), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3392)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 424), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3456)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 221184)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3520)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 440), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3584)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 448), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3648)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 456), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3712)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 464), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3776)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 472), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3840)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 480), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3904)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 488), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 3968)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 496), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 258048)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4096)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 512), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4160)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 520), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4224)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 528), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4288)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 536), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4352)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 544), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4416)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 552), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4480)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 560), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4544)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 568), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4608)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 294912)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4672)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 584), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4736)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 592), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4800)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 600), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4864)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 608), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4928)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 616), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 4992)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 624), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5056)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 632), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5120)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 640), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5184)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 331776)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5248)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 656), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5312)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 664), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5376)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 672), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5440)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 680), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5504)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 688), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5568)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 696), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5632)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 704), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5696)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 712), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5760)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 368640)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5824)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 728), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5888)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 736), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 5952)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 744), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6016)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 752), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6080)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 760), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6144)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 768), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6208)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 776), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6272)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 784), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6336)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 405504)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6400)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 800), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6464)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 808), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6528)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 816), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6592)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 824), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6656)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 832), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6720)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 840), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6784)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 848), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6848)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 856), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6912)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 442368)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 6976)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 872), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7040)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 880), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7104)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 888), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7168)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 896), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7232)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 904), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7296)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 912), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7360)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 920), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7424)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 928), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7488)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 479232)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7552)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 944), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7616)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 952), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7680)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 960), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7744)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 968), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7808)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 976), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7872)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 984), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 7936)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 992), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8000)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1000), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8064)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 516096)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8128)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1016), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8192)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1024), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8256)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1032), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8320)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1040), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8384)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1048), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8448)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1056), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8512)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1064), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8576)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1072), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8640)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + cse_var_1) + threadIdx.x_2) + 552960)]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8704)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1088), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 64), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8768)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1096), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 56), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8832)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1104), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 48), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8896)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 40), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 8960)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1120), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 32), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 9024)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1128), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 24), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 9088)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1136), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 72))]
-        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-        kernel.shared_1[(threadIdx.x_2 + 9152)] = kernel[((((floordiv(blockIdx.x, 7)*589824) + (floordiv((floordiv(threadIdx.x_2, 8) + 1144), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 72))]
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*72)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[81]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[81]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*72)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[82]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*72)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[83]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*72)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[84]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*72)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[85]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*72)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[86]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*72)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*72) + 4608)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 1)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*72) + 4609)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*72) + 2)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*72) + 4610)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 9)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*72) + 4617)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 10)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*72) + 4618)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*72) + 11)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*72) + 4619)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 18)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*72) + 4626)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 19)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*72) + 4627)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*72) + 20)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*72) + 4628)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 27)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[87]*kernel.shared_1[((threadIdx.x*72) + 4635)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 28)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[88]*kernel.shared_1[((threadIdx.x*72) + 4636)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[89]*kernel.shared_1[((threadIdx.x*72) + 29)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[89]*kernel.shared_1[((threadIdx.x*72) + 4637)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[90]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[90]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[91]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[92]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[93]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[94]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[95]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 3)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*72) + 4611)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 4)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*72) + 4612)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*72) + 5)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*72) + 4613)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 12)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*72) + 4620)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 13)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*72) + 4621)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*72) + 14)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*72) + 4622)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 21)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*72) + 4629)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 22)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*72) + 4630)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*72) + 23)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*72) + 4631)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 30)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[96]*kernel.shared_1[((threadIdx.x*72) + 4638)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 31)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[97]*kernel.shared_1[((threadIdx.x*72) + 4639)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[98]*kernel.shared_1[((threadIdx.x*72) + 32)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[98]*kernel.shared_1[((threadIdx.x*72) + 4640)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[72]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[72]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[99]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[99]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[73]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[100]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[74]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[101]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[75]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[102]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[76]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[103]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[77]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[104]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 6)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*72) + 4614)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 7)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*72) + 4615)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*72) + 8)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*72) + 4616)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 15)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*72) + 4623)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 16)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*72) + 4624)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*72) + 17)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*72) + 4625)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 24)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[78]*kernel.shared_1[((threadIdx.x*72) + 4632)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 25)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[79]*kernel.shared_1[((threadIdx.x*72) + 4633)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[80]*kernel.shared_1[((threadIdx.x*72) + 26)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[80]*kernel.shared_1[((threadIdx.x*72) + 4634)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 33)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[105]*kernel.shared_1[((threadIdx.x*72) + 4641)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 34)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[106]*kernel.shared_1[((threadIdx.x*72) + 4642)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[107]*kernel.shared_1[((threadIdx.x*72) + 35)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[107]*kernel.shared_1[((threadIdx.x*72) + 4643)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[108]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[108]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[135]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[135]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[162]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[162]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[189]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[189]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[109]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[136]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[163]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[190]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[110]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[137]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[164]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[191]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[111]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[138]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[165]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[192]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[112]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[139]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[166]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[193]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[113]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[140]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[167]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[194]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 36)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[114]*kernel.shared_1[((threadIdx.x*72) + 4644)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 37)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[115]*kernel.shared_1[((threadIdx.x*72) + 4645)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[116]*kernel.shared_1[((threadIdx.x*72) + 38)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[116]*kernel.shared_1[((threadIdx.x*72) + 4646)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 45)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[141]*kernel.shared_1[((threadIdx.x*72) + 4653)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 46)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[142]*kernel.shared_1[((threadIdx.x*72) + 4654)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[143]*kernel.shared_1[((threadIdx.x*72) + 47)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[143]*kernel.shared_1[((threadIdx.x*72) + 4655)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 54)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[168]*kernel.shared_1[((threadIdx.x*72) + 4662)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 55)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[169]*kernel.shared_1[((threadIdx.x*72) + 4663)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[170]*kernel.shared_1[((threadIdx.x*72) + 56)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[170]*kernel.shared_1[((threadIdx.x*72) + 4664)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 63)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[195]*kernel.shared_1[((threadIdx.x*72) + 4671)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 64)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[196]*kernel.shared_1[((threadIdx.x*72) + 4672)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[197]*kernel.shared_1[((threadIdx.x*72) + 65)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[197]*kernel.shared_1[((threadIdx.x*72) + 4673)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[117]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[117]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[144]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[144]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[171]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[171]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[198]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[198]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[118]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[145]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[172]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[199]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[119]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[146]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[173]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[200]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[120]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[147]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[174]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[201]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[121]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[148]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[175]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[202]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[122]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[149]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[176]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[203]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 39)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[123]*kernel.shared_1[((threadIdx.x*72) + 4647)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 40)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[124]*kernel.shared_1[((threadIdx.x*72) + 4648)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[125]*kernel.shared_1[((threadIdx.x*72) + 41)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[125]*kernel.shared_1[((threadIdx.x*72) + 4649)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 48)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[150]*kernel.shared_1[((threadIdx.x*72) + 4656)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 49)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[151]*kernel.shared_1[((threadIdx.x*72) + 4657)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[152]*kernel.shared_1[((threadIdx.x*72) + 50)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[152]*kernel.shared_1[((threadIdx.x*72) + 4658)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 57)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[177]*kernel.shared_1[((threadIdx.x*72) + 4665)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 58)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[178]*kernel.shared_1[((threadIdx.x*72) + 4666)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[179]*kernel.shared_1[((threadIdx.x*72) + 59)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[179]*kernel.shared_1[((threadIdx.x*72) + 4667)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 66)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[204]*kernel.shared_1[((threadIdx.x*72) + 4674)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 67)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[205]*kernel.shared_1[((threadIdx.x*72) + 4675)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[206]*kernel.shared_1[((threadIdx.x*72) + 68)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[206]*kernel.shared_1[((threadIdx.x*72) + 4676)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[126]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[126]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[153]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[153]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[180]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[180]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[207]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[207]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[127]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[154]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[181]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[208]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-        conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[128]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[155]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[182]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[209]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-        conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[129]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[156]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[183]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[210]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-        conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[130]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[157]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[184]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[211]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-        conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[131]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[158]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[185]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[212]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-        conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 42)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[132]*kernel.shared_1[((threadIdx.x*72) + 4650)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 43)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[133]*kernel.shared_1[((threadIdx.x*72) + 4651)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[134]*kernel.shared_1[((threadIdx.x*72) + 44)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[134]*kernel.shared_1[((threadIdx.x*72) + 4652)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 51)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[159]*kernel.shared_1[((threadIdx.x*72) + 4659)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 52)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[160]*kernel.shared_1[((threadIdx.x*72) + 4660)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[161]*kernel.shared_1[((threadIdx.x*72) + 53)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[161]*kernel.shared_1[((threadIdx.x*72) + 4661)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 60)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[186]*kernel.shared_1[((threadIdx.x*72) + 4668)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 61)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[187]*kernel.shared_1[((threadIdx.x*72) + 4669)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[188]*kernel.shared_1[((threadIdx.x*72) + 62)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[188]*kernel.shared_1[((threadIdx.x*72) + 4670)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 69)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[213]*kernel.shared_1[((threadIdx.x*72) + 4677)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 70)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[214]*kernel.shared_1[((threadIdx.x*72) + 4678)]))
-        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[215]*kernel.shared_1[((threadIdx.x*72) + 71)]))
-        conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[215]*kernel.shared_1[((threadIdx.x*72) + 4679)]))
       }
     }
-    for (i3.inner: int32, 0, 7) {
-      compute[((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[i3.inner] + bias[((floordiv(blockIdx.x, 7)*128) + threadIdx.x)]), 0f32)
-      compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner) + 3136)] = max((conv2d_nchw_1[(i3.inner + 7)] + bias[(((floordiv(blockIdx.x, 7)*128) + threadIdx.x) + 64)]), 0f32)
+    for (i1.inner: int32, 0, 2) {
+      for (i3.inner: int32, 0, 7) {
+        compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*98)) + (i1.inner*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[((i1.inner*7) + i3.inner)] + bias[(((floordiv(blockIdx.x, 7)*128) + (threadIdx.x*2)) + i1.inner)]), 0f32)
+      }
     }
   }
 }
@@ -1840,7 +983,7 @@ cooperative fetching, unrolling and operator fusion.</p>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.255 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.371 ms
 </pre></div>
 </div>
 </div>
@@ -1871,9 +1014,9 @@ conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o
 conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
 conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
 conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
-conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
+conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
 conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
-conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
+conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
 conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
 conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
 conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
@@ -1882,19 +1025,19 @@ conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, fact
 conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
 conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
 conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
 conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
-conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
-conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
+conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
 s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
 compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
 compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
 compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
 compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
-compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
+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=1)
 compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
@@ -1922,11 +1065,11 @@ s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
 kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
 s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis(&quot;threadIdx.x&quot;))
 pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=4)
 s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
 pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
 s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis(&quot;threadIdx.x&quot;))
-s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, &quot;auto_unroll_max_step&quot;, 1024)
+s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, &quot;auto_unroll_max_step&quot;, 512)
 s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, &quot;unroll_explicit&quot;, True)
 
 CUDA source code:
@@ -1946,1187 +1089,428 @@ CUDA source code:
 #endif
 extern &quot;C&quot; __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
   float conv2d_nchw[14];
-  __shared__ float pad_temp_shared[216];
-  __shared__ float kernel_shared[9216];
+  __shared__ float pad_temp_shared[72];
+  __shared__ float kernel_shared[3072];
   conv2d_nchw[0] = 0.000000e+00f;
-  conv2d_nchw[7] = 0.000000e+00f;
   conv2d_nchw[1] = 0.000000e+00f;
-  conv2d_nchw[8] = 0.000000e+00f;
   conv2d_nchw[2] = 0.000000e+00f;
-  conv2d_nchw[9] = 0.000000e+00f;
   conv2d_nchw[3] = 0.000000e+00f;
-  conv2d_nchw[10] = 0.000000e+00f;
   conv2d_nchw[4] = 0.000000e+00f;
-  conv2d_nchw[11] = 0.000000e+00f;
   conv2d_nchw[5] = 0.000000e+00f;
-  conv2d_nchw[12] = 0.000000e+00f;
   conv2d_nchw[6] = 0.000000e+00f;
+  conv2d_nchw[7] = 0.000000e+00f;
+  conv2d_nchw[8] = 0.000000e+00f;
+  conv2d_nchw[9] = 0.000000e+00f;
+  conv2d_nchw[10] = 0.000000e+00f;
+  conv2d_nchw[11] = 0.000000e+00f;
+  conv2d_nchw[12] = 0.000000e+00f;
   conv2d_nchw[13] = 0.000000e+00f;
   for (int rc_outer_outer = 0; rc_outer_outer &lt; 64; ++rc_outer_outer) {
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = (((((1 &lt;= (((((int)threadIdx.x) % 27) / 9) + (((int)blockIdx.x) % 7))) &amp;&amp; ((((((int)threadIdx.x) % 27) / 9) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 9))) &amp;&amp; ((((int)threadIdx.x) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
-    pad_temp_shared[(((int)threadIdx.x) + 64)] = (((((1 &lt;= ((((((int)threadIdx.x) + 10) % 27) / 9) + (((int)blockIdx.x) % 7))) &amp;&amp; (((((((int)threadIdx.x) + 10) % 27) / 9) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)threadIdx.x) + 1) % 9))) &amp;&amp; (((((int)threadIdx.x) + 1) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 64) / 27) * 49)) + ((((((int)threadIdx.x) + 10) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int) [...]
-    pad_temp_shared[(((int)threadIdx.x) + 128)] = (((((1 &lt;= ((((((int)threadIdx.x) + 20) % 27) / 9) + (((int)blockIdx.x) % 7))) &amp;&amp; (((((((int)threadIdx.x) + 20) % 27) / 9) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)threadIdx.x) + 2) % 9))) &amp;&amp; (((((int)threadIdx.x) + 2) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 128) / 27) * 49)) + ((((((int)threadIdx.x) + 20) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((in [...]
-    if (((int)threadIdx.x) &lt; 24) {
-      pad_temp_shared[(((int)threadIdx.x) + 192)] = (((((1 &lt;= ((((((int)threadIdx.x) + 3) % 27) / 9) + (((int)blockIdx.x) % 7))) &amp;&amp; (((((((int)threadIdx.x) + 3) % 27) / 9) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)threadIdx.x) + 3) % 9))) &amp;&amp; (((((int)threadIdx.x) + 3) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 192) / 27) * 49)) + ((((((int)threadIdx.x) + 3) % 27) / 9) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int [...]
+    for (int ry_outer_outer = 0; ry_outer_outer &lt; 3; ++ry_outer_outer) {
+      __syncthreads();
+      if (((int)threadIdx.x) &lt; 18) {
+        pad_temp_shared[(((int)threadIdx.x) * 4)] = (((((1 &lt;= (ry_outer_outer + (((int)blockIdx.x) % 7))) &amp;&amp; ((ry_outer_outer + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)threadIdx.x) * 4) % 9))) &amp;&amp; (((((int)threadIdx.x) * 4) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 4) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 4) % 9)) - 8)] : 0.000000e+00f);
+      }
+      if (((int)threadIdx.x) &lt; 18) {
+        pad_temp_shared[((((int)threadIdx.x) * 4) + 1)] = (((((1 &lt;= (ry_outer_outer + (((int)blockIdx.x) % 7))) &amp;&amp; ((ry_outer_outer + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= (((((int)threadIdx.x) * 4) + 1) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 4) + 1) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 1) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 1) % 9)) - 8)] : 0.000000e+00f);
+      }
+      if (((int)threadIdx.x) &lt; 18) {
+        pad_temp_shared[((((int)threadIdx.x) * 4) + 2)] = (((((1 &lt;= (ry_outer_outer + (((int)blockIdx.x) % 7))) &amp;&amp; ((ry_outer_outer + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= (((((int)threadIdx.x) * 4) + 2) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 4) + 2) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 2) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 2) % 9)) - 8)] : 0.000000e+00f);
+      }
+      if (((int)threadIdx.x) &lt; 18) {
+        pad_temp_shared[((((int)threadIdx.x) * 4) + 3)] = (((((1 &lt;= (ry_outer_outer + (((int)blockIdx.x) % 7))) &amp;&amp; ((ry_outer_outer + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= (((((int)threadIdx.x) * 4) + 3) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 4) + 3) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 4) + 3) / 9) * 49)) + (ry_outer_outer * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 4) + 3) % 9)) - 8)] : 0.000000e+00f);
+      }
+      kernel_shared[((int)threadIdx.x)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 64)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 192)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 36864)];
+      kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 384)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 73728)];
+      kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 576)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 110592)];
+      kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 768)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 147456)];
+      kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 960)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 184320)];
+      kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 221184)];
+      kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
+      kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 294912)];
+      kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 331776)];
+      kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 368640)];
+      kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 405504)];
+      kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 442368)];
+      kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 479232)];
+      kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 516096)];
+      kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[((((((((((int)blockIdx.x) / 7) * 589824) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 552960)];
+      kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      __syncthreads();
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 48)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 48)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 48)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 48)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 48)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 48)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 48)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 3)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 24)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 27)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 1)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 4)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 25)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 28)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 2)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 5)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 48) + 26)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 48) + 29)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 6)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 9)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 30)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 33)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 7)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 10)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 31)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 34)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 8)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 11)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 48) + 32)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 48) + 35)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 12)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 15)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 36)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 39)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 13)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 16)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 37)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 40)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 14)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 17)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 48) + 38)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 48) + 41)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 18)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 21)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 42)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 45)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 19)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 22)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 43)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 46)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 20)]));
+      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 23)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 48) + 44)]));
+      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 48) + 47)]));
     }
-    kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x))];
-    kernel_shared[(((int)threadIdx.x) + 64)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 64) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 128)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 128) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 192)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 192) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 256)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 256) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 320)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 320) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 384)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 384) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 448) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 512)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 512) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 576)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 36864)];
-    kernel_shared[(((int)threadIdx.x) + 640)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 640) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 704)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 704) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 768)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 768) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 832)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 832) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 896) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 960)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 960) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1024)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1024) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1088)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1088) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 1152)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 73728)];
-    kernel_shared[(((int)threadIdx.x) + 1216)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1216) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1280)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1280) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1344) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1408)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1408) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1472)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1472) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1536)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1536) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1600)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1600) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1664)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1664) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 1728)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 110592)];
-    kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1792) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1856)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1856) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1920)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1920) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 1984)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 1984) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2048)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2048) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2112)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2112) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2176)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2176) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2240) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 2304)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 147456)];
-    kernel_shared[(((int)threadIdx.x) + 2368)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2368) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2432)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2432) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2496)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2496) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2560)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2560) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2624)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2624) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2688) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2752)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2752) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 2816)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2816) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 2880)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 184320)];
-    kernel_shared[(((int)threadIdx.x) + 2944)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 2944) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3008)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3008) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3072)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3072) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3136) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3200)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3200) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3264)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3264) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3328)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3328) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3392)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3392) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 3456)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 221184)];
-    kernel_shared[(((int)threadIdx.x) + 3520)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3520) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3584) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3648)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3648) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3712)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3712) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3776)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3776) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3840)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3840) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3904)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3904) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 3968)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 3968) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 258048)];
-    kernel_shared[(((int)threadIdx.x) + 4096)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4096) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4160)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4160) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4224)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4224) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4288)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4288) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4352)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4352) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4416)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4416) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4480) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4544)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4544) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 4608)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 294912)];
-    kernel_shared[(((int)threadIdx.x) + 4672)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4672) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4736)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4736) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4800)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4800) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4864)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4864) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4928)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4928) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 4992)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 4992) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5056)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5056) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5120)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5120) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 5184)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 331776)];
-    kernel_shared[(((int)threadIdx.x) + 5248)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5248) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5312)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5312) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5376)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5376) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5440)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5440) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5504)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5504) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5568)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5568) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5632)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5632) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5696)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5696) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 5760)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 368640)];
-    kernel_shared[(((int)threadIdx.x) + 5824)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5824) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5888)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5888) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 5952)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 5952) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6016)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6016) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6080)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6080) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6144)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6144) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6208)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6208) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6272)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6272) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 6336)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 405504)];
-    kernel_shared[(((int)threadIdx.x) + 6400)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6400) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6464)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6464) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6528)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6528) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6592)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6592) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6656)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6656) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6720)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6720) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6784)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6784) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 6848)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6848) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 6912)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 442368)];
-    kernel_shared[(((int)threadIdx.x) + 6976)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 6976) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7040)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7040) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7104)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7104) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7168)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7168) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7232)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7232) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7296)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7296) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7360)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7360) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7424)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7424) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 7488)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 479232)];
-    kernel_shared[(((int)threadIdx.x) + 7552)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7552) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7616)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7616) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7680)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7680) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7744)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7744) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7808)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7808) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7872)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7872) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 7936)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 7936) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8000)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8000) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 8064)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 516096)];
-    kernel_shared[(((int)threadIdx.x) + 8128)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8128) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8192)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8192) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8256)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8256) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8320)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8320) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8384)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8384) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8448)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8448) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8512)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8512) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8576)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8576) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    kernel_shared[(((int)threadIdx.x) + 8640)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (rc_outer_outer * 72)) + ((int)threadIdx.x)) + 552960)];
-    kernel_shared[(((int)threadIdx.x) + 8704)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8704) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 64) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8768)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8768) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 56) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8832)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8832) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 48) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8896)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8896) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 40) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 8960)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 8960) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 32) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 9024)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 9024) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 24) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 9088)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 9088) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) + 16) % 72))];
-    kernel_shared[(((int)threadIdx.x) + 9152)] = kernel[(((((((int)blockIdx.x) / 7) * 589824) + (((((int)threadIdx.x) + 9152) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) + 8))];
-    __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[0] * kernel_shared[(((int)threadIdx.x) * 72)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[0] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[27] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[54] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[81] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[1] * kernel_shared[(((int)threadIdx.x) * 72)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[1] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[28] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[55] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[82] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[2] * kernel_shared[(((int)threadIdx.x) * 72)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[2] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[29] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[56] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[83] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[3] * kernel_shared[(((int)threadIdx.x) * 72)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[3] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[30] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[57] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[84] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[4] * kernel_shared[(((int)threadIdx.x) * 72)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[4] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[31] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[58] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[85] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[5] * kernel_shared[(((int)threadIdx.x) * 72)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[5] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[32] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[59] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[86] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[6] * kernel_shared[(((int)threadIdx.x) * 72)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[6] * kernel_shared[((((int)threadIdx.x) * 72) + 4608)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 1)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[7] * kernel_shared[((((int)threadIdx.x) * 72) + 4609)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 72) + 2)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[8] * kernel_shared[((((int)threadIdx.x) * 72) + 4610)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 9)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[33] * kernel_shared[((((int)threadIdx.x) * 72) + 4617)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 10)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[34] * kernel_shared[((((int)threadIdx.x) * 72) + 4618)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 72) + 11)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[35] * kernel_shared[((((int)threadIdx.x) * 72) + 4619)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 18)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[60] * kernel_shared[((((int)threadIdx.x) * 72) + 4626)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 19)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[61] * kernel_shared[((((int)threadIdx.x) * 72) + 4627)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 72) + 20)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[62] * kernel_shared[((((int)threadIdx.x) * 72) + 4628)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 27)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[87] * kernel_shared[((((int)threadIdx.x) * 72) + 4635)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 28)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[88] * kernel_shared[((((int)threadIdx.x) * 72) + 4636)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 72) + 29)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[89] * kernel_shared[((((int)threadIdx.x) * 72) + 4637)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[9] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[36] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[63] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[90] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[10] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[37] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[64] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[91] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[11] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[38] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[65] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[92] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[12] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[39] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[66] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[93] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[13] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[40] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[67] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[94] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[14] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[41] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[68] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[95] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 3)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[15] * kernel_shared[((((int)threadIdx.x) * 72) + 4611)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 4)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[16] * kernel_shared[((((int)threadIdx.x) * 72) + 4612)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 72) + 5)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[17] * kernel_shared[((((int)threadIdx.x) * 72) + 4613)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 12)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[42] * kernel_shared[((((int)threadIdx.x) * 72) + 4620)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 13)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[43] * kernel_shared[((((int)threadIdx.x) * 72) + 4621)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 72) + 14)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[44] * kernel_shared[((((int)threadIdx.x) * 72) + 4622)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 21)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[69] * kernel_shared[((((int)threadIdx.x) * 72) + 4629)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 22)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[70] * kernel_shared[((((int)threadIdx.x) * 72) + 4630)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 72) + 23)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[71] * kernel_shared[((((int)threadIdx.x) * 72) + 4631)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 30)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[96] * kernel_shared[((((int)threadIdx.x) * 72) + 4638)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 31)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[97] * kernel_shared[((((int)threadIdx.x) * 72) + 4639)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 72) + 32)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[98] * kernel_shared[((((int)threadIdx.x) * 72) + 4640)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[18] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[45] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[72] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[99] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[19] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[46] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[73] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[100] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[20] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[47] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[74] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[101] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[21] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[48] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[75] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[102] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[22] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[49] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[76] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[103] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[23] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[50] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[77] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[104] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 6)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[24] * kernel_shared[((((int)threadIdx.x) * 72) + 4614)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 7)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[25] * kernel_shared[((((int)threadIdx.x) * 72) + 4615)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 72) + 8)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[26] * kernel_shared[((((int)threadIdx.x) * 72) + 4616)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 15)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[51] * kernel_shared[((((int)threadIdx.x) * 72) + 4623)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 16)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[52] * kernel_shared[((((int)threadIdx.x) * 72) + 4624)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 72) + 17)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[53] * kernel_shared[((((int)threadIdx.x) * 72) + 4625)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 24)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[78] * kernel_shared[((((int)threadIdx.x) * 72) + 4632)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 25)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[79] * kernel_shared[((((int)threadIdx.x) * 72) + 4633)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 72) + 26)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[80] * kernel_shared[((((int)threadIdx.x) * 72) + 4634)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 33)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[105] * kernel_shared[((((int)threadIdx.x) * 72) + 4641)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 34)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[106] * kernel_shared[((((int)threadIdx.x) * 72) + 4642)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 72) + 35)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[107] * kernel_shared[((((int)threadIdx.x) * 72) + 4643)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[108] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[135] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[162] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[189] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[109] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[136] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[163] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[190] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[110] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[137] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[164] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[191] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[111] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[138] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[165] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[192] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[112] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[139] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[166] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[193] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[113] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[140] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[167] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[194] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 36)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[114] * kernel_shared[((((int)threadIdx.x) * 72) + 4644)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 37)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[115] * kernel_shared[((((int)threadIdx.x) * 72) + 4645)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 72) + 38)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[116] * kernel_shared[((((int)threadIdx.x) * 72) + 4646)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 45)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[141] * kernel_shared[((((int)threadIdx.x) * 72) + 4653)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 46)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[142] * kernel_shared[((((int)threadIdx.x) * 72) + 4654)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 72) + 47)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[143] * kernel_shared[((((int)threadIdx.x) * 72) + 4655)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 54)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[168] * kernel_shared[((((int)threadIdx.x) * 72) + 4662)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 55)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[169] * kernel_shared[((((int)threadIdx.x) * 72) + 4663)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 72) + 56)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[170] * kernel_shared[((((int)threadIdx.x) * 72) + 4664)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 63)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[195] * kernel_shared[((((int)threadIdx.x) * 72) + 4671)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 64)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[196] * kernel_shared[((((int)threadIdx.x) * 72) + 4672)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 72) + 65)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[197] * kernel_shared[((((int)threadIdx.x) * 72) + 4673)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[117] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[144] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[171] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[198] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[118] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[145] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[172] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[199] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[119] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[146] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[173] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[200] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[120] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[147] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[174] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[201] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[121] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[148] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[175] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[202] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[122] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[149] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[176] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[203] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 39)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[123] * kernel_shared[((((int)threadIdx.x) * 72) + 4647)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 40)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[124] * kernel_shared[((((int)threadIdx.x) * 72) + 4648)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 72) + 41)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[125] * kernel_shared[((((int)threadIdx.x) * 72) + 4649)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 48)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[150] * kernel_shared[((((int)threadIdx.x) * 72) + 4656)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 49)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[151] * kernel_shared[((((int)threadIdx.x) * 72) + 4657)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 72) + 50)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[152] * kernel_shared[((((int)threadIdx.x) * 72) + 4658)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 57)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[177] * kernel_shared[((((int)threadIdx.x) * 72) + 4665)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 58)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[178] * kernel_shared[((((int)threadIdx.x) * 72) + 4666)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 72) + 59)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[179] * kernel_shared[((((int)threadIdx.x) * 72) + 4667)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 66)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[204] * kernel_shared[((((int)threadIdx.x) * 72) + 4674)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 67)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[205] * kernel_shared[((((int)threadIdx.x) * 72) + 4675)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 72) + 68)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[206] * kernel_shared[((((int)threadIdx.x) * 72) + 4676)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[126] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[153] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[180] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[207] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[127] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[154] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[181] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[208] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-    conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[128] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[155] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[182] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[209] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-    conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[129] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[156] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[183] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[210] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-    conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[130] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[157] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[184] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[211] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-    conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[131] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[158] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[185] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[212] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-    conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 42)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[132] * kernel_shared[((((int)threadIdx.x) * 72) + 4650)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 43)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[133] * kernel_shared[((((int)threadIdx.x) * 72) + 4651)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 72) + 44)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[134] * kernel_shared[((((int)threadIdx.x) * 72) + 4652)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 51)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[159] * kernel_shared[((((int)threadIdx.x) * 72) + 4659)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 52)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[160] * kernel_shared[((((int)threadIdx.x) * 72) + 4660)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 72) + 53)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[161] * kernel_shared[((((int)threadIdx.x) * 72) + 4661)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 60)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[186] * kernel_shared[((((int)threadIdx.x) * 72) + 4668)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 61)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[187] * kernel_shared[((((int)threadIdx.x) * 72) + 4669)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 72) + 62)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[188] * kernel_shared[((((int)threadIdx.x) * 72) + 4670)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 69)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[213] * kernel_shared[((((int)threadIdx.x) * 72) + 4677)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 70)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[214] * kernel_shared[((((int)threadIdx.x) * 72) + 4678)]));
-    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 72) + 71)]));
-    conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[215] * kernel_shared[((((int)threadIdx.x) * 72) + 4679)]));
   }
-  for (int i3_inner = 0; i3_inner &lt; 7; ++i3_inner) {
-    compute[(((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[i3_inner] + bias[(((((int)blockIdx.x) / 7) * 128) + ((int)threadIdx.x))]), 0.000000e+00f);
-    compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner) + 3136)] = max((conv2d_nchw[(i3_inner + 7)] + bias[((((((int)blockIdx.x) / 7) * 128) + ((int)threadIdx.x)) + 64)]), 0.000000e+00f);
+  for (int i1_inner = 0; i1_inner &lt; 2; ++i1_inner) {
+    for (int i3_inner = 0; i3_inner &lt; 7; ++i3_inner) {
+      compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
+    }
   }
 }
 </pre></div>
@@ -3164,7 +1548,7 @@ In the example below we resume the status and do more 5 trials.</p>
 Get devices for measurement successfully!
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  24.018 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  29.882 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py">
 <div class="sphx-glr-download docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/e3e540f3b477c0c52d8eb73e674e8ffd/tune_conv2d_layer_cuda.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_conv2d_layer_cuda.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
index 2f5c41b..34bdb16 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -876,7 +876,7 @@ so we can read the log file and load the best schedules.</p>
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-   9.8818       9.8835       9.8960       9.8659       0.0123
+  10.0844      10.0845      10.1364      10.0324       0.0425
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
index b1976a9..831559f 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -895,7 +895,7 @@ so we can read the log file and load the best schedules.</p>
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  764.9774     762.1082     772.5468     760.2772      5.4043
+  765.4736     767.2281     767.7041     761.4887      2.8245
 </pre></div>
 </div>
 </div>
@@ -917,7 +917,7 @@ to learn how to use the RPC Tracker and RPC Server.
 To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
 with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
 </ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  20.658 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  19.112 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-x86-py">
 <div class="sphx-glr-download docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/e416b94ca1090b0897c0f6e0df95b911/tune_network_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_network_x86.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
index 5a28f4a..f3732ba 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -600,23 +600,27 @@ 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} {
-  for (i0.outer.i1.outer.fused: int32, 0, 64) &quot;parallel&quot; {
-    allocate(compute_3: Pointer(global float32), float32, [1024]), storage_scope = global {
-      for (i.outer.inner: int32, 0, 64) {
-        for (j.init: int32, 0, 16) {
-          compute_4: Buffer(compute_3, float32, [1024], [])[((i.outer.inner*16) + j.init)] = 0f32
+  for (i0.outer.i1.outer.fused: int32, 0, 128) &quot;parallel&quot; {
+    allocate(compute_3: Pointer(global float32), float32, [512]), storage_scope = global {
+      for (nb_j.inner: int32, 0, 2) {
+        for (i.inner.init: int32, 0, 16) {
+          for (j.init: int32, 0, 16) {
+            compute_4: Buffer(compute_3, float32, [512], [])[(((i.inner.init*32) + (nb_j.inner*16)) + j.init)] = 0f32
+          }
         }
-        for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
-          for (j: int32, 0, 16) {
-            let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
-            let cse_var_2: int32 = ((i.outer.inner*16) + j)
-            compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i.outer.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+        for (elem_idx: int32, 0, let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+          for (i.inner: int32, 0, 16) {
+            for (j: int32, 0, 16) {
+              let cse_var_3: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
+              let cse_var_2: int32 = (((i.inner*32) + (nb_j.inner*16)) + j)
+              compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((floordiv(i0.outer.i1.outer.fused, 16)*4096) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+            }
           }
         }
       }
-      for (i0.inner: int32, 0, 64) {
-        let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*32768) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
-        compute[ramp(cse_var_4, 1, 16)] = max((compute_4[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_4, 1, 16)]), broadcast(0f32, 16))
+      for (i0.inner: int32, 0, 16) {
+        let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
+        compute[ramp(cse_var_4, 1, 32)] = max((compute_4[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
       }
     }
   }
@@ -655,7 +659,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 2.262 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.491 ms
 </pre></div>
 </div>
 <div class="admonition note">
diff --git a/docs/how_to/tune_with_autotvm/sg_execution_times.html b/docs/how_to/tune_with_autotvm/sg_execution_times.html
index b798f6b..367d402 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -300,13 +300,13 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:43.629</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:44.335</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
 <ul class="simple">
-<li><p><strong>00:42.830</strong>: <a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></li>
-<li><p><strong>00:00.210</strong>: <a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></li>
-<li><p><strong>00:00.199</strong>: <a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></li>
-<li><p><strong>00:00.196</strong>: <a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></li>
-<li><p><strong>00:00.194</strong>: <a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></li>
+<li><p><strong>00:43.478</strong>: <a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></li>
+<li><p><strong>00:00.224</strong>: <a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></li>
+<li><p><strong>00:00.215</strong>: <a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></li>
+<li><p><strong>00:00.212</strong>: <a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></li>
+<li><p><strong>00:00.207</strong>: <a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></li>
 </ul>
 </div>
 
diff --git a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
index 61371e3..6dc7dc2 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -1142,8 +1142,8 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 4, 32]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 128]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2885496
-No: 6   GFLOPS: 103.84/103.84   result: MeasureResult(costs=(0.0022294135833333334,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5974090099334717, timestamp=1648924642.0972478)      [(&#39;tile_f&#39;, [-1, 1, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 4]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,3754080
-No: 7   GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 6   GFLOPS: 109.34/109.34   result: MeasureResult(costs=(0.0021172386666666663,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6623878479003906, timestamp=1648932944.1037817)      [(&#39;tile_f&#39;, [-1, 1, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 4]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,3754080
+No: 7   GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -1266,7 +1266,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 16, 32]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 256, 1]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6225319
-No: 8   GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 8   GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -1389,7 +1389,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 1, 32]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 64]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,943546
-No: 9   GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 9   GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -1512,7 +1512,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 16, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 16, 32]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2868708
-No: 10  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 10  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 142, in build
     res = future.result()
   File &quot;/usr/lib/python3.7/concurrent/futures/_base.py&quot;, line 435, in result
@@ -1530,7 +1530,7 @@ No: 10  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
 TimeoutError
 
         [(&#39;tile_f&#39;, [-1, 32, 2, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 4, 2]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4691833
-No: 11  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 11  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -1653,7 +1653,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 2, 64]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 4, 4]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1042124
-No: 12  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 12  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -1776,7 +1776,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 32, 1, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 32, 16]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,10013405
-No: 13  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 13  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -1899,7 +1899,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 32]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6732082
-No: 14  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 14  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -2022,7 +2022,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 4, 32]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 4, 128]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,7536735
-No: 15  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 15  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -2145,7 +2145,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 1, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 128, 4]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,482121
-No: 16  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 16  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -2268,7 +2268,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 1, 16]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 32, 8]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2824525
-No: 17  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 17  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -2391,7 +2391,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 8]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4559286
-No: 18  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 18  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 571, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 523, in _build_func_common
@@ -2514,7 +2514,7 @@ Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 854, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
 tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 32, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 512]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9677544
-No: 19  GFLOPS: 0.00/103.84     result: Traceback (most recent call last):
+No: 19  GFLOPS: 0.00/109.34     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 721, in __call__
     yield remote, remote.load_module(os.path.split(build_result.filename)[1])
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 685, in run_through_rpc
@@ -2602,7 +2602,7 @@ tvm._ffi.base.TVMError: Traceback (most recent call last):
   15: _PyEval_EvalFrameDefault
   14: 0x0000000000537c30
   13: _PyObject_FastCallKeywords
-  12: 0x00007f4b0f68efa2
+  12: 0x00007f4585a94fa2
   11: _ctypes_callproc
   10: ffi_call
   9: ffi_call_unix64
@@ -2667,7 +2667,7 @@ Traceback (most recent call last):
   21: _PyFunction_FastCallKeywords
   20: _PyEval_EvalFrameDefault
   19: _PyFunction_FastCall      [(&#39;tile_f&#39;, [-1, 8, 2, 16]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 1]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6390073
-No: 20  GFLOPS: 144.80/144.80   result: MeasureResult(costs=(0.0015987917100000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3352305889129639, timestamp=1648924667.7228255)      [(&#39;tile_f&#39;, [-1, 1, 4, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 4, 1]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9881539
+No: 20  GFLOPS: 141.51/141.51   result: MeasureResult(costs=(0.00163588642,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.414628505706787, timestamp=1648932970.4464226)       [(&#39;tile_f&#39;, [-1, 1, 4, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 4, 1]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9881539
 </pre></div>
 </div>
 <p>Finally we can inspect the best config from log file, check correctness,
@@ -2706,7 +2706,7 @@ and measure running time.</p>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Best config:
 [(&#39;tile_f&#39;, [-1, 1, 4, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 4, 1]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9881539
-Time cost of this operator: 0.001954
+Time cost of this operator: 0.002028
 </pre></div>
 </div>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autotvm-tune-conv2d-cuda-py">
diff --git a/docs/how_to/work_with_microtvm/micro_autotune.html b/docs/how_to/work_with_microtvm/micro_autotune.html
index 68dc4a2..9a7c3dc 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -553,10 +553,10 @@ the tuned operator.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
 Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs
 ---------                                     ---                                           --------  -------  -----              ------  -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  313.2     98.712   (1, 2, 10, 10, 3)  2       1
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.144     0.991    (1, 6, 10, 10)     1       1
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.944     0.298    (1, 1, 10, 10, 3)  1       1
-Total_time                                    -                                             317.288   -        -                  -       -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  312.5     98.719   (1, 2, 10, 10, 3)  2       1
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.153     0.996    (1, 6, 10, 10)     1       1
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.901     0.285    (1, 1, 10, 10, 3)  1       1
+Total_time                                    -                                             316.554   -        -                  -       -
 </pre></div>
 </div>
 </div>
@@ -608,10 +608,10 @@ Total_time                                    -
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
 Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs
 ---------                                     ---                                           --------  -------  -----              ------  -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  95.25     97.289   (1, 6, 10, 10, 1)  2       1
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.736     1.773    (1, 6, 10, 10)     1       1
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.918     0.938    (1, 1, 10, 10, 3)  1       1
-Total_time                                    -                                             97.905    -        -                  -       -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  123.6     97.871   (1, 6, 10, 10, 1)  2       1
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.741     1.379    (1, 6, 10, 10)     1       1
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.947     0.75     (1, 1, 10, 10, 3)  1       1
+Total_time                                    -                                             126.289   -        -                  -       -
 </pre></div>
 </div>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/sg_execution_times.html b/docs/how_to/work_with_microtvm/sg_execution_times.html
index 1bb50a0..7837319 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -300,13 +300,13 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:44.863</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>00:44.217</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
 <ul class="simple">
-<li><p><strong>00:40.719</strong>: <a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></li>
-<li><p><strong>00:03.546</strong>: <a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></li>
-<li><p><strong>00:00.200</strong>: <a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
-<li><p><strong>00:00.200</strong>: <a class="reference internal" href="micro_reference_vm.html#sphx-glr-how-to-work-with-microtvm-micro-reference-vm-py"><span class="std std-ref">microTVM Reference Virtual Machines</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_reference_vm.py</span></code>)</p></li>
-<li><p><strong>00:00.197</strong>: <a class="reference internal" href="micro_tvmc.html#sphx-glr-how-to-work-with-microtvm-micro-tvmc-py"><span class="std std-ref">Executing a Tiny Model with TVMC Micro</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tvmc.py</span></code>)</p></li>
+<li><p><strong>00:40.189</strong>: <a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></li>
+<li><p><strong>00:03.475</strong>: <a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></li>
+<li><p><strong>00:00.190</strong>: <a class="reference internal" href="micro_tvmc.html#sphx-glr-how-to-work-with-microtvm-micro-tvmc-py"><span class="std std-ref">Executing a Tiny Model with TVMC Micro</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tvmc.py</span></code>)</p></li>
+<li><p><strong>00:00.183</strong>: <a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
+<li><p><strong>00:00.179</strong>: <a class="reference internal" href="micro_reference_vm.html#sphx-glr-how-to-work-with-microtvm-micro-reference-vm-py"><span class="std std-ref">microTVM Reference Virtual Machines</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_reference_vm.py</span></code>)</p></li>
 </ul>
 </div>
 
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index d02f443..c47adaf 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -300,11 +300,11 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:08.912</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:10.469</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
 <ul class="simple">
-<li><p><strong>00:06.761</strong>: <a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></li>
-<li><p><strong>00:01.947</strong>: <a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></li>
-<li><p><strong>00:00.205</strong>: <a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></li>
+<li><p><strong>00:08.368</strong>: <a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></li>
+<li><p><strong>00:01.899</strong>: <a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></li>
+<li><p><strong>00:00.202</strong>: <a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></li>
 </ul>
 </div>
 
diff --git a/docs/how_to/work_with_schedules/sg_execution_times.html b/docs/how_to/work_with_schedules/sg_execution_times.html
index 23e683c..f59f0c5 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -300,16 +300,16 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:05.449</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:05.564</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
 <ul class="simple">
-<li><p><strong>00:02.002</strong>: <a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></li>
-<li><p><strong>00:01.130</strong>: <a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></li>
-<li><p><strong>00:00.699</strong>: <a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></li>
-<li><p><strong>00:00.681</strong>: <a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></li>
-<li><p><strong>00:00.295</strong>: <a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></li>
-<li><p><strong>00:00.227</strong>: <a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></li>
-<li><p><strong>00:00.211</strong>: <a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></li>
-<li><p><strong>00:00.204</strong>: <a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></li>
+<li><p><strong>00:02.021</strong>: <a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></li>
+<li><p><strong>00:01.203</strong>: <a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></li>
+<li><p><strong>00:00.707</strong>: <a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></li>
+<li><p><strong>00:00.698</strong>: <a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></li>
+<li><p><strong>00:00.284</strong>: <a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></li>
+<li><p><strong>00:00.223</strong>: <a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></li>
+<li><p><strong>00:00.216</strong>: <a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></li>
+<li><p><strong>00:00.211</strong>: <a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></li>
 </ul>
 </div>
 
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index a7c8a79..d6f0c4a 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -548,8 +548,8 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
              B: Buffer(B_2: Pointer(float32), float32, [32768], []),
              C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
   buffer_map = {A_1: A, B_1: B, C_1: C} {
-  attr [IterVar(i: int32, (nullptr), &quot;DataPar&quot;, &quot;&quot;)] &quot;pragma_import_llvm&quot; = &quot;; ModuleID = &#39;/tmp/tmpjiuprxtx/input0.cc&#39;
-source_filename = &quot;/tmp/tmpjiuprxtx/input0.cc&quot;
+  attr [IterVar(i: int32, (nullptr), &quot;DataPar&quot;, &quot;&quot;)] &quot;pragma_import_llvm&quot; = &quot;; ModuleID = &#39;/tmp/tmp5a5wc43t/input0.cc&#39;
+source_filename = &quot;/tmp/tmp5a5wc43t/input0.cc&quot;
 target datalayout = &quot;e-m:e-i64:64-f80:128-n8:16:32:64-S128&quot;
 target triple = &quot;x86_64-pc-linux-gnu&quot;
 
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index 2006671..9781658 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1713,7 +1713,7 @@ Can be the a function or the function name.</p></li>
 
 <dl class="py function">
 <dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
 <dd><p>THIS API IS DEPRECATED.</p>
 <p>Run auto scheduling search for a task.</p>
 <dl class="field-list simple">
@@ -1750,7 +1750,7 @@ the initial naive schedule (state).</p>
 
 <dl class="py class">
 <dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
 <dd><p>The search policy that searches in a hierarchical search space defined by sketches.
 The policy randomly samples programs from the space defined by sketches and use evolutionary
 search to fine-tune them.</p>
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 4090833..063b952 100644
--- a/docs/reference/api/typedoc/classes/bytestreamreader.html
+++ b/docs/reference/api/typedoc/classes/bytestreamreader.html
@@ -119,7 +119,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
 					<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -151,7 +151,7 @@
 					<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -168,7 +168,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index e165cdb..2634aff 100644
--- a/docs/reference/api/typedoc/classes/cachedcallstack.html
+++ b/docs/reference/api/typedoc/classes/cachedcallstack.html
@@ -144,7 +144,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L223">memory.ts:223</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L223">memory.ts:223</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
 					<div class="tsd-signature tsd-kind-icon">temp<wbr>Args<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol">&lt;</span><a href="../interfaces/disposable.html" class="tsd-signature-type">Disposable</a><span class="tsd-signature-symbol">&gt;</span><span class="tsd-signature-symbol"> = []</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L208">memory.ts:208</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L208">memory.ts:208</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -194,7 +194,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L312">memory.ts:312</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L312">memory.ts:312</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -226,7 +226,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L284">memory.ts:284</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L284">memory.ts:284</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L388">memory.ts:388</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L388">memory.ts:388</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -300,7 +300,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L376">memory.ts:376</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L376">memory.ts:376</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -340,7 +340,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L267">memory.ts:267</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L267">memory.ts:267</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -373,7 +373,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L243">memory.ts:243</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L243">memory.ts:243</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -390,7 +390,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L321">memory.ts:321</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L321">memory.ts:321</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -422,7 +422,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L252">memory.ts:252</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L252">memory.ts:252</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -444,7 +444,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L359">memory.ts:359</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L359">memory.ts:359</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -470,7 +470,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L342">memory.ts:342</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L342">memory.ts:342</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -496,7 +496,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L350">memory.ts:350</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L350">memory.ts:350</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -522,7 +522,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L326">memory.ts:326</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L326">memory.ts:326</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -548,7 +548,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L363">memory.ts:363</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L363">memory.ts:363</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -574,7 +574,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L346">memory.ts:346</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L346">memory.ts:346</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -600,7 +600,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/memory.ts#L334">memory.ts:334</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/memory.ts#L334">memory.ts:334</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/dldatatype.html b/docs/reference/api/typedoc/classes/dldatatype.html
index 3e70351..49403e4 100644
--- a/docs/reference/api/typedoc/classes/dldatatype.html
+++ b/docs/reference/api/typedoc/classes/dldatatype.html
@@ -119,7 +119,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/runtime.ts#L262">runtime.ts:262</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
 					<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/264ee08ae/web/src/runtime.ts#L260">runtime.ts:260</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/daa068979/web/src/runtime.ts#L260">runtime.ts:260</a></li>
... 2504 lines suppressed ...