You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2022/05/26 02:07:50 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@814f5501bf7d65f759135d214572388b0ddadefc)
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 37f2394cd deploying docs (apache/tvm@814f5501bf7d65f759135d214572388b0ddadefc)
37f2394cd is described below
commit 37f2394cdfab627e2c0a8565ea5074d4bf0fbef3
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Thu May 26 02:07:39 2022 +0000
deploying docs (apache/tvm@814f5501bf7d65f759135d214572388b0ddadefc)
---
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.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 | 5 -
.../compile_models/sg_execution_times.rst.txt | 22 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 18 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 2 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 10 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 16 +-
.../tune_conv2d_layer_cuda.rst.txt | 1038 +++++++--
.../tune_network_cuda.rst.txt | 2 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 88 +-
.../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 | 2 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 6 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 9 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 56 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 26 +-
.../tutorial/tensor_expr_get_started.rst.txt | 45 +-
docs/commit_hash | 2 +-
docs/genindex.html | 2 +
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 82 +-
docs/how_to/compile_models/from_paddle.html | 2 +-
docs/how_to/compile_models/from_pytorch.html | 6 +-
docs/how_to/compile_models/from_tensorflow.html | 1 -
docs/how_to/compile_models/sg_execution_times.html | 22 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 20 +-
docs/how_to/deploy_models/deploy_prequantized.html | 9 +-
.../deploy_models/deploy_prequantized_tflite.html | 4 +-
docs/how_to/deploy_models/deploy_quantized.html | 2 +-
docs/how_to/deploy_models/deploy_ssd_gluoncv.html | 38 +-
docs/how_to/deploy_models/sg_execution_times.html | 18 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
docs/how_to/extend_tvm/sg_execution_times.html | 10 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 1038 +++++++--
.../tune_with_autoscheduler/tune_network_cuda.html | 2 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 88 +-
.../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 | 2 +-
docs/objects.inv | Bin 22384 -> 22395 bytes
.../api/doxygen/affine__type_8h__incl.svg | 1112 +++++----
docs/reference/api/doxygen/algorithm_8h__incl.svg | 976 ++++----
docs/reference/api/doxygen/algorithms_8h.html | 2 +-
docs/reference/api/doxygen/algorithms_8h__incl.svg | 2045 ++++++++---------
docs/reference/api/doxygen/analyzer_8h.html | 2 +-
.../api/doxygen/analyzer_8h__dep__incl.svg | 644 +++---
docs/reference/api/doxygen/analyzer_8h__incl.svg | 2084 ++++++++---------
docs/reference/api/doxygen/annotation_8h.html | 2 +-
docs/reference/api/doxygen/annotation_8h__incl.svg | 1914 ++++++++--------
docs/reference/api/doxygen/aot__executor_8h.html | 2 +-
.../api/doxygen/aot__executor_8h__incl.svg | 102 +-
.../api/doxygen/aot__executor_8h_source.html | 2 +-
.../api/doxygen/apply__history__best_8h__incl.svg | 884 ++++----
docs/reference/api/doxygen/arg__info_8h.html | 2 +-
.../api/doxygen/arg__info_8h__dep__incl.svg | 108 +-
docs/reference/api/doxygen/arg__info_8h__incl.svg | 2273 +++++++++----------
docs/reference/api/doxygen/array_8h__dep__incl.svg | 772 +++----
docs/reference/api/doxygen/array_8h__incl.svg | 126 +-
.../api/doxygen/array__utils_8h__dep__incl.svg | 44 +-
.../api/doxygen/array__utils_8h__incl.svg | 804 +++----
.../doxygen/attr__registry__map_8h__dep__incl.svg | 536 ++---
.../api/doxygen/attr__registry__map_8h__incl.svg | 262 +--
docs/reference/api/doxygen/auto__schedule_8h.html | 2 +-
.../api/doxygen/auto__schedule_8h__incl.svg | 2104 ++++++++---------
.../doxygen/auto__scheduler_2cost__model_8h.html | 2 +-
.../auto__scheduler_2cost__model_8h__incl.svg | 1982 ++++++++--------
.../api/doxygen/auto__scheduler_2feature_8h.html | 2 +-
.../doxygen/auto__scheduler_2feature_8h__incl.svg | 2146 +++++++++---------
docs/reference/api/doxygen/autodiff_8h.html | 2 +-
docs/reference/api/doxygen/autodiff_8h__incl.svg | 2288 +++++++++----------
docs/reference/api/doxygen/bias__add_8h__incl.svg | 876 ++++----
docs/reference/api/doxygen/bitserial_8h__incl.svg | 810 +++----
docs/reference/api/doxygen/block__scope_8h.html | 2 +-
.../api/doxygen/block__scope_8h__dep__incl.svg | 128 +-
.../api/doxygen/block__scope_8h__incl.svg | 2036 ++++++++---------
docs/reference/api/doxygen/bound_8h.html | 2 +-
docs/reference/api/doxygen/bound_8h__dep__incl.svg | 628 +++---
docs/reference/api/doxygen/bound_8h__incl.svg | 2115 +++++++++---------
.../api/doxygen/broadcast_8h__dep__incl.svg | 92 +-
docs/reference/api/doxygen/broadcast_8h__incl.svg | 928 ++++----
.../reference/api/doxygen/buffer_8h__dep__incl.svg | 640 +++---
docs/reference/api/doxygen/buffer_8h__incl.svg | 1140 +++++-----
.../api/doxygen/builder_8h__dep__incl.svg | 40 +-
docs/reference/api/doxygen/builder_8h__incl.svg | 864 +++----
docs/reference/api/doxygen/builtin_8h.html | 2 +-
.../api/doxygen/builtin_8h__dep__incl.svg | 108 +-
docs/reference/api/doxygen/builtin_8h__incl.svg | 1992 ++++++++---------
docs/reference/api/doxygen/builtin__fp16_8h.html | 2 +-
.../api/doxygen/builtin__fp16_8h__incl.svg | 78 +-
docs/reference/api/doxygen/bytecode_8h.html | 2 +-
.../api/doxygen/bytecode_8h__dep__incl.svg | 24 +-
docs/reference/api/doxygen/bytecode_8h__incl.svg | 194 +-
docs/reference/api/doxygen/c__backend__api_8h.html | 2 +-
.../api/doxygen/c__backend__api_8h__dep__incl.svg | 40 +-
.../api/doxygen/c__backend__api_8h__incl.svg | 58 +-
.../api/doxygen/c__backend__api_8h_source.html | 4 +-
docs/reference/api/doxygen/c__runtime__api_8h.html | 20 +-
.../api/doxygen/c__runtime__api_8h__dep__incl.svg | 652 +++---
.../api/doxygen/c__runtime__api_8h__incl.svg | 48 +-
.../api/doxygen/c__runtime__api_8h_source.html | 95 +-
docs/reference/api/doxygen/call_8h.html | 2 +-
docs/reference/api/doxygen/call_8h__incl.svg | 1914 ++++++++--------
docs/reference/api/doxygen/closure_8h.html | 2 +-
.../api/doxygen/closure_8h__dep__incl.svg | 20 +-
docs/reference/api/doxygen/closure_8h__incl.svg | 294 ++-
docs/reference/api/doxygen/codegen_8h__incl.svg | 904 ++++----
.../doxygen/compilation__config_8h__dep__incl.svg | 20 +-
.../api/doxygen/compilation__config_8h__incl.svg | 836 +++----
docs/reference/api/doxygen/compute__dag_8h.html | 2 +-
.../api/doxygen/compute__dag_8h__dep__incl.svg | 76 +-
.../api/doxygen/compute__dag_8h__incl.svg | 2065 ++++++++---------
.../api/doxygen/constant__utils_8h__dep__incl.svg | 168 +-
.../api/doxygen/constant__utils_8h__incl.svg | 984 ++++----
.../api/doxygen/crt_2packed__func_8h.html | 2 +-
.../doxygen/crt_2packed__func_8h__dep__incl.svg | 12 +-
.../api/doxygen/crt_2packed__func_8h__incl.svg | 282 ++-
.../api/doxygen/crt_2packed__func_8h_source.html | 10 +-
.../reference/api/doxygen/cublas_8h__dep__incl.svg | 20 +-
docs/reference/api/doxygen/cublas_8h__incl.svg | 840 +++----
.../api/doxygen/cuda_2dense_8h__dep__incl.svg | 12 +-
.../reference/api/doxygen/cuda_2dense_8h__incl.svg | 952 ++++----
.../api/doxygen/cuda_2injective_8h__dep__incl.svg | 12 +-
.../api/doxygen/cuda_2injective_8h__incl.svg | 992 ++++----
.../api/doxygen/cuda_2pooling_8h__dep__incl.svg | 12 +-
.../api/doxygen/cuda_2pooling_8h__incl.svg | 964 ++++----
.../api/doxygen/cuda_2reduction_8h__dep__incl.svg | 12 +-
.../api/doxygen/cuda_2reduction_8h__incl.svg | 992 ++++----
.../api/doxygen/cuda_2softmax_8h__dep__incl.svg | 12 +-
.../api/doxygen/cuda_2softmax_8h__incl.svg | 992 ++++----
docs/reference/api/doxygen/data__layout_8h.html | 2 +-
.../api/doxygen/data__layout_8h__dep__incl.svg | 112 +-
.../api/doxygen/data__layout_8h__incl.svg | 2057 ++++++++---------
docs/reference/api/doxygen/data__type_8h.html | 2 +-
.../api/doxygen/data__type_8h__dep__incl.svg | 716 +++---
docs/reference/api/doxygen/data__type_8h__incl.svg | 114 +-
.../api/doxygen/data__type_8h_source.html | 4 +-
.../api/doxygen/database_8h__dep__incl.svg | 20 +-
docs/reference/api/doxygen/database_8h__incl.svg | 948 ++++----
.../api/doxygen/dataflow__matcher_8h__incl.svg | 964 ++++----
.../doxygen/dataflow__pattern_8h__dep__incl.svg | 24 +-
.../api/doxygen/dataflow__pattern_8h__incl.svg | 984 ++++----
.../dataflow__pattern__functor_8h__dep__incl.svg | 12 +-
.../dataflow__pattern__functor_8h__incl.svg | 964 ++++----
.../doxygen/detail_2broadcast_8h__dep__incl.svg | 104 +-
.../api/doxygen/detail_2broadcast_8h__incl.svg | 900 ++++----
.../api/doxygen/detail_2extern_8h__dep__incl.svg | 40 +-
.../api/doxygen/detail_2extern_8h__incl.svg | 864 +++----
docs/reference/api/doxygen/device__api_8h.html | 2 +-
.../api/doxygen/device__api_8h__dep__incl.svg | 64 +-
.../reference/api/doxygen/device__api_8h__incl.svg | 1294 ++++++-----
.../api/doxygen/device__api_8h_source.html | 12 +-
.../api/doxygen/device__copy_8h__incl.svg | 872 ++++----
.../api/doxygen/diagnostic_8h__dep__incl.svg | 536 ++---
docs/reference/api/doxygen/diagnostic_8h__incl.svg | 820 +++----
docs/reference/api/doxygen/dilate_8h__incl.svg | 840 +++----
.../reference/api/doxygen/driver__api_8h__incl.svg | 1008 ++++-----
docs/reference/api/doxygen/einsum_8h__incl.svg | 884 ++++----
docs/reference/api/doxygen/elemwise_8h.html | 2 +-
.../api/doxygen/elemwise_8h__dep__incl.svg | 48 +-
docs/reference/api/doxygen/elemwise_8h__incl.svg | 1930 ++++++++--------
.../api/doxygen/env__func_8h__dep__incl.svg | 556 ++---
docs/reference/api/doxygen/env__func_8h__incl.svg | 930 ++++----
docs/reference/api/doxygen/error_8h__dep__incl.svg | 524 ++---
docs/reference/api/doxygen/error_8h__incl.svg | 844 +++----
.../api/doxygen/executable_8h__dep__incl.svg | 12 +-
docs/reference/api/doxygen/executable_8h__incl.svg | 762 +++----
.../api/doxygen/executable_8h_source.html | 2 +-
docs/reference/api/doxygen/executor_8h__incl.svg | 884 ++++----
.../api/doxygen/extracted__task_8h__incl.svg | 888 ++++----
.../api/doxygen/feature__extractor_8h__incl.svg | 864 +++----
docs/reference/api/doxygen/flatten_8h__incl.svg | 904 ++++----
docs/reference/api/doxygen/func__registry_8h.html | 2 +-
.../api/doxygen/func__registry_8h__dep__incl.svg | 28 +-
.../api/doxygen/func__registry_8h__incl.svg | 94 +-
docs/reference/api/doxygen/functor_8h.html | 2 +-
.../api/doxygen/functor_8h__dep__incl.svg | 660 +++---
docs/reference/api/doxygen/functor_8h__incl.svg | 208 +-
docs/reference/api/doxygen/fuse_8h__dep__incl.svg | 156 +-
docs/reference/api/doxygen/fuse_8h__incl.svg | 804 +++----
.../api/doxygen/generic_2default_8h__incl.svg | 992 ++++----
.../api/doxygen/generic_2extern_8h__dep__incl.svg | 24 +-
.../api/doxygen/generic_2extern_8h__incl.svg | 980 ++++----
.../doxygen/generic_2injective_8h__dep__incl.svg | 32 +-
.../api/doxygen/generic_2injective_8h__incl.svg | 992 ++++----
docs/reference/api/doxygen/generic__func_8h.html | 2 +-
.../api/doxygen/generic__func_8h__dep__incl.svg | 196 +-
.../api/doxygen/generic__func_8h__incl.svg | 2347 +++++++++----------
.../api/doxygen/generic__func_8h_source.html | 2 +-
docs/reference/api/doxygen/globals_t.html | 3 -
docs/reference/api/doxygen/globals_type.html | 3 -
docs/reference/api/doxygen/graph__executor_8h.html | 2 +-
.../api/doxygen/graph__executor_8h__incl.svg | 338 ++-
.../api/doxygen/graph__executor_8h_source.html | 2 +-
docs/reference/api/doxygen/greedy_8h.html | 2 +-
docs/reference/api/doxygen/greedy_8h__incl.svg | 2161 +++++++++---------
docs/reference/api/doxygen/image_8h__incl.svg | 810 +++----
.../api/doxygen/index__map_8h__dep__incl.svg | 568 ++---
docs/reference/api/doxygen/index__map_8h__incl.svg | 1136 +++++-----
.../api/doxygen/instruction_8h__dep__incl.svg | 148 +-
.../reference/api/doxygen/instruction_8h__incl.svg | 924 ++++----
.../api/doxygen/instrument_8h__dep__incl.svg | 516 ++---
docs/reference/api/doxygen/instrument_8h__incl.svg | 936 ++++----
docs/reference/api/doxygen/int__set_8h.html | 2 +-
.../api/doxygen/int__set_8h__dep__incl.svg | 660 +++---
docs/reference/api/doxygen/int__set_8h__incl.svg | 2020 ++++++++---------
docs/reference/api/doxygen/int__solver_8h.html | 2 +-
.../reference/api/doxygen/int__solver_8h__incl.svg | 2101 ++++++++---------
.../reference/api/doxygen/interpreter_8h__incl.svg | 864 +++----
.../api/doxygen/ir_2adt_8h__dep__incl.svg | 620 ++---
docs/reference/api/doxygen/ir_2adt_8h__incl.svg | 1144 +++++-----
.../api/doxygen/ir_2attrs_8h__dep__incl.svg | 580 ++---
docs/reference/api/doxygen/ir_2attrs_8h__incl.svg | 1138 +++++-----
.../reference/api/doxygen/ir_2attrs_8h_source.html | 4 +-
.../api/doxygen/ir_2expr_8h__dep__incl.svg | 604 ++---
docs/reference/api/doxygen/ir_2expr_8h__incl.svg | 1084 +++++----
docs/reference/api/doxygen/ir_2expr_8h_source.html | 4 +-
docs/reference/api/doxygen/ir_2function_8h.html | 2 +-
.../api/doxygen/ir_2function_8h__dep__incl.svg | 628 +++---
.../api/doxygen/ir_2function_8h__incl.svg | 1954 ++++++++--------
.../api/doxygen/ir_2module_8h__dep__incl.svg | 616 ++---
docs/reference/api/doxygen/ir_2module_8h__incl.svg | 832 +++----
.../reference/api/doxygen/ir_2op_8h__dep__incl.svg | 576 ++---
docs/reference/api/doxygen/ir_2op_8h__incl.svg | 880 ++++----
.../api/doxygen/ir_2span_8h__dep__incl.svg | 708 +++---
docs/reference/api/doxygen/ir_2span_8h__incl.svg | 1000 ++++-----
.../api/doxygen/ir_2transform_8h__dep__incl.svg | 536 ++---
.../api/doxygen/ir_2transform_8h__incl.svg | 872 ++++----
docs/reference/api/doxygen/ir_2type_8h.html | 2 +-
.../api/doxygen/ir_2type_8h__dep__incl.svg | 696 +++---
docs/reference/api/doxygen/ir_2type_8h__incl.svg | 1028 +++++----
.../api/doxygen/iter__affine__map_8h__incl.svg | 836 +++----
.../api/doxygen/libtorch__runtime_8h__incl.svg | 790 ++++---
.../api/doxygen/local__response__norm_8h__incl.svg | 820 +++----
docs/reference/api/doxygen/loop__state_8h.html | 2 +-
.../api/doxygen/loop__state_8h__dep__incl.svg | 88 +-
.../reference/api/doxygen/loop__state_8h__incl.svg | 2087 ++++++++---------
docs/reference/api/doxygen/map_8h__dep__incl.svg | 632 +++---
docs/reference/api/doxygen/map_8h__incl.svg | 152 +-
docs/reference/api/doxygen/mapping_8h__incl.svg | 820 +++----
docs/reference/api/doxygen/measure_8h.html | 2 +-
.../api/doxygen/measure_8h__dep__incl.svg | 48 +-
docs/reference/api/doxygen/measure_8h__incl.svg | 1908 ++++++++--------
.../doxygen/measure__callback_8h__dep__incl.svg | 12 +-
.../api/doxygen/measure__callback_8h__incl.svg | 788 +++----
docs/reference/api/doxygen/measure__record_8h.html | 2 +-
.../api/doxygen/measure__record_8h__incl.svg | 1726 +++++++-------
docs/reference/api/doxygen/memory__manager_8h.html | 2 +-
.../api/doxygen/memory__manager_8h__dep__incl.svg | 12 +-
.../api/doxygen/memory__manager_8h__incl.svg | 954 ++++----
docs/reference/api/doxygen/memory__pools_8h.html | 2 +-
.../api/doxygen/memory__pools_8h__dep__incl.svg | 44 +-
.../api/doxygen/memory__pools_8h__incl.svg | 2303 +++++++++----------
.../meta__schedule_2cost__model_8h__dep__incl.svg | 12 +-
.../meta__schedule_2cost__model_8h__incl.svg | 868 +++----
docs/reference/api/doxygen/metadata_8h.html | 2 +-
docs/reference/api/doxygen/metadata_8h__incl.svg | 2109 ++++++++---------
.../api/doxygen/metadata__base_8h__dep__incl.svg | 12 +-
.../api/doxygen/metadata__base_8h__incl.svg | 1110 +++++----
docs/reference/api/doxygen/metadata__types_8h.html | 2 +-
.../api/doxygen/metadata__types_8h__dep__incl.svg | 20 +-
.../api/doxygen/metadata__types_8h__incl.svg | 78 +-
.../api/doxygen/mutator_8h__dep__incl.svg | 32 +-
docs/reference/api/doxygen/mutator_8h__incl.svg | 776 +++----
.../api/doxygen/namespacemembers_func_s.html | 6 +-
docs/reference/api/doxygen/namespacemembers_s.html | 6 +-
.../api/doxygen/ndarray_8h__dep__incl.svg | 580 ++---
docs/reference/api/doxygen/ndarray_8h__incl.svg | 542 +++--
docs/reference/api/doxygen/ndarray_8h_source.html | 6 +-
docs/reference/api/doxygen/nn_2bnn_8h__incl.svg | 912 ++++----
.../api/doxygen/nn_2dense_8h__dep__incl.svg | 24 +-
docs/reference/api/doxygen/nn_2dense_8h__incl.svg | 820 +++----
.../reference/api/doxygen/nn_2pooling_8h__incl.svg | 964 ++++----
.../reference/api/doxygen/nn_2softmax_8h__incl.svg | 940 ++++----
docs/reference/api/doxygen/node_8h.html | 2 +-
docs/reference/api/doxygen/node_8h__dep__incl.svg | 676 +++---
docs/reference/api/doxygen/node_8h__incl.svg | 1620 +++++++-------
docs/reference/api/doxygen/object_8h.html | 2 +-
.../reference/api/doxygen/object_8h__dep__incl.svg | 748 +++----
docs/reference/api/doxygen/object_8h__incl.svg | 130 +-
docs/reference/api/doxygen/on__device_8h__incl.svg | 872 ++++----
.../api/doxygen/op__strategy_8h__incl.svg | 1020 ++++-----
.../api/doxygen/operation_8h__dep__incl.svg | 592 ++---
docs/reference/api/doxygen/operation_8h__incl.svg | 806 +++----
.../api/doxygen/optional_8h__dep__incl.svg | 624 +++---
docs/reference/api/doxygen/optional_8h__incl.svg | 136 +-
docs/reference/api/doxygen/packed__func_8h.html | 4 +-
.../api/doxygen/packed__func_8h__dep__incl.svg | 544 ++---
.../api/doxygen/packed__func_8h__incl.svg | 792 ++++---
.../api/doxygen/packed__func_8h_source.html | 34 +-
docs/reference/api/doxygen/pad__utils_8h.html | 2 +-
.../api/doxygen/pad__utils_8h__dep__incl.svg | 12 +-
docs/reference/api/doxygen/pad__utils_8h__incl.svg | 1976 ++++++++--------
docs/reference/api/doxygen/page__allocator_8h.html | 2 +-
.../api/doxygen/page__allocator_8h__incl.svg | 92 +-
docs/reference/api/doxygen/papi_8h.html | 2 +-
docs/reference/api/doxygen/papi_8h__incl.svg | 822 ++++---
docs/reference/api/doxygen/parallel__for_8h.html | 2 +-
.../api/doxygen/parallel__for_8h__incl.svg | 98 +-
docs/reference/api/doxygen/parser_8h__incl.svg | 872 ++++----
docs/reference/api/doxygen/pattern_8h.html | 2 +-
docs/reference/api/doxygen/pattern_8h__incl.svg | 1998 ++++++++---------
.../api/doxygen/pattern__functor_8h__incl.svg | 896 ++++----
docs/reference/api/doxygen/platform_8h.html | 2 +-
.../api/doxygen/platform_8h__dep__incl.svg | 20 +-
docs/reference/api/doxygen/platform_8h__incl.svg | 72 +-
.../api/doxygen/postproc_8h__dep__incl.svg | 32 +-
docs/reference/api/doxygen/postproc_8h__incl.svg | 776 +++----
docs/reference/api/doxygen/profiling_8h.html | 2 +-
.../api/doxygen/profiling_8h__dep__incl.svg | 12 +-
docs/reference/api/doxygen/profiling_8h__incl.svg | 1488 ++++++------
docs/reference/api/doxygen/random_8h.html | 2 +-
docs/reference/api/doxygen/random_8h__incl.svg | 1896 ++++++++--------
.../api/doxygen/ravel__unravel_8h__dep__incl.svg | 84 +-
.../api/doxygen/ravel__unravel_8h__incl.svg | 808 +++----
docs/reference/api/doxygen/reduce_8h.html | 2 +-
docs/reference/api/doxygen/reduce_8h__incl.svg | 1914 ++++++++--------
.../api/doxygen/reduction_8h__dep__incl.svg | 40 +-
docs/reference/api/doxygen/reduction_8h__incl.svg | 964 ++++----
.../api/doxygen/reflection_8h__dep__incl.svg | 700 +++---
docs/reference/api/doxygen/reflection_8h__incl.svg | 918 ++++----
.../api/doxygen/registry_8h__dep__incl.svg | 580 ++---
docs/reference/api/doxygen/registry_8h__incl.svg | 816 ++++---
.../api/doxygen/relay_2adt_8h__dep__incl.svg | 36 +-
docs/reference/api/doxygen/relay_2adt_8h__incl.svg | 944 ++++----
.../api/doxygen/relay_2analysis_8h__incl.svg | 948 ++++----
.../api/doxygen/relay_2attrs_2debug_8h.html | 2 +-
.../api/doxygen/relay_2attrs_2debug_8h__incl.svg | 1878 ++++++++--------
.../api/doxygen/relay_2attrs_2memory_8h__incl.svg | 996 ++++-----
.../api/doxygen/relay_2attrs_2nn_8h__incl.svg | 810 +++----
.../relay_2attrs_2transform_8h__dep__incl.svg | 20 +-
.../doxygen/relay_2attrs_2transform_8h__incl.svg | 976 ++++----
.../reference/api/doxygen/relay_2attrs_2vm_8h.html | 2 +-
.../api/doxygen/relay_2attrs_2vm_8h__incl.svg | 1894 ++++++++--------
docs/reference/api/doxygen/relay_2base_8h.html | 2 +-
.../api/doxygen/relay_2base_8h__dep__incl.svg | 296 +--
.../reference/api/doxygen/relay_2base_8h__incl.svg | 2034 ++++++++---------
.../api/doxygen/relay_2expr_8h__dep__incl.svg | 212 +-
.../reference/api/doxygen/relay_2expr_8h__incl.svg | 1000 ++++-----
.../api/doxygen/relay_2expr__functor_8h__incl.svg | 900 ++++----
.../api/doxygen/relay_2feature_8h__incl.svg | 860 +++----
.../api/doxygen/relay_2function_8h__dep__incl.svg | 36 +-
.../api/doxygen/relay_2function_8h__incl.svg | 920 ++++----
.../api/doxygen/relay_2op_8h__dep__incl.svg | 36 +-
docs/reference/api/doxygen/relay_2op_8h__incl.svg | 972 ++++----
.../relay_2op__attr__types_8h__dep__incl.svg | 28 +-
.../doxygen/relay_2op__attr__types_8h__incl.svg | 1004 ++++-----
.../api/doxygen/relay_2qnn_2attrs_8h.html | 2 +-
.../api/doxygen/relay_2qnn_2attrs_8h__incl.svg | 1914 ++++++++--------
.../api/doxygen/relay_2qnn_2transform_8h.html | 2 +-
.../api/doxygen/relay_2qnn_2transform_8h__incl.svg | 2132 +++++++++---------
.../api/doxygen/relay_2transform_8h__dep__incl.svg | 12 +-
.../api/doxygen/relay_2transform_8h__incl.svg | 920 ++++----
.../api/doxygen/relay_2type_8h__dep__incl.svg | 240 +-
.../reference/api/doxygen/relay_2type_8h__incl.svg | 988 ++++----
docs/reference/api/doxygen/reorg_8h__incl.svg | 1004 ++++-----
docs/reference/api/doxygen/repr__printer_8h.html | 2 +-
.../api/doxygen/repr__printer_8h__dep__incl.svg | 672 +++---
.../api/doxygen/repr__printer_8h__incl.svg | 238 +-
.../api/doxygen/rocblas_8h__dep__incl.svg | 12 +-
docs/reference/api/doxygen/rocblas_8h__incl.svg | 840 +++----
.../reference/api/doxygen/rocm_2dense_8h__incl.svg | 920 ++++----
.../api/doxygen/rocm_2injective_8h__incl.svg | 992 ++++----
.../api/doxygen/rocm_2pooling_8h__incl.svg | 968 ++++----
.../api/doxygen/rocm_2reduction_8h__incl.svg | 992 ++++----
.../api/doxygen/rocm_2softmax_8h__incl.svg | 992 ++++----
docs/reference/api/doxygen/runner_8h.html | 2 +-
.../reference/api/doxygen/runner_8h__dep__incl.svg | 76 +-
docs/reference/api/doxygen/runner_8h__incl.svg | 2171 +++++++++---------
.../runtime_2container_2adt_8h__dep__incl.svg | 616 ++---
.../doxygen/runtime_2container_2adt_8h__incl.svg | 136 +-
.../api/doxygen/runtime_2container_2base_8h.html | 2 +-
.../runtime_2container_2base_8h__dep__incl.svg | 724 +++---
.../doxygen/runtime_2container_2base_8h__incl.svg | 286 ++-
.../api/doxygen/runtime_2crt_2module_8h.html | 2 +-
.../doxygen/runtime_2crt_2module_8h__dep__incl.svg | 20 +-
.../api/doxygen/runtime_2crt_2module_8h__incl.svg | 122 +-
.../doxygen/runtime_2crt_2module_8h_source.html | 2 +-
.../api/doxygen/runtime_2debug_8h__incl.svg | 484 ++--
docs/reference/api/doxygen/runtime_2memory_8h.html | 2 +-
.../api/doxygen/runtime_2memory_8h__dep__incl.svg | 692 +++---
.../api/doxygen/runtime_2memory_8h__incl.svg | 202 +-
docs/reference/api/doxygen/runtime_2module_8h.html | 2 +-
.../api/doxygen/runtime_2module_8h__dep__incl.svg | 544 ++---
.../api/doxygen/runtime_2module_8h__incl.svg | 790 ++++---
docs/reference/api/doxygen/runtime_2vm_2vm_8h.html | 2 +-
.../api/doxygen/runtime_2vm_2vm_8h__incl.svg | 1646 +++++++-------
docs/reference/api/doxygen/runtime_8h__incl.svg | 884 ++++----
.../api/doxygen/schedule__pass_8h__dep__incl.svg | 136 +-
.../api/doxygen/schedule__pass_8h__incl.svg | 808 +++----
.../api/doxygen/schedule__rule_8h__dep__incl.svg | 32 +-
.../api/doxygen/schedule__rule_8h__incl.svg | 776 +++----
docs/reference/api/doxygen/search/all_13.js | 2 +-
docs/reference/api/doxygen/search/all_14.js | 2 +-
docs/reference/api/doxygen/search/all_15.js | 1 -
docs/reference/api/doxygen/search/functions_13.js | 2 +-
docs/reference/api/doxygen/search/typedefs_e.js | 1 -
docs/reference/api/doxygen/search__policy_8h.html | 2 +-
.../api/doxygen/search__policy_8h__dep__incl.svg | 12 +-
.../api/doxygen/search__policy_8h__incl.svg | 2140 +++++++++---------
.../api/doxygen/search__strategy_8h__dep__incl.svg | 56 +-
.../api/doxygen/search__strategy_8h__incl.svg | 864 +++----
docs/reference/api/doxygen/search__task_8h.html | 2 +-
.../api/doxygen/search__task_8h__dep__incl.svg | 60 +-
.../api/doxygen/search__task_8h__incl.svg | 1859 +++++++--------
docs/reference/api/doxygen/serialization_8h.html | 2 +-
.../api/doxygen/serialization_8h__incl.svg | 172 +-
docs/reference/api/doxygen/serializer_8h.html | 2 +-
.../api/doxygen/serializer_8h__dep__incl.svg | 580 ++---
docs/reference/api/doxygen/serializer_8h__incl.svg | 526 +++--
.../api/doxygen/shape__tuple_8h__dep__incl.svg | 548 ++---
.../api/doxygen/shape__tuple_8h__incl.svg | 136 +-
.../api/doxygen/source__map_8h__dep__incl.svg | 620 ++---
.../reference/api/doxygen/source__map_8h__incl.svg | 1082 +++++----
.../api/doxygen/space__generator_8h__dep__incl.svg | 32 +-
.../api/doxygen/space__generator_8h__incl.svg | 788 +++----
docs/reference/api/doxygen/state_8h__dep__incl.svg | 120 +-
docs/reference/api/doxygen/state_8h__incl.svg | 900 ++++----
docs/reference/api/doxygen/stmt_8h.html | 2 +-
docs/reference/api/doxygen/stmt_8h__dep__incl.svg | 596 ++---
docs/reference/api/doxygen/stmt_8h__incl.svg | 2002 ++++++++---------
docs/reference/api/doxygen/stmt__functor_8h.html | 2 +-
.../api/doxygen/stmt__functor_8h__dep__incl.svg | 12 +-
.../api/doxygen/stmt__functor_8h__incl.svg | 2358 ++++++++++----------
docs/reference/api/doxygen/strided__slice_8h.html | 2 +-
.../api/doxygen/strided__slice_8h__dep__incl.svg | 72 +-
.../api/doxygen/strided__slice_8h__incl.svg | 2124 +++++++++---------
docs/reference/api/doxygen/string_8h.html | 2 +-
.../reference/api/doxygen/string_8h__dep__incl.svg | 672 +++---
docs/reference/api/doxygen/string_8h__incl.svg | 406 ++--
.../doxygen/structural__equal_8h__dep__incl.svg | 644 +++---
.../api/doxygen/structural__equal_8h__incl.svg | 326 ++-
.../api/doxygen/structural__hash_8h__dep__incl.svg | 644 +++---
.../api/doxygen/structural__hash_8h__incl.svg | 580 +++--
docs/reference/api/doxygen/tag_8h.html | 2 +-
docs/reference/api/doxygen/tag_8h__incl.svg | 2086 ++++++++---------
.../reference/api/doxygen/target_8h__dep__incl.svg | 504 ++---
docs/reference/api/doxygen/target_8h__incl.svg | 896 ++++----
.../api/doxygen/target__info_8h__incl.svg | 1090 +++++----
.../api/doxygen/target__kind_8h__dep__incl.svg | 504 ++---
.../api/doxygen/target__kind_8h__incl.svg | 766 +++----
.../api/doxygen/task__scheduler_8h__incl.svg | 780 +++----
docs/reference/api/doxygen/te_2schedule_8h.html | 2 +-
.../api/doxygen/te_2schedule_8h__dep__incl.svg | 648 +++---
.../api/doxygen/te_2schedule_8h__incl.svg | 1832 +++++++--------
docs/reference/api/doxygen/tensor_8h.html | 2 +-
.../reference/api/doxygen/tensor_8h__dep__incl.svg | 664 +++---
docs/reference/api/doxygen/tensor_8h__incl.svg | 2051 ++++++++---------
docs/reference/api/doxygen/tensor__intrin_8h.html | 2 +-
.../api/doxygen/tensor__intrin_8h__dep__incl.svg | 640 +++---
.../api/doxygen/tensor__intrin_8h__incl.svg | 2109 ++++++++---------
.../api/doxygen/tensor__type_8h__dep__incl.svg | 248 +-
.../api/doxygen/tensor__type_8h__incl.svg | 1112 +++++----
.../api/doxygen/tensor__utils_8h__dep__incl.svg | 80 +-
.../api/doxygen/tensor__utils_8h__incl.svg | 808 +++----
.../api/doxygen/tir_2analysis_8h__dep__incl.svg | 176 +-
.../api/doxygen/tir_2analysis_8h__incl.svg | 940 ++++----
docs/reference/api/doxygen/tir_2expr_8h.html | 2 +-
.../api/doxygen/tir_2expr_8h__dep__incl.svg | 612 ++---
docs/reference/api/doxygen/tir_2expr_8h__incl.svg | 2036 +++++++++--------
.../api/doxygen/tir_2expr__functor_8h.html | 2 +-
.../doxygen/tir_2expr__functor_8h__dep__incl.svg | 20 +-
.../api/doxygen/tir_2expr__functor_8h__incl.svg | 2126 +++++++++---------
docs/reference/api/doxygen/tir_2function_8h.html | 2 +-
.../api/doxygen/tir_2function_8h__dep__incl.svg | 508 ++---
.../api/doxygen/tir_2function_8h__incl.svg | 2175 +++++++++---------
.../api/doxygen/tir_2op_8h__dep__incl.svg | 660 +++---
docs/reference/api/doxygen/tir_2op_8h__incl.svg | 880 ++++----
.../doxygen/tir_2op__attr__types_8h__dep__incl.svg | 184 +-
.../api/doxygen/tir_2op__attr__types_8h__incl.svg | 1096 +++++----
.../tir_2schedule_2schedule_8h__dep__incl.svg | 112 +-
.../doxygen/tir_2schedule_2schedule_8h__incl.svg | 788 +++----
.../api/doxygen/tir_2transform_8h__incl.svg | 840 +++----
.../api/doxygen/tir_2usmp_2analysis_8h__incl.svg | 896 ++++----
.../api/doxygen/tir_2usmp_2transform_8h.html | 2 +-
.../api/doxygen/tir_2usmp_2transform_8h__incl.svg | 2045 ++++++++---------
.../reference/api/doxygen/tir_2usmp_2utils_8h.html | 2 +-
.../api/doxygen/tir_2usmp_2utils_8h__dep__incl.svg | 36 +-
.../api/doxygen/tir_2usmp_2utils_8h__incl.svg | 2096 ++++++++---------
.../api/doxygen/topi_2nn_8h__dep__incl.svg | 12 +-
docs/reference/api/doxygen/topi_2nn_8h__incl.svg | 956 ++++----
.../api/doxygen/topi_2transform_8h__dep__incl.svg | 64 +-
.../api/doxygen/topi_2transform_8h__incl.svg | 924 ++++----
.../reference/api/doxygen/topi_2utils_8h__incl.svg | 1090 +++++----
docs/reference/api/doxygen/trace_8h__dep__incl.svg | 140 +-
docs/reference/api/doxygen/trace_8h__incl.svg | 924 ++++----
docs/reference/api/doxygen/transform__step_8h.html | 2 +-
.../api/doxygen/transform__step_8h__dep__incl.svg | 96 +-
.../api/doxygen/transform__step_8h__incl.svg | 2068 ++++++++---------
.../api/doxygen/tune__context_8h__dep__incl.svg | 24 +-
.../api/doxygen/tune__context_8h__incl.svg | 792 +++----
.../api/doxygen/type__functor_8h__incl.svg | 996 ++++-----
.../api/doxygen/type__relation_8h__dep__incl.svg | 564 ++---
.../api/doxygen/type__relation_8h__incl.svg | 872 ++++----
docs/reference/api/doxygen/var_8h__dep__incl.svg | 616 ++---
docs/reference/api/doxygen/var_8h__incl.svg | 1098 +++++----
.../api/doxygen/virtual__device_8h__dep__incl.svg | 256 +--
.../api/doxygen/virtual__device_8h__incl.svg | 852 +++----
docs/reference/api/doxygen/vision_8h__incl.svg | 810 +++----
docs/reference/api/doxygen/x86_2bnn_8h__incl.svg | 988 ++++----
.../api/doxygen/x86_2default_8h__incl.svg | 992 ++++----
.../api/doxygen/x86_2injective_8h__incl.svg | 988 ++++----
docs/reference/api/python/auto_scheduler.html | 4 +-
docs/reference/api/python/tir.html | 126 +-
.../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 | 4 +-
docs/tutorial/autotvm_relay_x86.html | 270 +--
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 | 41 +-
579 files changed, 140359 insertions(+), 138765 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 99c77db2d..e8016f331 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.zipb1879542-46bf-46e7-b009-9e5784efc22b from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip3b17ef11-bae1-49ae-ad3c-9de9e65665f0 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
x (1, 3, 224, 224)
diff --git a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
index 45a8d6636..1c0400ba4 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -100,7 +100,7 @@ Load a pretrained OneFlow model and save model
.. code-block:: none
Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
0%| | 0.00/41.5M [00:00<?, ?B/s]
0%| | 16.0k/41.5M [00:00<07:48, 92.7kB/s]
0%| | 48.0k/41.5M [00:00<04:56, 147kB/s]
0%| | 96.0k/41.5M [00:00<03:30, 206kB/s]
0%| | 160k/41.5M [00:00<02:40, 270kB/s]
1%| | 312k/41.5M [00:00<01:28, 489kB/s]
1%|1 | 624k/41.5M [00:01<00:45, 934kB/s]
3%|2 | 1.21M/41.5M [00:01<00:23, 1.79MB/s]
6%|5 | 2.42M/41.5M [00:01<00:11, 3.49MB/s]
9%|9 | 3.89M/41.5M [00:01<00:07, 5.11MB/s]
13%|#2 | 5.37M/41.5M [00:01<00:06, 6.22MB/s]
16%|#6 | 6.84M/41.5M [00:01<00:04, 7.98MB/s]
19%|#9 | 7.91M/41.5M [00:01<00:04, 8.66MB/s]
21%|##1 | 8.82M/41.5M [00:02<00:04, 7.88MB/s]
24%|##3 | 9.77M/41.5M [00:02<00:03, 8.36MB/s]
26%|##6 | 10.8M/41.5M [00:02<00:03, 9.01MB/s]
28%|##8 | 11.8M/41.5M [00:02<00:03, 8.06MB/s]
31%|### | 12.7M/41.5M [00:02<00
:03, 8.18MB/s]
34%|###4 | 14.2M/41.5M [00:02<00:03, 9.37MB/s]
36%|###6 | 15.1M/41.5M [00:02<00:03, 9.08MB/s]
38%|###8 | 16.0M/41.5M [00:03<00:03, 7.68MB/s]
41%|####1 | 17.1M/41.5M [00:03<00:03, 7.46MB/s]
45%|####4 | 18.6M/41.5M [00:03<00:03, 7.87MB/s]
48%|####8 | 20.1M/41.5M [00:03<00:02, 8.14MB/s]
52%|#####1 | 21.6M/41.5M [00:03<00:02, 8.31MB/s]
55%|#####5 | 23.0M/41.5M [00:03<00:02, 9.46MB/s]
58%|#####7 | 24.1M/41.5M [00:03<00:01, 9.68MB/s]
60%|###### | 25.0M/41.5M [00:04<00:01, 8.89MB/s]
63%|######2 | 25.9M/41.5M [00:04<00:01, 8.91MB/s]
65%|######5 | 27.0M/41.5M [00:04<00:01, 9.30MB/s]
67%|######7 | 27.9M/41.5M [00:04<00:01, 8.45MB/s]
70%|######9 | 28.9M/41.5M [00:04<00:01, 8.75MB/s]
72%|#######2 | 29.9M/41.5M [00:04<00:01, 9.18MB/s]
74%|#######4 | 30.8M/41.5M [00:04<00:01, 8.30MB/s]
77%|#######6 | 31.8M/41.5M [00:04<00:01, 8.41MB/s]
80%|####
#### | 33.2M/41.5M [00:05<00:00, 9.73MB/s]
82%|########2 | 34.2M/41.5M [00:05<00:00, 9.11MB/s]
85%|########4 | 35.1M/41.5M [00:05<00:00, 7.76MB/s]
87%|########7 | 36.2M/41.5M [00:05<00:00, 7.53MB/s]
91%|######### | 37.7M/41.5M [00:05<00:00, 9.26MB/s]
93%|#########3| 38.7M/41.5M [00:05<00:00, 9.33MB/s]
95%|#########5| 39.6M/41.5M [00:05<00:00, 8.33MB/s]
98%|#########7| 40.7M/41.5M [00:06<00:00, 7.61MB/s]
100%|##########| 41.5M/41.5M [00:06<00:00, 7.19MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
0%| | 16.0k/41.5M [00:00<07:56, 91.3kB/s]
0%| | 48.0k/41.5M [00:00<05:00, 144kB/s]
0%| | 96.0k/41.5M [00:00<03:33, 203kB/s]
0%| | 144k/41.5M [00:00<03:08, 230kB/s]
1%| | 296k/41.5M [00:00<01:34, 458kB/s]
1%|1 | 592k/41.5M [00:01<00:49, 873kB/s]
3%|2 | 1.11M/41.5M [00:01<00:26, 1.59MB/s]
5%|5 | 2.22M/41.5M [00:01<00:13, 3.14MB/s]
9%|8 | 3.70M/41.5M [00:01<00:08, 4.85MB/s]
12%|#2 | 5.18M/41.5M [00:01<00:06, 6.02MB/s]
16%|#6 | 6.66M/41.5M [00:01<00:05, 6.80MB/s]
20%|#9 | 8.13M/41.5M [00:02<00:04, 7.34MB/s]
23%|##3 | 9.61M/41.5M [00:02<00:04, 7.71MB/s]
27%|##6 | 11.1M/41.5M [00:02<00:03, 7.98MB/s]
30%|### | 12.6M/41.5M [00:02<00:03, 8.16MB/s]
34%|###3 | 14.0M/41.5M [00:02<00:03, 8.29MB/s]
37%|###7 | 15.5M/41.5M [00:03<00
:03, 8.36MB/s]
41%|#### | 17.0M/41.5M [00:03<00:03, 8.43MB/s]
44%|####4 | 18.5M/41.5M [00:03<00:02, 8.46MB/s]
48%|####8 | 19.9M/41.5M [00:03<00:02, 8.50MB/s]
52%|#####1 | 21.4M/41.5M [00:03<00:02, 8.52MB/s]
55%|#####5 | 22.9M/41.5M [00:03<00:02, 8.54MB/s]
59%|#####8 | 24.4M/41.5M [00:04<00:02, 8.55MB/s]
62%|######2 | 25.8M/41.5M [00:04<00:01, 9.85MB/s]
65%|######4 | 26.9M/41.5M [00:04<00:01, 9.98MB/s]
67%|######7 | 27.9M/41.5M [00:04<00:01, 8.91MB/s]
69%|######9 | 28.8M/41.5M [00:04<00:01, 7.71MB/s]
73%|#######2 | 30.3M/41.5M [00:04<00:01, 9.37MB/s]
75%|#######5 | 31.3M/41.5M [00:04<00:01, 9.55MB/s]
78%|#######7 | 32.3M/41.5M [00:05<00:01, 8.49MB/s]
80%|######## | 33.2M/41.5M [00:05<00:01, 7.50MB/s]
84%|########3 | 34.7M/41.5M [00:05<00:00, 9.17MB/s]
86%|########6 | 35.7M/41.5M [00:05<00:00, 9.44MB/s]
88%|########8 | 36.7M/41.5M [00:05<00:00, 8.37MB/s]
91%|####
##### | 37.7M/41.5M [00:05<00:00, 7.45MB/s]
94%|#########4| 39.1M/41.5M [00:05<00:00, 9.15MB/s]
97%|#########6| 40.1M/41.5M [00:05<00:00, 9.46MB/s]
99%|#########9| 41.1M/41.5M [00:06<00:00, 8.37MB/s]
100%|##########| 41.5M/41.5M [00:06<00:00, 7.08MB/s]
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 4b7de7316..ec850ebd8 100644
--- a/docs/_sources/how_to/compile_models/from_paddle.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
@@ -210,7 +210,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 6.130 seconds)
+ **Total running time of the script:** ( 1 minutes 8.766 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 5e31ee53f..b77001b4c 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]
39%|###9 | 17.5M/44.7M [00:00<00:00, 184MB/s]
88%|########8 | 39.3M/44.7M [00:00<00:00, 210MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 208MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
28%|##7 | 12.4M/44.7M [00:00<00:00, 130MB/s]
67%|######7 | 30.0M/44.7M [00:00<00:00, 162MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 139MB/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 a70738353..0f21926a1 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -379,11 +379,6 @@ Run the corresponding model on tensorflow
-.. rst-class:: sphx-glr-timing
-
- **Total running time of the script:** ( 1 minutes 6.315 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 ca6a09269..f8d8fa672 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
@@ -5,15 +5,15 @@
Computation times
=================
-**05:24.769** total execution time for **how_to_compile_models** files:
+**05:35.761** total execution time for **how_to_compile_models** files:
-- **01:06.315**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
-- **01:06.130**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
-- **00:57.446**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
-- **00:30.970**: :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)
-- **00:24.441**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
-- **00:21.955**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
-- **00:21.567**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
-- **00:19.348**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
-- **00:14.099**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
-- **00:02.499**: :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)
+- **01:08.766**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
+- **00:59.888**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
+- **00:57.086**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
+- **00:41.220**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
+- **00:32.201**: :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)
+- **00:20.726**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
+- **00:20.624**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
+- **00:19.288**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
+- **00:13.291**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
+- **00:02.670**: :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 cc87c7abe..0f7015970 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
@@ -402,7 +402,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.1705 16.1716 16.2298 16.1183 0.0334
+ 15.8685 15.8805 16.1054 15.5230 0.1629
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 47767fa59..325b8fa16 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]
8%|8 | 14.0M/170M [00:00<00:01, 147MB/s]
18%|#7 | 30.5M/170M [00:00<00:00, 162MB/s]
29%|##8 | 48.7M/170M [00:00<00:00, 175MB/s]
45%|####4 | 75.7M/170M [00:00<00:00, 218MB/s]
60%|###### | 102M/170M [00:00<00:00, 240MB/s]
75%|#######4 | 127M/170M [00:00<00:00, 247MB/s]
89%|########8 | 151M/170M [00:00<00:00, 247MB/s]
100%|##########| 170M/170M [00:00<00:00, 231MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
10%|9 | 16.2M/170M [00:00<00:00, 168MB/s]
19%|#8 | 32.2M/170M [00:00<00:01, 141MB/s]
28%|##8 | 48.3M/170M [00:00<00:00, 152MB/s]
37%|###7 | 63.0M/170M [00:00<00:00, 142MB/s]
45%|####5 | 76.8M/170M [00:00<00:00, 130MB/s]
59%|#####8 | 99.6M/170M [00:00<00:00, 163MB/s]
73%|#######2 | 123M/170M [00:00<00:00, 188MB/s]
85%|########4 | 144M/170M [00:00<00:00, 197MB/s]
96%|#########6| 163M/170M [00:01<00:00, 174MB/s]
100%|##########| 170M/170M [00:01<00:00, 167MB/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').
@@ -262,7 +262,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 14.835 seconds)
+ **Total running time of the script:** ( 2 minutes 54.422 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 bc1aa6566..a7d615b0b 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]
100%|##########| 13.6M/13.6M [00:00<00:00, 176MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
34%|###3 | 4.56M/13.6M [00:00<00:00, 46.5MB/s]
66%|######6 | 9.01M/13.6M [00:00<00:00, 33.4MB/s]
91%|#########1| 12.4M/13.6M [00:00<00:00, 22.2MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 25.2MB/s]
@@ -353,7 +353,7 @@ 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)
- 90.5915 90.4739 91.5880 90.2648 0.2938
+ 88.0771 87.9550 89.8937 87.7442 0.3431
@@ -393,7 +393,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 7.292 seconds)
+ **Total running time of the script:** ( 1 minutes 3.247 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 c02160f58..4cd93af95 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
@@ -360,7 +360,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 123.6896 123.6542 124.7379 123.0264 0.4053
+ 115.5216 115.2153 121.6813 114.2261 1.1246
@@ -394,7 +394,7 @@ 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 1.358 seconds)
+ **Total running time of the script:** ( 1 minutes 58.743 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 82b0a5934..9cd5b80ba 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -223,7 +223,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 37.312 seconds)
+ **Total running time of the script:** ( 1 minutes 16.460 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 0bc502714..2e49a7fcd 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%|2 | 3029/132723 [00:00<00:04, 30287.14KB/s]
6%|5 | 7561/132723 [00:00<00:03, 36819.82KB/s]
9%|8 | 11789/132723 [00:00<00:03, 39229.18KB/s]
15%|#5 | 20312/132723 [00:00<00:01, 57019.24KB/s]
22%|##1 | 28979/132723 [00:00<00:01, 67569.57KB/s]
28%|##8 | 37708/132723 [00:00<00:01, 74215.28KB/s]
35%|###5 | 46489/132723 [00:00<00:01, 78629.73KB/s]
42%|####1 | 55274/132723 [00:00<00:00, 81551.82KB/s]
48%|####8 | 64118/132723 [00:00<00:00, 83697.12KB/s]
55%|#####4 | 72765/132723 [00:01<00:00, 84547.73KB/s]
61%|######1 | 81621/132723 [00:01<00:00, 85770.83KB/s]
68%|######8 | 90437/132723 [00:01<00:00, 86494.75KB/s]
75%|#######4 | 99291/132723 [00:01<00:00, 87111.53KB/s]
81%|########1 | 108006/132723 [00:01<00:00, 82950.03KB/s]
88%|########7 | 116784/132723 [00:01<00:00, 84353.61KB/s]
95%|#########4
| 125589/132723 [00:01<00:00, 85437.12KB/s]
100%|##########| 132723/132723 [00:01<00:00, 77615.30KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
5%|4 | 6112/132723 [00:00<00:02, 61115.54KB/s]
11%|#1 | 14734/132723 [00:00<00:01, 75879.60KB/s]
17%|#6 | 22322/132723 [00:00<00:01, 57385.81KB/s]
23%|##3 | 30849/132723 [00:00<00:01, 66594.41KB/s]
29%|##8 | 37940/132723 [00:00<00:01, 47783.49KB/s]
34%|###3 | 45094/132723 [00:00<00:01, 53567.03KB/s]
39%|###8 | 51247/132723 [00:00<00:01, 44700.90KB/s]
45%|####4 | 59383/132723 [00:01<00:01, 53080.04KB/s]
49%|####9 | 65523/132723 [00:01<00:01, 50456.66KB/s]
55%|#####5 | 73081/132723 [00:01<00:01, 56572.75KB/s]
60%|#####9 | 79285/132723 [00:01<00:00, 56407.72KB/s]
64%|######4 | 85304/132723 [00:01<00:00, 51535.07KB/s]
70%|####### | 93194/132723 [00:01<00:00, 58455.23KB/s]
75%|#######4 | 99395/132723 [00:01<00:00, 48942.51KB/s]
81%|########1 | 107525/132723 [00:01<00:00, 56644.15KB/s]
86%|########5
| 113724/132723 [00:02<00:00, 55712.46KB/s]
90%|######### | 119663/132723 [00:02<00:00, 42008.93KB/s]
96%|#########6| 127879/132723 [00:02<00:00, 50624.68KB/s]
100%|##########| 132723/132723 [00:02<00:00, 53534.86KB/s]
@@ -211,7 +211,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 28.269 seconds)
+ **Total running time of the script:** ( 2 minutes 20.338 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 bd17df106..134177385 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
=================
-**11:22.479** total execution time for **how_to_deploy_models** files:
+**10:22.528** total execution time for **how_to_deploy_models** files:
-- **03:14.835**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
-- **02:28.269**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
-- **02:01.358**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
-- **01:37.312**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
-- **01:07.292**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
-- **00:30.085**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
-- **00:23.120**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
-- **00:00.208**: :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)
+- **02:54.422**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
+- **02:20.338**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
+- **01:58.743**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
+- **01:16.460**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
+- **01:03.247**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
+- **00:27.628**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
+- **00:21.511**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
+- **00:00.179**: :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 b65299339..b05b238ec 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
@@ -425,7 +425,7 @@ 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.zip85795ee8-f62f-4165-9871-80cddc17ab20 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip50c0d070-59e0-4d71-9505-d0c79d0c4b7e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
diff --git a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
index 8f8e62447..7b86b7ab6 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
=================
-**00:39.421** total execution time for **how_to_extend_tvm** files:
+**00:37.391** total execution time for **how_to_extend_tvm** files:
-- **00:35.744**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
-- **00:02.352**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
-- **00:01.104**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
-- **00:00.221**: :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)
+- **00:33.995**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
+- **00:02.204**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
+- **00:01.005**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
+- **00:00.186**: :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 61bfd2593..eb6abf2de 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: 6065us [6065us] (40.24%; 40.24%)
- FoldScaleAxis: 9006us [3us] (59.76%; 59.76%)
- FoldConstant: 9003us [1500us] (59.74%; 99.97%)
- InferType: 7502us [7502us] (49.78%; 83.34%)
+ InferType: 6002us [6002us] (45.63%; 45.63%)
+ FoldScaleAxis: 7153us [2us] (54.37%; 54.37%)
+ FoldConstant: 7151us [1454us] (54.36%; 99.97%)
+ InferType: 5697us [5697us] (43.31%; 79.67%)
@@ -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: 9563us [9563us] (47.50%; 47.50%)
- FoldScaleAxis: 10570us [3us] (52.50%; 52.50%)
- FoldConstant: 10567us [2109us] (52.49%; 99.97%)
- InferType: 8459us [8459us] (42.01%; 80.05%)
+ InferType: 5739us [5739us] (44.57%; 44.57%)
+ FoldScaleAxis: 7137us [2us] (55.43%; 55.43%)
+ FoldConstant: 7136us [1498us] (55.41%; 99.97%)
+ InferType: 5637us [5637us] (43.78%; 79.00%)
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 297d51ecc..8aea4421c 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
@@ -299,7 +299,7 @@ latency of convolution.
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Convolution: 37.517698 ms
+ Convolution: 54.199556 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 e3b94287d..e530122c4 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
@@ -632,7 +632,7 @@ be able to run on our build server
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- conv2d with tensor core: 7.367135 ms
+ conv2d with tensor core: 9.389245 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 abddf5ed0..38e5f1a40 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -118,10 +118,10 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.019321
+ Numpy running time: 0.018460
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Baseline: 3.301291
+ Baseline: 3.445202
@@ -212,7 +212,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.317786
+ Opt1: 0.295180
@@ -311,7 +311,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.346788
+ Opt2: 0.333483
@@ -403,7 +403,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.132057
+ Opt3: 0.114889
@@ -522,7 +522,7 @@ flattening.
.. code-block:: none
- Opt4: 0.111363
+ Opt4: 0.110148
@@ -640,7 +640,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.112277
+ Opt5: 0.111236
@@ -761,7 +761,7 @@ Futhermore, we can also utilize multi-core processors to do the thread-level par
.. code-block:: none
- Opt6: 0.145991
+ Opt6: 0.145570
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 fc45e560b..6ea4760bc 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
=================
-**00:35.509** total execution time for **how_to_optimize_operators** files:
+**00:35.098** total execution time for **how_to_optimize_operators** files:
-- **00:32.787**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
-- **00:01.492**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
-- **00:01.230**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)
+- **00:32.373**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
+- **00:01.482**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
+- **00:01.243**: :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 91206e652..69f078363 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
=================
-**05:09.723** total execution time for **how_to_tune_with_autoscheduler** files:
-
-- **02:22.075**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
-- **01:20.816**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
-- **00:41.277**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
-- **00:27.315**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
-- **00:09.415**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
-- **00:08.825**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)
+**04:55.021** total execution time for **how_to_tune_with_autoscheduler** files:
+
+- **02:25.955**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
+- **01:16.455**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
+- **00:39.267**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
+- **00:16.716**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
+- **00:08.559**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
+- **00:08.070**: :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 77d7493c7..7a0725bf2 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
@@ -222,74 +222,483 @@ cooperative fetching, unrolling and operator fusion.
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 112;
- allocate(conv2d_nchw: Pointer(local float32), float32, [2]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [54]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [576]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [2], [], scope="local", align=8)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 28;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[1] = 0f32
- for (rc.outer.outer: int32, 0, 256) {
- let cse_var_1: int32 = (rc.outer.outer*18)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- if @tir.likely((threadIdx.x_1 < 54), dtype=bool) {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [54], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((3 <= floormod(threadIdx.x_1, 27)) && (floormod(threadIdx.x_1, 27) < 24)) && (1 <= (floormod(blockIdx.x, 7) + floormod(threadIdx.x_1, 3)))) && ((floormod(blockIdx.x, 7) + floormod(threadIdx.x_1, 3)) < 8)), data[((((((rc.outer.outer*98) + (floordiv(threadIdx.x_1, 27)*49)) + (floordiv(floormod(threadIdx.x_1, 27), 3)*7)) + floormod(blockIdx.x, 7)) + floormod(thre [...]
+ conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[3] = 0f32
+ conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[5] = 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) {
+ 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" = 112;
- kernel.shared_1: Buffer(kernel.shared, float32, [576], [], scope="shared")[threadIdx.x_2] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv(threadIdx.x_2, 18)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 56), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 168), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 12), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 224), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- if @tir.likely((threadIdx.x_2 < 16), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 280), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 18))]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*3)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 21)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 24)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 30)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 30)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 30)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 33)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 33)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 22)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 25)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 31)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 31)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 31)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 34)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 34)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 23)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 26)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 32)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 32)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 32)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 35)]))
}
}
for (i1.inner: int32, 0, 2) {
- compute[(((((floordiv(blockIdx.x, 7)*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (floormod(threadIdx.x, 7)*7)) + floormod(blockIdx.x, 7))] = max((conv2d_nchw_1[i1.inner] + bias[(((floordiv(blockIdx.x, 7)*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
+ 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)
+ }
}
}
}
@@ -346,7 +755,7 @@ We build the binary and check its correctness and performance.
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Execution time of this operator: 0.419 ms
+ Execution time of this operator: 0.361 ms
@@ -390,21 +799,21 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
- conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
- conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
- conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
+ conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+ conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
- conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
+ conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
- conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+ conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=1)
- conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+ conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
@@ -413,12 +822,12 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
- compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+ compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
- compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+ compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
@@ -439,12 +848,12 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
+ 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=112)
+ 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", 512)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
@@ -466,65 +875,430 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[2];
- __shared__ float pad_temp_shared[54];
- __shared__ float kernel_shared[576];
+ extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[14];
+ __shared__ float pad_temp_shared[72];
+ __shared__ float kernel_shared[3072];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 256; ++rc_outer_outer) {
- __syncthreads();
- if (((int)threadIdx.x) < 54) {
- pad_temp_shared[((int)threadIdx.x)] = (((((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 18)) + (((int)threadIdx.x) % 18))];
- kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 4) % 18))];
- kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 8) % 18))];
- kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 12) % 18))];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 16) % 18))];
- if (((int)threadIdx.x) < 16) {
- kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 18)) + (((int)threadIdx.x) + 2))];
+ conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[5] = 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) {
+ 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)]));
}
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[((((int)threadIdx.x) / 7) * 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 30)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 30)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 33)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 33)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 31)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 31)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 34)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 34)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 32)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 32)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 35)]));
}
for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i1_inner] + bias[((((((int)blockIdx.x) / 7) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ 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);
+ }
}
}
@@ -583,7 +1357,7 @@ 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 22.075 seconds)
+ **Total running time of the script:** ( 2 minutes 25.955 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 a1e316946..517f06cb0 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
@@ -616,7 +616,7 @@ 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)
- 10.1293 10.1448 10.1897 10.0533 0.0567
+ 9.9674 9.9812 10.0124 9.9087 0.0434
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 5c2e97df7..df6f0362f 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
@@ -635,7 +635,7 @@ 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)
- 768.4319 769.6622 773.6825 761.9511 4.8677
+ 736.8305 737.5163 738.4373 734.5379 1.6642
@@ -660,7 +660,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 20.816 seconds)
+ **Total running time of the script:** ( 1 minutes 16.455 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 85ad4b3c8..16bd1b482 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
@@ -362,32 +362,76 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_5: placeholder_15: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_7: placeholder_19: Buffer(placeholder_12, int32, [4916], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 16) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [4096]), storage_scope = global {
- for (i.outer.inner: int32, 0, 2) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 64) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [4096], [])[((((i.outer.inner*2048) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
- }
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_17: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_19: Buffer(placeholder_14, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 64) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global {
+ for (i.outer.inner: int32, 0, 8) {
+ for (i.inner.init: int32, 0, 8) {
+ let cse_var_1: int32 = ((i.outer.inner*128) + (i.inner.init*16))
+ {
+ compute_5: Buffer(compute_4, float32, [1024], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (i.inner: int32, 0, 64) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner)
- let cse_var_2: int32 = ((((i.outer.inner*2048) + (i.inner*32)) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
- }
+ }
+ for (elem_idx: int32, 0, let cse_var_2: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_2 + 1)] - placeholder_3[cse_var_2])) {
+ for (i.inner: int32, 0, 8) {
+ let cse_var_21: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ let cse_var_20: int32 = (elem_idx*16)
+ let cse_var_19: int32 = ((i.outer.inner*128) + (i.inner*16))
+ let cse_var_18: int32 = (cse_var_19 + 10)
+ let cse_var_17: int32 = (cse_var_19 + 11)
+ let cse_var_16: int32 = (cse_var_19 + 12)
+ let cse_var_15: int32 = (cse_var_19 + 13)
+ let cse_var_14: int32 = (cse_var_19 + 14)
+ let cse_var_13: int32 = (cse_var_19 + 15)
+ let cse_var_12: int32 = (cse_var_19 + 2)
+ let cse_var_11: int32 = (cse_var_19 + 3)
+ let cse_var_10: int32 = (cse_var_19 + 4)
+ let cse_var_9: int32 = (cse_var_19 + 5)
+ let cse_var_8: int32 = (cse_var_19 + 6)
+ let cse_var_7: int32 = (cse_var_19 + 7)
+ let cse_var_6: int32 = (cse_var_19 + 8)
+ let cse_var_5: int32 = (cse_var_19 + 9)
+ let cse_var_4: int32 = (cse_var_19 + 1)
+ let cse_var_3: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i.outer.inner*2048)) + (i.inner*256))
+ {
+ compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[((placeholder_3[cse_var_21]*16) + cse_var_20)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 1)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 2)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 3)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 4)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 5)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 6)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 7)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 8)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 9)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 10)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 11)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 12)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 13)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 14)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 15)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 128) {
- for (i1.inner: int32, 0, 32) {
- let cse_var_4: int32 = (((i0.inner*512) + (i0.outer.i1.outer.fused*32)) + i1.inner)
- compute[cse_var_4] = max((compute_5[((i0.inner*32) + i1.inner)] + placeholder_4[cse_var_4]), 0f32)
- }
+ for (i0.inner: int32, 0, 64) {
+ let cse_var_22: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*32768) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_22, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_22, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -443,7 +487,7 @@ We build the binary and check its correctness and performance.
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Execution time of this operator: 1.839 ms
+ Execution time of this operator: 1.688 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 76e29fb13..ba83f8ac1 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
=================
-**00:44.713** total execution time for **how_to_tune_with_autotvm** files:
+**00:44.226** total execution time for **how_to_tune_with_autotvm** files:
-- **00:43.770**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
-- **00:00.248**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
-- **00:00.233**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
-- **00:00.232**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
-- **00:00.230**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)
+- **00:43.390**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
+- **00:00.227**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
+- **00:00.213**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
+- **00:00.200**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
+- **00:00.196**: :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 b2f78385c..dafe3ed58 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: 110.31/110.31 result: MeasureResult(costs=(0.0020986534583333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6601579189300537, timestamp=1653441326.2048783) [('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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 6 GFLOPS: 100.05/100.05 result: MeasureResult(costs=(0.0023137704375,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.686079978942871, timestamp=1653528323.4934978) [('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: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 8 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 9 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 10 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 11 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 12 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 13 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 14 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 15 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 16 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 17 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 18 GFLOPS: 0.00/100.05 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: 0.00/110.31 result: Traceback (most recent call last):
+ No: 19 GFLOPS: 0.00/100.05 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: 0x00007f88e4504fa2
+ 12: 0x00007f0caee70fa2
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: 142.02/142.02 result: MeasureResult(costs=(0.0016300953799999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4533238410949707, timestamp=1653441352.8289824) [('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: 143.07/143.07 result: MeasureResult(costs=(0.0016180540999999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4196603298187256, timestamp=1653528349.8571422) [('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
@@ -2441,7 +2441,7 @@ and measure running time.
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Time cost of this operator: 0.002079
+ Time cost of this operator: 0.001979
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 b9ab36482..dcc90183f 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
@@ -294,10 +294,10 @@ 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 310.7 98.739 (1, 2, 10, 10, 3) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.018 0.959 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.949 0.302 (1, 1, 10, 10, 3) 1 1
- Total_time - 314.667 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.2 98.676 (1, 2, 10, 10, 3) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.258 1.03 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.932 0.295 (1, 1, 10, 10, 3) 1 1
+ Total_time - 316.39 - - - -
@@ -359,10 +359,10 @@ 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 80.95 96.807 (1, 6, 10, 10, 1) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.74 2.08 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.931 1.113 (1, 1, 10, 10, 3) 1 1
- Total_time - 83.62 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 131.3 97.967 (1, 6, 10, 10, 1) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.824 1.361 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.901 0.672 (1, 1, 10, 10, 3) 1 1
+ Total_time - 134.025 - - - -
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 4cfce7c53..4fe17884b 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
=================
-**00:48.363** total execution time for **how_to_work_with_microtvm** files:
+**00:44.855** total execution time for **how_to_work_with_microtvm** files:
-- **00:43.939**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
-- **00:03.794**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
-- **00:00.211**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)
-- **00:00.210**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
-- **00:00.210**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
+- **00:40.645**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
+- **00:03.636**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
+- **00:00.198**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
+- **00:00.191**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
+- **00:00.186**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.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 0910f4c9e..8977c1d4e 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
=================
-**00:09.558** total execution time for **how_to_work_with_relay** files:
+**00:06.279** total execution time for **how_to_work_with_relay** files:
-- **00:07.390**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
-- **00:01.940**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
-- **00:00.229**: :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)
+- **00:04.304**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
+- **00:01.763**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
+- **00:00.212**: :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 975519b8e..8f23ca621 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
=================
-**00:06.021** total execution time for **how_to_work_with_schedules** files:
+**00:05.851** total execution time for **how_to_work_with_schedules** files:
-- **00:02.221**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
-- **00:01.214**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
-- **00:00.765**: :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)
-- **00:00.758**: :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)
-- **00:00.323**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
-- **00:00.254**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
-- **00:00.252**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
-- **00:00.235**: :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)
+- **00:02.074**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
+- **00:01.350**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
+- **00:00.723**: :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``)
+- **00:00.308**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
+- **00:00.237**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
+- **00:00.230**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
+- **00:00.229**: :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 1775e3f34..f248b0ce2 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -318,7 +318,7 @@ The importing needs to happen before the tensorized GEMV being executed.
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp2ehf9h_2/input0.cc'\nsource_filename = \"/tmp/tmp2ehf9h_2/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdsdid5jn/input0.cc'\nsource_filename = \"/tmp/tmpdsdid5jn/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 13f37cda8..6d7cb8cb5 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
=================
-**00:21.578** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:20.087** total execution time for **topic_vta_tutorials_autotvm** files:
-- **00:21.359**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
-- **00:00.219**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)
+- **00:19.883**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
+- **00:00.204**: :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 43a2a7ee3..4d76c6853 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -267,7 +267,7 @@ 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 22.54s!
+ resnet18_v1 inference graph built in 21.46s!
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 95592af9d..546e58daf 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -303,7 +303,7 @@ The compilation steps are:
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/relay/build_module.py:389: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 15.54s!
+ yolov3-tiny inference graph built in 14.69s!
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 cabd11835..5a29f1b19 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
=================
-**01:30.787** total execution time for **topic_vta_tutorials_frontend** files:
+**01:28.988** total execution time for **topic_vta_tutorials_frontend** files:
-- **00:47.952**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
-- **00:42.834**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``)
+- **00:47.244**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
+- **00:41.745**: :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 1993f5b72..d4e4c9824 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
=================
-**00:03.602** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.531** total execution time for **topic_vta_tutorials_optimize** files:
-- **00:03.011**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
-- **00:00.591**: :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``)
+- **00:02.983**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
+- **00:00.547**: :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 3d1e3b632..95f251333 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
=================
-**00:01.072** total execution time for **topic_vta_tutorials** files:
+**00:01.039** total execution time for **topic_vta_tutorials** files:
-- **00:00.540**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
-- **00:00.531**: :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``)
+- **00:00.532**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
+- **00:00.507**: :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 9ebaa8c12..41019ad8f 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -308,7 +308,7 @@ We build the binary and check its correctness and performance.
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- Execution time of this operator: 93.896 ms
+ Execution time of this operator: 94.090 ms
@@ -404,7 +404,7 @@ resume the status and do more 5 trials.
Resume search:
/usr/local/lib/python3.7/dist-packages/xgboost/training.py:17: UserWarning: Old style callback is deprecated. See: https://xgboost.readthedocs.io/en/latest/python/callbacks.html
warnings.warn(f'Old style callback is deprecated. See: {link}', UserWarning)
-
+ .T
@@ -417,11 +417,6 @@ Expression (TE) language that demonstrates how TVM can optimize computational
operations.
-.. rst-class:: sphx-glr-timing
-
- **Total running time of the script:** ( 1 minutes 10.194 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 a78079675..8ce803b51 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -280,7 +280,7 @@ standard deviation.
.. code-block:: none
- {'mean': 497.0214020900312, 'median': 496.80926645005457, 'std': 1.3955343831042604}
+ {'mean': 483.11649433006096, 'median': 485.0085570498777, 'std': 8.787907520273365}
@@ -494,31 +494,31 @@ the tuning data to.
.. code-block:: none
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 17.53/ 17.53 GFLOPS | Progress: (4/20) | 6.12 s
[Task 1/25] Current/Best: 6.16/ 17.53 GFLOPS | Progress: (8/20) | 9.09 s
[Task 1/25] Current/Best: 11.53/ 22.63 GFLOPS | Progress: (12/20) | 11.61 s
[Task 1/25] Current/Best: 16.81/ 22.69 GFLOPS | Progress: (16/20) | 13.31 s
[Task 1/25] Current/Best: 11.58/ 23.88 GFLOPS | Progress: (20/20) | 15.05 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 11.99/ 12.83 GFLOPS | Progress: (4/20) | 3.99 s
[Task 2/25] Current/Best: 14.40/ 18.69 GFLOPS | Progress: (8/20) | 5.31 s
[Task 2/25] Current/Best: 20.79/ 20.79 GFLOPS | Progress: (12/20) | 6.64 s
[Task 2/25] Current/Best: 11.93/ 20.79 GFLOPS | Progress: (16/20) | 7.92 s
[Task 2/25] Current/Best: 19.99/ 20.79 GFLOPS | Progress: (20/20) | 9.54 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 1.63/ 10.52 GFLOPS | Progress: (4/20) | 5.85 s
[Task 3/25] Current/Best: 15.56/ 16.79 GFLOPS | Progress: (8/20) | 7.78 s
[Task 3/25] Current/Best: 14.82/ 16.79 GFLOPS | Progress: (12/20) | 9.51 s
[Task 3/25] Current/Best: 7.17/ 23.62 GFLOPS | Progress: (16/20) | 11.43 s
[Task 3/25] Current/Best: 12.61/ 23.62 GFLOPS | Progress: (20/20) | 16.01 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 9.45/ 20.02 GFLOPS | Progress: (4/20) | 2.40 s
[Task 4/25] Current/Best: 6.78/ 20.02 GFLOPS | Progress: (8/20) | 7.21 s
[Task 4/25] Current/Best: 21.48/ 21.48 GFLOPS | Progress: (12/20) | 12.31 s
[Task 4/25] Current/Best: 16.98/ 21.48 GFLOPS | Progress: (16/20) | 14.77 s
[Task 4/25] Current/Best: 12.69/ 21.48 GFLOPS | Progress: (20/20) | 16.90 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 9.95/ 10.38 GFLOPS | Progress: (4/20) | 2.56 s
[Task 5/25] Current/Best: 11.88/ 13.06 GFLOPS | Progress: (8/20) | 4.61 s
[Task 5/25] Current/Best: 10.17/ 17.60 GFLOPS | Progress: (12/20) | 7.89 s
[Task 5/25] Current/Best: 11.89/ 22.68 GFLOPS | Progress: (16/20) | 9.35 s
[Task 5/25] Current/Best: 12.04/ 22.68 GFLOPS | Progress: (20/20) | 11.26 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 12.29/ 20.75 GFLOPS | Progress: (4/20) | 4.13 s
[Task 6/25] Current/Best: 18.99/ 20.75 GFLOPS | Progress: (8/20) | 5.91 s
[Task 6/25] Current/Best: 13.23/ 20.75 GFLOPS | Progress: (12/20) | 7.86 s
[Task 6/25] Current/Best: 19.89/ 20.75 GFLOPS | Progress: (16/20) | 10.11 s
[Task 6/25] Current/Best: 3.76/ 20.75 GFLOPS | Progress: (20/20) | 12.62 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 10.00/ 12.75 GFLOPS | Progress: (4/20) | 3.60 s
[Task 7/25] Current/Best: 19.99/ 21.25 GFLOPS | Progress: (8/20) | 5.12 s
[Task 7/25] Current/Best: 15.83/ 21.25 GFLOPS | Progress: (12/20) | 7.04 s
[Task 7/25] Current/Best: 12.24/ 21.25 GFLOPS | Progress: (16/20) | 9.09 s
[Task 7/25] Current/Best: 6.36/ 21.71 GFLOPS | Progress: (20/20) | 11.55 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 10.50/ 14.22 GFLOPS | Progress: (4/20) | 2.85 s
[Task 8/25] Current/Best: 9.95/ 14.22 GFLOPS | Progress: (8/20) | 8.06 s
[Task 8/25] Current/Best: 13.17/ 14.22 GFLOPS | Progress: (12/20) | 14.66 s
[Task 8/25] Current/Best: 18.88/ 18.88 GFLOPS | Progress: (16/20) | 16.77 s
[Task 8/25] Current/Best: 20.18/ 20.18 GFLOPS | Progress: (20/20) | 24.00 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.24/ 15.74 GFLOPS | Progress: (4/20) | 19.47 s
[Task 9/25] Current/Best: 23.01/ 23.01 GFLOPS | Progress: (8/20) | 21.25 s
[Task 9/25] Current/Best: 8.21/ 23.01 GFLOPS | Progress: (12/20) | 23.85 s
[Task 9/25] Current/Best: 17.99/ 23.01 GFLOPS | Progress: (16/20) | 26.64 s
[Task 9/25] Current/Best: 8.98/ 23.01 GFLOPS | Progress: (20/20) | 35.45 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.46/ 18.46 GFLOPS | Progress: (4/20) | 2.55 s
[Task 10/25] Current/Best: 15.55/ 18.46 GFLOPS | Progress: (8/20) | 4.19 s
[Task 10/25] Current/Best: 12.74/ 19.10 GFLOPS | Progress: (12/20) | 5.75 s
[Task 10/25] Current/Best: 19.17/ 20.38 GFLOPS | Progress: (16/20) | 6.86 s
[Task 10/25] Current/Best: 8.86/ 20.38 GFLOPS | Progress: (20/20
) | 8.40 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 12.23/ 18.00 GFLOPS | Progress: (4/20) | 3.36 s
[Task 11/25] Current/Best: 16.80/ 18.00 GFLOPS | Progress: (8/20) | 6.23 s
[Task 11/25] Current/Best: 17.94/ 18.00 GFLOPS | Progress: (12/20) | 8.28 s
[Task 11/25] Current/Best: 13.32/ 21.19 GFLOPS | Progress: (16/20) | 11.26 s
[Task 11/25] Current/Best: 19.49/ 21.29 GFLOPS | Progress: (20/20) | 13.37 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.78/ 17.74 GFLOPS | Progress: (4/20) | 5.84 s
[Task 12/25] Current/Best: 5.37/ 17.74 GFLOPS | Progress: (8/20) | 9.83 s
[Task 12/25] Current/Best: 19.08/ 19.15 GFLOPS | Progress: (12/20) | 11.83 s
[Task 12/25] Current/Best: 14.91/ 19.15 GFLOPS | Progress: (16/20) | 14.81 s
[Task 12/25] Current/Best: 15.19/ 19.19 GFLOPS | Progress: (20/20) | 16.77 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.76/ 17.30 GFLOPS | Progress: (4/20) | 3.71 s
[Task 13/25] Current/Best: 15.47/ 20.84 GFLOPS | Progress: (8/20) | 6.35 s
[Task 13/25] Current/Best: 19.51/ 21.52 GFLOPS | Progress: (12/20) | 9.45 s
[Task 13/25] Current/Best: 12.21/ 21.52 GFLOPS | Progress: (16/20) | 12.92 s
[Task 13/25] Current/Best: 18.68/ 21.52 GFLOPS | Progress: (20/20) | 15.23 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 13.24/ 13.24 GFLOPS | Progress: (4/20) | 3.39 s
[Task 14/25] Current/Best: 6.08/ 13.29 GFLOPS | Progress: (8/20) | 5.56 s
[Task 14/25] Current/Best: 20.78/ 20.78 GFLOPS | Progress: (12/20) | 8.23 s
[Task 14/25] Current/Best: 16.46/ 20.78 GFLOPS | Progress: (16/20) | 10.11 s
[Task 14/25] Current/Best: 17.06/ 20.78 GFLOPS | Progress: (20/20) | 11.83 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 19.23/ 19.23 GFLOPS | Progress: (4/20) | 5.76 s
[Task 1/25] Current/Best: 6.25/ 19.23 GFLOPS | Progress: (8/20) | 8.60 s
[Task 1/25] Current/Best: 11.75/ 23.17 GFLOPS | Progress: (12/20) | 11.01 s
[Task 1/25] Current/Best: 18.34/ 23.17 GFLOPS | Progress: (16/20) | 12.63 s
[Task 1/25] Current/Best: 11.83/ 24.26 GFLOPS | Progress: (20/20) | 14.33 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.47/ 13.42 GFLOPS | Progress: (4/20) | 3.73 s
[Task 2/25] Current/Best: 14.46/ 19.06 GFLOPS | Progress: (8/20) | 5.02 s
[Task 2/25] Current/Best: 21.69/ 21.69 GFLOPS | Progress: (12/20) | 6.31 s
[Task 2/25] Current/Best: 12.60/ 21.69 GFLOPS | Progress: (16/20) | 7.53 s
[Task 2/25] Current/Best: 20.79/ 21.69 GFLOPS | Progress: (20/20) | 9.09 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 1.65/ 10.67 GFLOPS | Progress: (4/20) | 5.67 s
[Task 3/25] Current/Best: 17.09/ 18.16 GFLOPS | Progress: (8/20) | 7.53 s
[Task 3/25] Current/Best: 16.35/ 18.16 GFLOPS | Progress: (12/20) | 9.20 s
[Task 3/25] Current/Best: 7.33/ 24.20 GFLOPS | Progress: (16/20) | 11.05 s
[Task 3/25] Current/Best: 12.89/ 24.20 GFLOPS | Progress: (20/20) | 15.48 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 9.77/ 20.79 GFLOPS | Progress: (4/20) | 2.25 s
[Task 4/25] Current/Best: 6.94/ 20.79 GFLOPS | Progress: (8/20) | 6.80 s
[Task 4/25] Current/Best: 22.88/ 22.88 GFLOPS | Progress: (12/20) | 11.51 s
[Task 4/25] Current/Best: 19.02/ 22.88 GFLOPS | Progress: (16/20) | 13.83 s
[Task 4/25] Current/Best: 13.75/ 22.88 GFLOPS | Progress: (20/20) | 15.87 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 9.58/ 10.45 GFLOPS | Progress: (4/20) | 2.46 s
[Task 5/25] Current/Best: 11.85/ 12.84 GFLOPS | Progress: (8/20) | 4.50 s
[Task 5/25] Current/Best: 11.96/ 18.36 GFLOPS | Progress: (12/20) | 7.49 s
[Task 5/25] Current/Best: 11.84/ 23.23 GFLOPS | Progress: (16/20) | 8.88 s
[Task 5/25] Current/Best: 12.24/ 23.23 GFLOPS | Progress: (20/20) | 10.75 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 12.35/ 21.07 GFLOPS | Progress: (4/20) | 3.95 s
[Task 6/25] Current/Best: 19.36/ 21.07 GFLOPS | Progress: (8/20) | 5.68 s
[Task 6/25] Current/Best: 13.51/ 21.07 GFLOPS | Progress: (12/20) | 7.59 s
[Task 6/25] Current/Best: 20.32/ 21.07 GFLOPS | Progress: (16/20) | 9.78 s
[Task 6/25] Current/Best: 3.77/ 21.07 GFLOPS | Progress: (20/20) | 12.27 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 11.45/ 13.15 GFLOPS | Progress: (4/20) | 3.47 s
[Task 7/25] Current/Best: 20.65/ 21.52 GFLOPS | Progress: (8/20) | 4.93 s
[Task 7/25] Current/Best: 16.38/ 21.52 GFLOPS | Progress: (12/20) | 6.79 s
[Task 7/25] Current/Best: 12.46/ 21.52 GFLOPS | Progress: (16/20) | 8.80 s
[Task 7/25] Current/Best: 6.50/ 22.13 GFLOPS | Progress: (20/20) | 11.22 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 9.76/ 13.94 GFLOPS | Progress: (4/20) | 2.78 s
[Task 8/25] Current/Best: 9.28/ 13.94 GFLOPS | Progress: (8/20) | 7.81 s
[Task 8/25] Current/Best: 12.47/ 13.94 GFLOPS | Progress: (12/20) | 14.15 s
[Task 8/25] Current/Best: 19.20/ 19.20 GFLOPS | Progress: (16/20) | 16.22 s
[Task 8/25] Current/Best: 20.11/ 20.11 GFLOPS | Progress: (20/20) | 23.18 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.59/ 16.01 GFLOPS | Progress: (4/20) | 18.37 s
[Task 9/25] Current/Best: 23.82/ 23.82 GFLOPS | Progress: (8/20) | 20.11 s
[Task 9/25] Current/Best: 8.40/ 23.82 GFLOPS | Progress: (12/20) | 22.60 s
[Task 9/25] Current/Best: 18.29/ 23.82 GFLOPS | Progress: (16/20) | 25.39 s
[Task 9/25] Current/Best: 9.28/ 23.82 GFLOPS | Progress: (20/20) | 33.73 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.44/ 18.44 GFLOPS | Progress: (4/20) | 2.45 s
[Task 10/25] Current/Best: 15.78/ 18.44 GFLOPS | Progress: (8/20) | 4.08 s
[Task 10/25] Current/Best: 12.51/ 19.14 GFLOPS | Progress: (12/20) | 5.59 s
[Task 10/25] Current/Best: 19.41/ 20.74 GFLOPS | Progress: (16/20) | 6.67 s
[Task 10/25] Current/Best: 8.99/ 20.74 GFLOPS | Progress: (20/20
) | 8.17 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 11.99/ 19.83 GFLOPS | Progress: (4/20) | 3.19 s
[Task 11/25] Current/Best: 17.32/ 19.83 GFLOPS | Progress: (8/20) | 5.91 s
[Task 11/25] Current/Best: 18.56/ 19.83 GFLOPS | Progress: (12/20) | 7.93 s
[Task 11/25] Current/Best: 13.69/ 21.51 GFLOPS | Progress: (16/20) | 10.79 s
[Task 11/25] Current/Best: 19.80/ 21.94 GFLOPS | Progress: (20/20) | 12.84 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.95/ 18.38 GFLOPS | Progress: (4/20) | 5.57 s
[Task 12/25] Current/Best: 5.23/ 18.38 GFLOPS | Progress: (8/20) | 9.42 s
[Task 12/25] Current/Best: 19.21/ 20.54 GFLOPS | Progress: (12/20) | 11.37 s
[Task 12/25] Current/Best: 15.71/ 20.54 GFLOPS | Progress: (16/20) | 14.20 s
[Task 12/25] Current/Best: 15.40/ 20.54 GFLOPS | Progress: (20/20) | 16.08 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.76/ 18.96 GFLOPS | Progress: (4/20) | 3.55 s
[Task 13/25] Current/Best: 16.89/ 21.37 GFLOPS | Progress: (8/20) | 6.05 s
[Task 13/25] Current/Best: 19.89/ 21.86 GFLOPS | Progress: (12/20) | 9.12 s
[Task 13/25] Current/Best: 12.48/ 21.86 GFLOPS | Progress: (16/20) | 12.47 s
[Task 13/25] Current/Best: 19.04/ 21.86 GFLOPS | Progress: (20/20) | 14.77 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 13.94/ 13.94 GFLOPS | Progress: (4/20) | 3.20 s
[Task 14/25] Current/Best: 6.19/ 14.66 GFLOPS | Progress: (8/20) | 5.34 s
[Task 14/25] Current/Best: 19.90/ 19.90 GFLOPS | Progress: (12/20) | 8.01 s
[Task 14/25] Current/Best: 16.95/ 19.90 GFLOPS | Progress: (16/20) | 9.85 s
[Task 14/25] Current/Best: 17.09/ 19.90 GFLOPS | Progress: (20/20) | 11.61 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
Done.
-
[Task 15/25] Current/Best: 16.15/ 17.64 GFLOPS | Progress: (4/20) | 2.66 s
[Task 15/25] Current/Best: 14.21/ 17.99 GFLOPS | Progress: (8/20) | 4.14 s
[Task 15/25] Current/Best: 10.37/ 21.89 GFLOPS | Progress: (12/20) | 6.51 s
[Task 15/25] Current/Best: 20.42/ 21.89 GFLOPS | Progress: (16/20) | 9.76 s
[Task 15/25] Current/Best: 9.70/ 21.89 GFLOPS | Progress: (20/20) | 10.98 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 19.54/ 19.54 GFLOPS | Progress: (4/20) | 3.03 s
[Task 16/25] Current/Best: 3.02/ 19.54 GFLOPS | Progress: (8/20) | 4.67 s
[Task 16/25] Current/Best: 19.02/ 19.54 GFLOPS | Progress: (12/20) | 5.91 s
[Task 16/25] Current/Best: 17.71/ 19.54 GFLOPS | Progress: (16/20) | 7.28 s
[Task 16/25] Current/Best: 10.12/ 21.64 GFLOPS | Progress: (20/20) | 9.46 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 13.58/ 18.86 GFLOPS | Progress: (4/20) | 4.80 s
[Task 17/25] Current/Best: 12.91/ 23.05 GFLOPS | Progress: (8/20) | 7.76 s
[Task 17/25] Current/Best: 16.77/ 23.05 GFLOPS | Progress: (12/20) | 9.79 s
[Task 17/25] Current/Best: 16.51/ 23.05 GFLOPS | Progress: (16/20) | 12.00 s
[Task 17/25] Current/Best: 10.03/ 23.05 GFLOPS | Progress: (20/20) | 14.16 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 11.45/ 18.07 GFLOPS | Progress: (4/20) | 3.80 s
[Task 18/25] Current/Best: 10.62/ 19.80 GFLOPS | Progress: (8/20) | 7.52 s
[Task 18/25] Current/Best: 19.23/ 19.80 GFLOPS | Progress: (12/20) | 9.44 s
[Task 18/25] Current/Best: 9.91/ 19.80 GFLOPS | Progress: (16/20) | 13.35 s
[Task 18/25] Current/Best: 20.51/ 20.51 GFLOPS | Progress: (20/20) | 14.87 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 6.50/ 20.16 GFLOPS | Progress: (4/20) | 6.26 s
[Task 19/25] Current/Best: 2.60/ 20.16 GFLOPS | Progress: (8/20) | 9.60 s
[Task 19/25] Current/Best: 18.77/ 20.72 GFLOPS | Progress: (12/20) | 12.59 s
[Task 19/25] Current/Best: 15.22/ 21.73 GFLOPS | Progress: (16/20) | 15.64 s
[Task 19/25] Current/Best: 2.70/ 22.92 GFLOPS | Progress: (20/20) | 18.45 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 8.94/ 15.12 GFLOPS | Progress: (4/20) | 3.34 s
[Task 20/25] Current/Best: 10.48/ 15.12 GFLOPS | Progress: (8/20) | 6.88 s
[Task 20/25] Current/Best: 2.32/ 16.72 GFLOPS | Progress: (12/20) | 10.89 s
[Task 20/25] Current/Best: 11.11/ 16.72 GFLOPS | Progress: (16/20) | 14.94 s Done.
-
[Task 20/25] Current/Best: 12.56/ 21.61 GFLOPS | Progress: (20/20) | 17.07 s Done.
-
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.39/ 17.53 GFLOPS | Progress: (4/20) | 3.31 s
[Task 21/25] Current/Best: 14.33/ 17.53 GFLOPS | Progress: (8/20) | 4.94 s
[Task 21/25] Current/Best: 1.61/ 17.53 GFLOPS | Progress: (12/20) | 7.08 s
[Task 21/25] Current/Best: 18.11/ 18.11 GFLOPS | Progress: (16/20) | 10.64 s
[Task 21/25] Current/Best: 4.46/ 18.11 GFLOPS | Progress: (20/20) | 18.05 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.70/ 16.94 GFLOPS | Progress: (4/20) | 2.65 s
[Task 22/25] Current/Best: 8.83/ 21.69 GFLOPS | Progress: (8/20) | 4.69 s
[Task 22/25] Current/Best: 19.73/ 21.69 GFLOPS | Progress: (12/20) | 7.08 s
[Task 22/25] Current/Best: 15.36/ 21.69 GFLOPS | Progress: (16/20) | 9.22 s
[Task 22/25] Current/Best: 14.93/ 21.69 GFLOPS | Progress: (20/20) |
10.90 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 17.33/ 20.33 GFLOPS | Progress: (4/20) | 3.20 s
[Task 23/25] Current/Best: 15.62/ 20.33 GFLOPS | Progress: (8/20) | 6.50 s
[Task 23/25] Current/Best: 20.76/ 21.35 GFLOPS | Progress: (12/20) | 8.36 s
[Task 23/25] Current/Best: 6.04/ 21.35 GFLOPS | Progress: (16/20) | 15.50 s
[Task 23/25] Current/Best: 7.62/ 21.35 GFLOPS | Progress: (20/20) | 19.80 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.65/ 8.65 GFLOPS | Progress: (4/20) | 13.64 s
[Task 24/25] Current/Best: 1.98/ 8.65 GFLOPS | Progress: (8/20) | 31.06 s
[Task 24/25] Current/Best: 4.22/ 8.65 GFLOPS | Progress: (12/20) | 55.98 s
[Task 24/25] Current/Best: 7.28/ 8.65 GFLOPS | Progress: (16/20) | 61.79 s Done.
-
[Task 24/25] Current/Best: 3.27/ 8.83 GFLOPS | Progress: (20/20) | 67.82 s Done.
-
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 1.55/ 2.95 GFLOPS | Progress: (4/20) | 32.90 s
[Task 25/25] Current/Best: 5.63/ 7.83 GFLOPS | Progress: (8/20) | 384.42 s
[Task 25/25] Current/Best: 5.94/ 7.83 GFLOPS | Progress: (12/20) | 413.43 s
[Task 25/25] Current/Best: 5.71/ 9.42 GFLOPS | Progress: (16/20) | 415.19 s
[Task 25/25] Current/Best: 2.82/ 9.42 GFLOPS | Progress: (20/20) | 435.52 s
+
[Task 15/25] Current/Best: 17.60/ 17.91 GFLOPS | Progress: (4/20) | 2.52 s
[Task 15/25] Current/Best: 14.71/ 18.37 GFLOPS | Progress: (8/20) | 3.99 s
[Task 15/25] Current/Best: 10.51/ 22.46 GFLOPS | Progress: (12/20) | 6.18 s
[Task 15/25] Current/Best: 22.40/ 22.46 GFLOPS | Progress: (16/20) | 9.26 s
[Task 15/25] Current/Best: 9.84/ 22.46 GFLOPS | Progress: (20/20) | 10.42 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 20.86/ 20.86 GFLOPS | Progress: (4/20) | 2.77 s
[Task 16/25] Current/Best: 3.06/ 20.86 GFLOPS | Progress: (8/20) | 4.36 s
[Task 16/25] Current/Best: 19.63/ 20.86 GFLOPS | Progress: (12/20) | 5.55 s
[Task 16/25] Current/Best: 18.35/ 20.86 GFLOPS | Progress: (16/20) | 6.87 s
[Task 16/25] Current/Best: 10.39/ 22.97 GFLOPS | Progress: (20/20) | 8.96 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 13.00/ 19.20 GFLOPS | Progress: (4/20) | 4.64 s
[Task 17/25] Current/Best: 14.67/ 23.82 GFLOPS | Progress: (8/20) | 7.45 s
[Task 17/25] Current/Best: 18.22/ 23.82 GFLOPS | Progress: (12/20) | 9.47 s
[Task 17/25] Current/Best: 18.03/ 23.82 GFLOPS | Progress: (16/20) | 11.63 s
[Task 17/25] Current/Best: 10.22/ 23.82 GFLOPS | Progress: (20/20) | 13.69 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 11.55/ 18.86 GFLOPS | Progress: (4/20) | 3.62 s
[Task 18/25] Current/Best: 10.76/ 19.81 GFLOPS | Progress: (8/20) | 7.21 s
[Task 18/25] Current/Best: 19.24/ 19.81 GFLOPS | Progress: (12/20) | 9.11 s
[Task 18/25] Current/Best: 10.57/ 19.81 GFLOPS | Progress: (16/20) | 12.87 s
[Task 18/25] Current/Best: 21.26/ 21.26 GFLOPS | Progress: (20/20) | 14.32 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 7.49/ 22.41 GFLOPS | Progress: (4/20) | 5.81 s
[Task 19/25] Current/Best: 2.65/ 22.41 GFLOPS | Progress: (8/20) | 9.15 s
[Task 19/25] Current/Best: 20.62/ 22.41 GFLOPS | Progress: (12/20) | 12.09 s
[Task 19/25] Current/Best: 14.16/ 22.41 GFLOPS | Progress: (16/20) | 15.12 s
[Task 19/25] Current/Best: 2.74/ 24.24 GFLOPS | Progress: (20/20) | 17.91 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 9.33/ 15.46 GFLOPS | Progress: (4/20) | 3.18 s
[Task 20/25] Current/Best: 10.12/ 15.46 GFLOPS | Progress: (8/20) | 6.67 s
[Task 20/25] Current/Best: 2.36/ 16.77 GFLOPS | Progress: (12/20) | 10.50 s
[Task 20/25] Current/Best: 12.58/ 16.77 GFLOPS | Progress: (16/20) | 14.34 s
[Task 20/25] Current/Best: 12.36/ 22.74 GFLOPS | Progress: (20/20) | 16.39 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+ Done.
+
[Task 21/25] Current/Best: 6.51/ 19.46 GFLOPS | Progress: (4/20) | 3.14 s
[Task 21/25] Current/Best: 14.90/ 19.46 GFLOPS | Progress: (8/20) | 4.69 s
[Task 21/25] Current/Best: 1.64/ 19.46 GFLOPS | Progress: (12/20) | 6.79 s
[Task 21/25] Current/Best: 18.39/ 19.46 GFLOPS | Progress: (16/20) | 10.18 s
[Task 21/25] Current/Best: 4.54/ 19.46 GFLOPS | Progress: (20/20) | 17.29 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.75/ 18.70 GFLOPS | Progress: (4/20) | 2.54 s
[Task 22/25] Current/Best: 8.80/ 22.31 GFLOPS | Progress: (8/20) | 4.49 s
[Task 22/25] Current/Best: 20.37/ 22.31 GFLOPS | Progress: (12/20) | 6.85 s
[Task 22/25] Current/Best: 15.76/ 22.31 GFLOPS | Progress: (16/20) | 8.92 s
[Task 22/25] Current/Best: 13.44/ 22.31 GFLOPS | Progress: (20/20) | 10.61 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 17.93/ 21.30 GFLOPS | Progress: (4/20) | 3.10 s
[Task 23/25] Current/Best: 14.31/ 21.63 GFLOPS | Progress: (8/20) | 6.51 s
[Task 23/25] Current/Best: 21.01/ 21.94 GFLOPS | Progress: (12/20) | 8.33 s
[Task 23/25] Current/Best: 6.51/ 21.94 GFLOPS | Progress: (16/20) | 15.44 s
[Task 23/25] Current/Best: 7.67/ 21.94 GFLOPS | Progress: (20/20) | 19.66 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.53/ 8.53 GFLOPS | Progress: (4/20) | 13.22 s
[Task 24/25] Current/Best: 3.66/ 8.53 GFLOPS | Progress: (8/20) | 28.90 s
[Task 24/25] Current/Best: 4.50/ 8.53 GFLOPS | Progress: (12/20) | 52.64 s
[Task 24/25] Current/Best: 6.11/ 9.00 GFLOPS | Progress: (16/20) | 58.23 s
[Task 24/25] Current/Best: 3.40/ 9.00 GFLOPS | Progress: (20/20) | 64.24 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+ Done.
+
[Task 25/25] Current/Best: 1.55/ 2.72 GFLOPS | Progress: (4/20) | 30.65 s
[Task 25/25] Current/Best: 6.30/ 8.44 GFLOPS | Progress: (8/20) | 326.72 s
[Task 25/25] Current/Best: 6.22/ 8.44 GFLOPS | Progress: (12/20) | 354.75 s
[Task 25/25] Current/Best: 6.03/ 9.05 GFLOPS | Progress: (16/20) | 356.54 s
[Task 25/25] Current/Best: 2.86/ 9.05 GFLOPS | Progress: (20/20) | 376.27 s
The output from this tuning process will look something like this:
@@ -660,8 +660,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 410.6132505199639, 'median': 410.6432591000157, 'std': 0.7856641110521063}
- unoptimized: {'mean': 497.0214020900312, 'median': 496.80926645005457, 'std': 1.3955343831042604}
+ optimized: {'mean': 402.6397288399676, 'median': 402.2587899999053, 'std': 1.1484816420935953}
+ unoptimized: {'mean': 483.11649433006096, 'median': 485.0085570498777, 'std': 8.787907520273365}
@@ -681,7 +681,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 17 minutes 27.703 seconds)
+ **Total running time of the script:** ( 16 minutes 3.197 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 2946b906f..f8c071bbb 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -244,7 +244,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.316e-07 secs/op
+ 1.428e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 120a34e79..d06c3817e 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -233,7 +233,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x12608a60)), stage(b, placeholder(b, 0x23179c70)), 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, 0xbb751b0)), stage(b, placeholder(b, 0x20dca650)), 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 707c9fbba..5ac59ae98 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
=================
-**20:31.276** total execution time for **tutorial** files:
+**18:53.193** total execution time for **tutorial** files:
-- **17:27.703**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
-- **01:10.194**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
-- **00:59.951**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
-- **00:26.839**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
-- **00:24.787**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
-- **00:00.748**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
-- **00:00.614**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
-- **00:00.243**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
-- **00:00.051**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
-- **00:00.049**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
-- **00:00.049**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
-- **00:00.048**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
+- **16:03.197**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
+- **01:01.006**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
+- **00:58.669**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
+- **00:25.468**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
+- **00:23.307**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
+- **00:00.687**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
+- **00:00.545**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
+- **00:00.178**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
+- **00:00.039**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
+- **00:00.035**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
+- **00:00.033**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
+- **00:00.030**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index 10f163ebb..730ca596f 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -344,7 +344,7 @@ compile and run this new schedule with the parallel operation applied:
.. code-block:: none
- parallel: 0.000009
+ parallel: 0.000007
@@ -447,10 +447,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 8.202829994843342e-06 1.0
- naive 6.4613e-06 0.7876915654794572
- parallel 9.1397e-06 1.1142130223039628
- vector 2.54533e-05 3.1029900675743685
+ numpy 8.412130009674001e-06 1.0
+ naive 5.8355e-06 0.6937006433910483
+ parallel 6.7793000000000005e-06 0.8058957710120699
+ vector 2.45968e-05 2.9239681236159605
@@ -839,7 +839,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.019028
+ Numpy running time: 0.017713
@@ -897,7 +897,7 @@ optimizations.
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
- none: 3.283674
+ none: 3.437374
@@ -996,7 +996,7 @@ schedule.
.. code-block:: none
- blocking: 0.320353
+ blocking: 0.297494
@@ -1088,7 +1088,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.341757
+ vectorization: 0.332753
@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], []),
@@ -1160,7 +1160,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.138530
+ loop permutation: 0.114928
@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], []),
@@ -1257,7 +1257,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.112607
+ array packing: 0.109230
@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], []),
@@ -1348,7 +1348,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.112169
+ block caching: 0.109880
@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], []),
@@ -1432,7 +1432,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.143844
+ parallelization: 0.144129
@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], []),
@@ -1511,13 +1511,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.2836744515 1.0
- blocking 0.32035340999999995 0.097559430671838
- vectorization 0.3417574647 0.1040777548894603
- loop permutation 0.13853012120000002 0.04218753206083469
- array packing 0.11260731929999998 0.03429308263142845
- block caching 0.1121686897 0.03415950373788143
- parallelization 0.1438435707 0.043805673438270194
+ none 3.4373738605999997 1.0
+ blocking 0.29749398039999997 0.08654687923532178
+ vectorization 0.3327531902 0.09680448030809118
+ loop permutation 0.1149282127 0.03343488877288986
+ array packing 0.1092296609 0.03177706741533601
+ block caching 0.10988034279999999 0.03196636364157961
+ parallelization 0.1441291052 0.04193000559294473
@@ -1552,6 +1552,11 @@ 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 1.006 seconds)
+
+
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 534ab930f..2bd04a158 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-8a93eaffd193c824ee320e71e8a8049d2c0d2ef0
+814f5501bf7d65f759135d214572388b0ddadefc
diff --git a/docs/genindex.html b/docs/genindex.html
index c08a49193..25d090bde 100644
--- a/docs/genindex.html
+++ b/docs/genindex.html
@@ -1553,6 +1553,8 @@
<li><a href="reference/api/python/ir.html#tvm.ir.IRModule.from_expr">from_expr() (tvm.ir.IRModule static method)</a>
</li>
<li><a href="reference/api/python/tir.html#tvm.tir.IndexMap.from_func">from_func() (tvm.tir.IndexMap static method)</a>
+</li>
+ <li><a href="reference/api/python/tir.html#tvm.tir.IndexMap.from_func_with_separators">from_func_with_separators() (tvm.tir.IndexMap static method)</a>
</li>
<li><a href="reference/api/python/runtime.html#tvm.runtime.Report.from_json">from_json() (tvm.runtime.Report class method)</a>
</li>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 9d288380d..900dde209 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -401,7 +401,7 @@
</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.zipb1879542-46bf-46e7-b009-9e5784efc22b 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.zip3b17ef11-bae1-49ae-ad3c-9de9e65665f0 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
x (1, 3, 224, 224)
</pre></div>
</div>
diff --git a/docs/how_to/compile_models/from_oneflow.html b/docs/how_to/compile_models/from_oneflow.html
index 24ea6703a..876590d3a 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -406,49 +406,45 @@ python3 -m pip install -f https://release.oneflow.info <span class="nv">oneflow<
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
0%| | 0.00/41.5M [00:00<?, ?B/s]
- 0%| | 16.0k/41.5M [00:00<07:48, 92.7kB/s]
- 0%| | 48.0k/41.5M [00:00<04:56, 147kB/s]
- 0%| | 96.0k/41.5M [00:00<03:30, 206kB/s]
- 0%| | 160k/41.5M [00:00<02:40, 270kB/s]
- 1%| | 312k/41.5M [00:00<01:28, 489kB/s]
- 1%|1 | 624k/41.5M [00:01<00:45, 934kB/s]
- 3%|2 | 1.21M/41.5M [00:01<00:23, 1.79MB/s]
- 6%|5 | 2.42M/41.5M [00:01<00:11, 3.49MB/s]
- 9%|9 | 3.89M/41.5M [00:01<00:07, 5.11MB/s]
- 13%|#2 | 5.37M/41.5M [00:01<00:06, 6.22MB/s]
- 16%|#6 | 6.84M/41.5M [00:01<00:04, 7.98MB/s]
- 19%|#9 | 7.91M/41.5M [00:01<00:04, 8.66MB/s]
- 21%|##1 | 8.82M/41.5M [00:02<00:04, 7.88MB/s]
- 24%|##3 | 9.77M/41.5M [00:02<00:03, 8.36MB/s]
- 26%|##6 | 10.8M/41.5M [00:02<00:03, 9.01MB/s]
- 28%|##8 | 11.8M/41.5M [00:02<00:03, 8.06MB/s]
- 31%|### | 12.7M/41.5M [00:02<00:03, 8.18MB/s]
- 34%|###4 | 14.2M/41.5M [00:02<00:03, 9.37MB/s]
- 36%|###6 | 15.1M/41.5M [00:02<00:03, 9.08MB/s]
- 38%|###8 | 16.0M/41.5M [00:03<00:03, 7.68MB/s]
- 41%|####1 | 17.1M/41.5M [00:03<00:03, 7.46MB/s]
- 45%|####4 | 18.6M/41.5M [00:03<00:03, 7.87MB/s]
- 48%|####8 | 20.1M/41.5M [00:03<00:02, 8.14MB/s]
- 52%|#####1 | 21.6M/41.5M [00:03<00:02, 8.31MB/s]
- 55%|#####5 | 23.0M/41.5M [00:03<00:02, 9.46MB/s]
- 58%|#####7 | 24.1M/41.5M [00:03<00:01, 9.68MB/s]
- 60%|###### | 25.0M/41.5M [00:04<00:01, 8.89MB/s]
- 63%|######2 | 25.9M/41.5M [00:04<00:01, 8.91MB/s]
- 65%|######5 | 27.0M/41.5M [00:04<00:01, 9.30MB/s]
- 67%|######7 | 27.9M/41.5M [00:04<00:01, 8.45MB/s]
- 70%|######9 | 28.9M/41.5M [00:04<00:01, 8.75MB/s]
- 72%|#######2 | 29.9M/41.5M [00:04<00:01, 9.18MB/s]
- 74%|#######4 | 30.8M/41.5M [00:04<00:01, 8.30MB/s]
- 77%|#######6 | 31.8M/41.5M [00:04<00:01, 8.41MB/s]
- 80%|######## | 33.2M/41.5M [00:05<00:00, 9.73MB/s]
- 82%|########2 | 34.2M/41.5M [00:05<00:00, 9.11MB/s]
- 85%|########4 | 35.1M/41.5M [00:05<00:00, 7.76MB/s]
- 87%|########7 | 36.2M/41.5M [00:05<00:00, 7.53MB/s]
- 91%|######### | 37.7M/41.5M [00:05<00:00, 9.26MB/s]
- 93%|#########3| 38.7M/41.5M [00:05<00:00, 9.33MB/s]
- 95%|#########5| 39.6M/41.5M [00:05<00:00, 8.33MB/s]
- 98%|#########7| 40.7M/41.5M [00:06<00:00, 7.61MB/s]
-100%|##########| 41.5M/41.5M [00:06<00:00, 7.19MB/s]
+ 0%| | 16.0k/41.5M [00:00<07:56, 91.3kB/s]
+ 0%| | 48.0k/41.5M [00:00<05:00, 144kB/s]
+ 0%| | 96.0k/41.5M [00:00<03:33, 203kB/s]
+ 0%| | 144k/41.5M [00:00<03:08, 230kB/s]
+ 1%| | 296k/41.5M [00:00<01:34, 458kB/s]
+ 1%|1 | 592k/41.5M [00:01<00:49, 873kB/s]
+ 3%|2 | 1.11M/41.5M [00:01<00:26, 1.59MB/s]
+ 5%|5 | 2.22M/41.5M [00:01<00:13, 3.14MB/s]
+ 9%|8 | 3.70M/41.5M [00:01<00:08, 4.85MB/s]
+ 12%|#2 | 5.18M/41.5M [00:01<00:06, 6.02MB/s]
+ 16%|#6 | 6.66M/41.5M [00:01<00:05, 6.80MB/s]
+ 20%|#9 | 8.13M/41.5M [00:02<00:04, 7.34MB/s]
+ 23%|##3 | 9.61M/41.5M [00:02<00:04, 7.71MB/s]
+ 27%|##6 | 11.1M/41.5M [00:02<00:03, 7.98MB/s]
+ 30%|### | 12.6M/41.5M [00:02<00:03, 8.16MB/s]
+ 34%|###3 | 14.0M/41.5M [00:02<00:03, 8.29MB/s]
+ 37%|###7 | 15.5M/41.5M [00:03<00:03, 8.36MB/s]
+ 41%|#### | 17.0M/41.5M [00:03<00:03, 8.43MB/s]
+ 44%|####4 | 18.5M/41.5M [00:03<00:02, 8.46MB/s]
+ 48%|####8 | 19.9M/41.5M [00:03<00:02, 8.50MB/s]
+ 52%|#####1 | 21.4M/41.5M [00:03<00:02, 8.52MB/s]
+ 55%|#####5 | 22.9M/41.5M [00:03<00:02, 8.54MB/s]
+ 59%|#####8 | 24.4M/41.5M [00:04<00:02, 8.55MB/s]
+ 62%|######2 | 25.8M/41.5M [00:04<00:01, 9.85MB/s]
+ 65%|######4 | 26.9M/41.5M [00:04<00:01, 9.98MB/s]
+ 67%|######7 | 27.9M/41.5M [00:04<00:01, 8.91MB/s]
+ 69%|######9 | 28.8M/41.5M [00:04<00:01, 7.71MB/s]
+ 73%|#######2 | 30.3M/41.5M [00:04<00:01, 9.37MB/s]
+ 75%|#######5 | 31.3M/41.5M [00:04<00:01, 9.55MB/s]
+ 78%|#######7 | 32.3M/41.5M [00:05<00:01, 8.49MB/s]
+ 80%|######## | 33.2M/41.5M [00:05<00:01, 7.50MB/s]
+ 84%|########3 | 34.7M/41.5M [00:05<00:00, 9.17MB/s]
+ 86%|########6 | 35.7M/41.5M [00:05<00:00, 9.44MB/s]
+ 88%|########8 | 36.7M/41.5M [00:05<00:00, 8.37MB/s]
+ 91%|######### | 37.7M/41.5M [00:05<00:00, 7.45MB/s]
+ 94%|#########4| 39.1M/41.5M [00:05<00:00, 9.15MB/s]
+ 97%|#########6| 40.1M/41.5M [00:05<00:00, 9.46MB/s]
+ 99%|#########9| 41.1M/41.5M [00:06<00:00, 8.37MB/s]
+100%|##########| 41.5M/41.5M [00:06<00:00, 7.08MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_paddle.html b/docs/how_to/compile_models/from_paddle.html
index 5f89cddfd..a0b7beca1 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -469,7 +469,7 @@ 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: 'tiger cat',
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 6.130 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 8.766 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 6fa883784..a1d25dfee 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -387,9 +387,9 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
- 39%|###9 | 17.5M/44.7M [00:00<00:00, 184MB/s]
- 88%|########8 | 39.3M/44.7M [00:00<00:00, 210MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 208MB/s]
+ 28%|##7 | 12.4M/44.7M [00:00<00:00, 130MB/s]
+ 67%|######7 | 30.0M/44.7M [00:00<00:00, 162MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 139MB/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 ad0804609..56c3d273b 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -612,7 +612,6 @@ 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 6.315 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 bf0c2dc4b..0ccfd8011 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -300,18 +300,18 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:24.769</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:35.761</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<ul class="simple">
-<li><p><strong>01:06.315</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>01:06.130</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>00:57.446</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>00:30.970</strong>: <a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></li>
-<li><p><strong>00:24.441</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>00:21.955</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>00:21.567</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>00:19.348</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>00:14.099</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>00:02.499</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>01:08.766</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>00:59.888</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>00:57.086</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>00:41.220</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>00:32.201</strong>: <a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></li>
+<li><p><strong>00:20.726</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>00:20.624</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>00:19.288</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>00:13.291</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>00:02.670</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 f5c6c163d..4ed3fbe61 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -627,7 +627,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.1705 16.1716 16.2298 16.1183 0.0334
+ 15.8685 15.8805 16.1054 15.5230 0.1629
</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 511da2794..69bf85872 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -409,14 +409,16 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 8%|8 | 14.0M/170M [00:00<00:01, 147MB/s]
- 18%|#7 | 30.5M/170M [00:00<00:00, 162MB/s]
- 29%|##8 | 48.7M/170M [00:00<00:00, 175MB/s]
- 45%|####4 | 75.7M/170M [00:00<00:00, 218MB/s]
- 60%|###### | 102M/170M [00:00<00:00, 240MB/s]
- 75%|#######4 | 127M/170M [00:00<00:00, 247MB/s]
- 89%|########8 | 151M/170M [00:00<00:00, 247MB/s]
-100%|##########| 170M/170M [00:00<00:00, 231MB/s]
+ 10%|9 | 16.2M/170M [00:00<00:00, 168MB/s]
+ 19%|#8 | 32.2M/170M [00:00<00:01, 141MB/s]
+ 28%|##8 | 48.3M/170M [00:00<00:00, 152MB/s]
+ 37%|###7 | 63.0M/170M [00:00<00:00, 142MB/s]
+ 45%|####5 | 76.8M/170M [00:00<00:00, 130MB/s]
+ 59%|#####8 | 99.6M/170M [00:00<00:00, 163MB/s]
+ 73%|#######2 | 123M/170M [00:00<00:00, 188MB/s]
+ 85%|########4 | 144M/170M [00:00<00:00, 197MB/s]
+ 96%|#########6| 163M/170M [00:01<00:00, 174MB/s]
+100%|##########| 170M/170M [00:01<00:00, 167MB/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').
@@ -514,7 +516,7 @@ 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 14.835 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 54.422 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 af04c95e8..4448a9bb5 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -450,7 +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: "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]
-100%|##########| 13.6M/13.6M [00:00<00:00, 176MB/s]
+ 34%|###3 | 4.56M/13.6M [00:00<00:00, 46.5MB/s]
+ 66%|######6 | 9.01M/13.6M [00:00<00:00, 33.4MB/s]
+ 91%|#########1| 12.4M/13.6M [00:00<00:00, 22.2MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 25.2MB/s]
</pre></div>
</div>
</div>
@@ -544,7 +547,7 @@ 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)
- 90.5915 90.4739 91.5880 90.2648 0.2938
+ 88.0771 87.9550 89.8937 87.7442 0.3431
</pre></div>
</div>
<div class="admonition note">
@@ -583,7 +586,7 @@ 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 7.292 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 3.247 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 277bba100..4f108f291 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -545,7 +545,7 @@ 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)
- 123.6896 123.6542 124.7379 123.0264 0.4053
+ 115.5216 115.2153 121.6813 114.2261 1.1246
</pre></div>
</div>
<div class="admonition note">
@@ -573,7 +573,7 @@ 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 1.358 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 58.743 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 f9130324e..7618645fb 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -482,7 +482,7 @@ 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 37.312 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 16.460 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 4785cd61c..d987de4ad 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -415,23 +415,25 @@ to your device.</p>
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
0%| | 0/132723 [00:00<?, ?KB/s]
- 2%|2 | 3029/132723 [00:00<00:04, 30287.14KB/s]
- 6%|5 | 7561/132723 [00:00<00:03, 36819.82KB/s]
- 9%|8 | 11789/132723 [00:00<00:03, 39229.18KB/s]
- 15%|#5 | 20312/132723 [00:00<00:01, 57019.24KB/s]
- 22%|##1 | 28979/132723 [00:00<00:01, 67569.57KB/s]
- 28%|##8 | 37708/132723 [00:00<00:01, 74215.28KB/s]
- 35%|###5 | 46489/132723 [00:00<00:01, 78629.73KB/s]
- 42%|####1 | 55274/132723 [00:00<00:00, 81551.82KB/s]
- 48%|####8 | 64118/132723 [00:00<00:00, 83697.12KB/s]
- 55%|#####4 | 72765/132723 [00:01<00:00, 84547.73KB/s]
- 61%|######1 | 81621/132723 [00:01<00:00, 85770.83KB/s]
- 68%|######8 | 90437/132723 [00:01<00:00, 86494.75KB/s]
- 75%|#######4 | 99291/132723 [00:01<00:00, 87111.53KB/s]
- 81%|########1 | 108006/132723 [00:01<00:00, 82950.03KB/s]
- 88%|########7 | 116784/132723 [00:01<00:00, 84353.61KB/s]
- 95%|#########4| 125589/132723 [00:01<00:00, 85437.12KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 77615.30KB/s]
+ 5%|4 | 6112/132723 [00:00<00:02, 61115.54KB/s]
+ 11%|#1 | 14734/132723 [00:00<00:01, 75879.60KB/s]
+ 17%|#6 | 22322/132723 [00:00<00:01, 57385.81KB/s]
+ 23%|##3 | 30849/132723 [00:00<00:01, 66594.41KB/s]
+ 29%|##8 | 37940/132723 [00:00<00:01, 47783.49KB/s]
+ 34%|###3 | 45094/132723 [00:00<00:01, 53567.03KB/s]
+ 39%|###8 | 51247/132723 [00:00<00:01, 44700.90KB/s]
+ 45%|####4 | 59383/132723 [00:01<00:01, 53080.04KB/s]
+ 49%|####9 | 65523/132723 [00:01<00:01, 50456.66KB/s]
+ 55%|#####5 | 73081/132723 [00:01<00:01, 56572.75KB/s]
+ 60%|#####9 | 79285/132723 [00:01<00:00, 56407.72KB/s]
+ 64%|######4 | 85304/132723 [00:01<00:00, 51535.07KB/s]
+ 70%|####### | 93194/132723 [00:01<00:00, 58455.23KB/s]
+ 75%|#######4 | 99395/132723 [00:01<00:00, 48942.51KB/s]
+ 81%|########1 | 107525/132723 [00:01<00:00, 56644.15KB/s]
+ 86%|########5 | 113724/132723 [00:02<00:00, 55712.46KB/s]
+ 90%|######### | 119663/132723 [00:02<00:00, 42008.93KB/s]
+ 96%|#########6| 127879/132723 [00:02<00:00, 50624.68KB/s]
+100%|##########| 132723/132723 [00:02<00:00, 53534.86KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -476,7 +478,7 @@ 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 28.269 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 20.338 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 1c4ce9363..1cdc163ec 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>11:22.479</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>10:22.528</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<ul class="simple">
-<li><p><strong>03:14.835</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>02:28.269</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>02:01.358</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>01:37.312</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>01:07.292</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>00:30.085</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>00:23.120</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>00:00.208</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>02:54.422</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>02:20.338</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>01:58.743</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>01:16.460</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>01:03.247</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>00:27.628</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>00:21.511</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>00:00.179</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 1f0681db5..1e32a13ca 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -590,7 +590,7 @@ 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.zip85795ee8-f62f-4165-9871-80cddc17ab20 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.zip50c0d070-59e0-4d71-9505-d0c79d0c4b7e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
</pre></div>
</div>
<p>It’s easy to execute MobileNet with native TVM:</p>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index 694dab1e4..6148c9778 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>00:39.421</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:37.391</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:35.744</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>00:02.352</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>00:01.104</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>00:00.221</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>00:33.995</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>00:02.204</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>00:01.005</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>00:00.186</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 30784bf9a..67007b785 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: 6065us [6065us] (40.24%; 40.24%)
-FoldScaleAxis: 9006us [3us] (59.76%; 59.76%)
- FoldConstant: 9003us [1500us] (59.74%; 99.97%)
- InferType: 7502us [7502us] (49.78%; 83.34%)
+InferType: 6002us [6002us] (45.63%; 45.63%)
+FoldScaleAxis: 7153us [2us] (54.37%; 54.37%)
+ FoldConstant: 7151us [1454us] (54.36%; 99.97%)
+ InferType: 5697us [5697us] (43.31%; 79.67%)
</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: 9563us [9563us] (47.50%; 47.50%)
-FoldScaleAxis: 10570us [3us] (52.50%; 52.50%)
- FoldConstant: 10567us [2109us] (52.49%; 99.97%)
- InferType: 8459us [8459us] (42.01%; 80.05%)
+InferType: 5739us [5739us] (44.57%; 44.57%)
+FoldScaleAxis: 7137us [2us] (55.43%; 55.43%)
+ FoldConstant: 7136us [1498us] (55.41%; 99.97%)
+ InferType: 5637us [5637us] (43.78%; 79.00%)
</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 638bc0bce..38723bee0 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -538,7 +538,7 @@ latency of convolution.</p>
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-Convolution: 37.517698 ms
+Convolution: 54.199556 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 4673aa0d1..e540c7d07 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -882,7 +882,7 @@ be able to run on our build server</p>
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-conv2d with tensor core: 7.367135 ms
+conv2d with tensor core: 9.389245 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 cd28db151..21366b896 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -431,10 +431,10 @@ 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: 0.019321
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018460
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-Baseline: 3.301291
+Baseline: 3.445202
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -496,7 +496,7 @@ 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: 0.317786
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.295180
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -565,7 +565,7 @@ 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: 0.346788
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.333483
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -628,7 +628,7 @@ 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: 0.132057
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.114889
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -713,7 +713,7 @@ 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: 0.111363
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.110148
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -801,7 +801,7 @@ 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: 0.112277
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111236
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -893,7 +893,7 @@ 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: 0.145991
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.145570
</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 8a82acb25..6c3623db2 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>00:35.509</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.098</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:32.787</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>00:01.492</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>00:01.230</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>00:32.373</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>00:01.482</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>00:01.243</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 957b68731..fc2e2c538 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>05:09.723</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>04:55.021</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<ul class="simple">
-<li><p><strong>02:22.075</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>01:20.816</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>00:41.277</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>00:27.315</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>00:09.415</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>00:08.825</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>02:25.955</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>01:16.455</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>00:39.267</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>00:16.716</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>00:08.559</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>00:08.070</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 b03fd1e78..99b7b9893 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
@@ -470,74 +470,483 @@ cooperative fetching, unrolling and operator fusion.</p>
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 112;
- allocate(conv2d_nchw: Pointer(local float32), float32, [2]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [54]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [576]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [2], [], scope="local", align=8)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 28;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[1] = 0f32
- for (rc.outer.outer: int32, 0, 256) {
- let cse_var_1: int32 = (rc.outer.outer*18)
- {
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- if @tir.likely((threadIdx.x_1 < 54), dtype=bool) {
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [54], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((3 <= floormod(threadIdx.x_1, 27)) && (floormod(threadIdx.x_1, 27) < 24)) && (1 <= (floormod(blockIdx.x, 7) + floormod(threadIdx.x_1, 3)))) && ((floormod(blockIdx.x, 7) + floormod(threadIdx.x_1, 3)) < 8)), data[((((((rc.outer.outer*98) + (floordiv(threadIdx.x_1, 27)*49)) + (floordiv(floormod(threadIdx.x_1, 27), 3)*7)) [...]
+ conv2d_nchw_1[2] = 0f32
+ conv2d_nchw_1[3] = 0f32
+ conv2d_nchw_1[4] = 0f32
+ conv2d_nchw_1[5] = 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) {
+ for (ry.outer.outer: int32, 0, 3) {
+ let cse_var_2: int32 = (rc.outer.outer*72)
+ let cse_var_1: int32 = (ry.outer.outer*3)
+ {
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
+ if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [72], [], scope="shared")[(threadIdx.x_1*4)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod((threadIdx.x_1*4), 9))) && (floormod((threadIdx.x_1*4), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*4), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*4) + 1)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 1), 9))) && (floormod(((threadIdx.x_1*4) + 1), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 1), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 1), 9)) - 8)], 0 [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*4) + 2)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 2), 9))) && (floormod(((threadIdx.x_1*4) + 2), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 2), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 2), 9)) - 8)], 0 [...]
+ }
+ if @tir.likely((threadIdx.x_1 < 18), dtype=bool) {
+ pad_temp.shared_1[((threadIdx.x_1*4) + 3)] = @tir.if_then_else(((((1 <= (ry.outer.outer + floormod(blockIdx.x, 7))) && ((ry.outer.outer + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*4) + 3), 9))) && (floormod(((threadIdx.x_1*4) + 3), 9) < 8)), data[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*4) + 3), 9)*49)) + (ry.outer.outer*7)) + (floormod(blockIdx.x, 7)*7)) + floormod(((threadIdx.x_1*4) + 3), 9)) - 8)], 0 [...]
+ }
+ }
+ 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" = 112;
- kernel.shared_1: Buffer(kernel.shared, float32, [576], [], scope="shared")[threadIdx.x_2] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv(threadIdx.x_2, 18)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 56), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 4), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 112), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 8), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 168), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 12), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 224), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 16), 18))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
- if @tir.likely((threadIdx.x_2 < 16), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[((((floordiv(blockIdx.x, 7)*147456) + (floordiv((floordiv(threadIdx.x_2, 2) + 280), 9)*4608)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 18))]
- }
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*3)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*36)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 21)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 24)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 27)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 30)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 30)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 30)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 33)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 33)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 22)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 25)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 28)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 31)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 31)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 31)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 34)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 34)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 23)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 26)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 29)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 32)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 32)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 32)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*3) + 35)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*36) + 35)]))
}
}
for (i1.inner: int32, 0, 2) {
- compute[(((((floordiv(blockIdx.x, 7)*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (floormod(threadIdx.x, 7)*7)) + floormod(blockIdx.x, 7))] = max((conv2d_nchw_1[i1.inner] + bias[(((floordiv(blockIdx.x, 7)*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
+ 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)
+ }
}
}
}
@@ -579,7 +988,7 @@ cooperative fetching, unrolling and operator fusion.</p>
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-Execution time of this operator: 0.419 ms
+Execution time of this operator: 0.361 ms
</pre></div>
</div>
</div>
@@ -609,21 +1018,21 @@ conv2d_nchw_nn_o_i, conv2d_nchw_nn_i = s[conv2d_nchw].split(conv2d_nchw_nn, fact
conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
-conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
-conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
+conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=64)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
-conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
+conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
-conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=1)
-conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
@@ -632,12 +1041,12 @@ compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
-compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
-compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
@@ -658,12 +1067,12 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
+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=112)
+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", 512)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
@@ -685,65 +1094,430 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[2];
- __shared__ float pad_temp_shared[54];
- __shared__ float kernel_shared[576];
+extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[14];
+ __shared__ float pad_temp_shared[72];
+ __shared__ float kernel_shared[3072];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 256; ++rc_outer_outer) {
- __syncthreads();
- if (((int)threadIdx.x) < 54) {
- pad_temp_shared[((int)threadIdx.x)] = (((((3 <= (((int)threadIdx.x) % 27)) && ((((int)threadIdx.x) % 27) < 24)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 98) + ((((int)threadIdx.x) / 27) * 49)) + (((((int)threadIdx.x) % 27) / 3) * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) - 8)] : 0.000000e+00f);
- }
- kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + ((((int)threadIdx.x) / 18) * 4608)) + (rc_outer_outer * 18)) + (((int)threadIdx.x) % 18))];
- kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 112) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 4) % 18))];
- kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 224) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 8) % 18))];
- kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 336) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 12) % 18))];
- kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 448) / 18) * 4608)) + (rc_outer_outer * 18)) + ((((int)threadIdx.x) + 16) % 18))];
- if (((int)threadIdx.x) < 16) {
- kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 560) / 18) * 4608)) + (rc_outer_outer * 18)) + (((int)threadIdx.x) + 2))];
+ conv2d_nchw[2] = 0.000000e+00f;
+ conv2d_nchw[3] = 0.000000e+00f;
+ conv2d_nchw[4] = 0.000000e+00f;
+ conv2d_nchw[5] = 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) {
+ 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)]));
}
- __syncthreads();
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[((((int)threadIdx.x) / 7) * 36)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 18)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 3)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 21)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 6)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 24)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 9)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 27)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 27)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 30)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 12)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 30)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 30)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 33)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 15)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 33)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 33)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 1)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 19)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 4)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 22)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 7)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 25)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 10)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 28)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 28)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 31)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 13)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 31)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 31)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 34)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 16)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 34)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 34)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 2)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 20)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 5)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 23)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 8)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 26)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 11)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 29)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 29)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 32)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 14)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 32)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 32)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 17)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 35)] * kernel_shared[(((((int)threadIdx.x) / 7) * 36) + 35)]));
}
for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i1_inner] + bias[((((((int)blockIdx.x) / 7) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+ 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);
+ }
}
}
</pre></div>
@@ -781,7 +1555,7 @@ 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 22.075 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 25.955 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 d80a6eb91..feac5468b 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -878,7 +878,7 @@ 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)
- 10.1293 10.1448 10.1897 10.0533 0.0567
+ 9.9674 9.9812 10.0124 9.9087 0.0434
</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 890f635c9..933322dbe 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -897,7 +897,7 @@ 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)
- 768.4319 769.6622 773.6825 761.9511 4.8677
+ 736.8305 737.5163 738.4373 734.5379 1.6642
</pre></div>
</div>
</div>
@@ -919,7 +919,7 @@ 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 20.816 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 16.455 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 ce72c63a6..8be7f99db 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -600,32 +600,76 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_5: placeholder_15: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_7: placeholder_19: Buffer(placeholder_12, int32, [4916], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 16) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [4096]), storage_scope = global {
- for (i.outer.inner: int32, 0, 2) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 64) {
- for (j.init: int32, 0, 16) {
- compute_5: Buffer(compute_4, float32, [4096], [])[((((i.outer.inner*2048) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
- }
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_17: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_19: Buffer(placeholder_14, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 64) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global {
+ for (i.outer.inner: int32, 0, 8) {
+ for (i.inner.init: int32, 0, 8) {
+ let cse_var_1: int32 = ((i.outer.inner*128) + (i.inner.init*16))
+ {
+ compute_5: Buffer(compute_4, float32, [1024], [])[cse_var_1] = 0f32
+ compute_5[(cse_var_1 + 1)] = 0f32
+ compute_5[(cse_var_1 + 2)] = 0f32
+ compute_5[(cse_var_1 + 3)] = 0f32
+ compute_5[(cse_var_1 + 4)] = 0f32
+ compute_5[(cse_var_1 + 5)] = 0f32
+ compute_5[(cse_var_1 + 6)] = 0f32
+ compute_5[(cse_var_1 + 7)] = 0f32
+ compute_5[(cse_var_1 + 8)] = 0f32
+ compute_5[(cse_var_1 + 9)] = 0f32
+ compute_5[(cse_var_1 + 10)] = 0f32
+ compute_5[(cse_var_1 + 11)] = 0f32
+ compute_5[(cse_var_1 + 12)] = 0f32
+ compute_5[(cse_var_1 + 13)] = 0f32
+ compute_5[(cse_var_1 + 14)] = 0f32
+ compute_5[(cse_var_1 + 15)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_1: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- for (i.inner: int32, 0, 64) {
- for (j: int32, 0, 16) {
- let cse_var_3: int32 = ((i0.outer.i1.outer.fused*2) + nb_j.inner)
- let cse_var_2: int32 = ((((i.outer.inner*2048) + (i.inner*32)) + (nb_j.inner*16)) + j)
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[(((i.outer.inner*16384) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
- }
+ }
+ for (elem_idx: int32, 0, let cse_var_2: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_2 + 1)] - placeholder_3[cse_var_2])) {
+ for (i.inner: int32, 0, 8) {
+ let cse_var_21: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ let cse_var_20: int32 = (elem_idx*16)
+ let cse_var_19: int32 = ((i.outer.inner*128) + (i.inner*16))
+ let cse_var_18: int32 = (cse_var_19 + 10)
+ let cse_var_17: int32 = (cse_var_19 + 11)
+ let cse_var_16: int32 = (cse_var_19 + 12)
+ let cse_var_15: int32 = (cse_var_19 + 13)
+ let cse_var_14: int32 = (cse_var_19 + 14)
+ let cse_var_13: int32 = (cse_var_19 + 15)
+ let cse_var_12: int32 = (cse_var_19 + 2)
+ let cse_var_11: int32 = (cse_var_19 + 3)
+ let cse_var_10: int32 = (cse_var_19 + 4)
+ let cse_var_9: int32 = (cse_var_19 + 5)
+ let cse_var_8: int32 = (cse_var_19 + 6)
+ let cse_var_7: int32 = (cse_var_19 + 7)
+ let cse_var_6: int32 = (cse_var_19 + 8)
+ let cse_var_5: int32 = (cse_var_19 + 9)
+ let cse_var_4: int32 = (cse_var_19 + 1)
+ let cse_var_3: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i.outer.inner*2048)) + (i.inner*256))
+ {
+ compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[((placeholder_3[cse_var_21]*16) + cse_var_20)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 1)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 2)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 3)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 4)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 5)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 6)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 7)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 8)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 9)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 10)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 11)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 12)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 13)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 14)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
+ compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_21]*16) + cse_var_20) + 15)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_21] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 128) {
- for (i1.inner: int32, 0, 32) {
- let cse_var_4: int32 = (((i0.inner*512) + (i0.outer.i1.outer.fused*32)) + i1.inner)
- compute[cse_var_4] = max((compute_5[((i0.inner*32) + i1.inner)] + placeholder_4[cse_var_4]), 0f32)
- }
+ for (i0.inner: int32, 0, 64) {
+ let cse_var_22: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*32768) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_22, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_22, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -666,7 +710,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-Execution time of this operator: 1.839 ms
+Execution time of this operator: 1.688 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 50b941e9d..b98669dc0 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>00:44.713</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:44.226</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:43.770</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>00:00.248</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>00:00.233</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>00:00.232</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>00:00.230</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>00:43.390</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>00:00.227</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>00:00.213</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>00:00.200</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>00:00.196</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 c512c9187..c2100c50c 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 "/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: 110.31/110.31 result: MeasureResult(costs=(0.0020986534583333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6601579189300537, timestamp=1653441326.2048783) [('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: 0.00/110.31 result: Traceback (most recent call last):
+No: 6 GFLOPS: 100.05/100.05 result: MeasureResult(costs=(0.0023137704375,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.686079978942871, timestamp=1653528323.4934978) [('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: 0.00/100.05 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
@@ -1266,7 +1266,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 8 GFLOPS: 0.00/100.05 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
@@ -1389,7 +1389,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 9 GFLOPS: 0.00/100.05 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
@@ -1512,7 +1512,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 10 GFLOPS: 0.00/100.05 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
@@ -1530,7 +1530,7 @@ No: 10 GFLOPS: 0.00/110.31 result: Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 11 GFLOPS: 0.00/100.05 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
@@ -1653,7 +1653,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 12 GFLOPS: 0.00/100.05 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
@@ -1776,7 +1776,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 13 GFLOPS: 0.00/100.05 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
@@ -1899,7 +1899,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 14 GFLOPS: 0.00/100.05 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
@@ -2022,7 +2022,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 15 GFLOPS: 0.00/100.05 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
@@ -2145,7 +2145,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 16 GFLOPS: 0.00/100.05 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
@@ -2268,7 +2268,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 17 GFLOPS: 0.00/100.05 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
@@ -2391,7 +2391,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 18 GFLOPS: 0.00/100.05 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
@@ -2514,7 +2514,7 @@ Traceback (most recent call last):
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: 0.00/110.31 result: Traceback (most recent call last):
+No: 19 GFLOPS: 0.00/100.05 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
@@ -2602,7 +2602,7 @@ tvm._ffi.base.TVMError: Traceback (most recent call last):
15: _PyEval_EvalFrameDefault
14: 0x0000000000537c30
13: _PyObject_FastCallKeywords
- 12: 0x00007f88e4504fa2
+ 12: 0x00007f0caee70fa2
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 [('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: 142.02/142.02 result: MeasureResult(costs=(0.0016300953799999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4533238410949707, timestamp=1653441352.8289824) [('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: 143.07/143.07 result: MeasureResult(costs=(0.0016180540999999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4196603298187256, timestamp=1653528349.8571422) [('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
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2710,7 +2710,7 @@ and measure running time.</p>
"target_host parameter is going to be deprecated. "
/workspace/python/tvm/target/target.py:317: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
"target_host parameter is going to be deprecated. "
-Time cost of this operator: 0.002079
+Time cost of this operator: 0.001979
</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 888e820a7..9845148da 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -555,10 +555,10 @@ the tuned operator.</p>
########## 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 310.7 98.739 (1, 2, 10, 10, 3) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.018 0.959 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.949 0.302 (1, 1, 10, 10, 3) 1 1
-Total_time - 314.667 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.2 98.676 (1, 2, 10, 10, 3) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.258 1.03 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.932 0.295 (1, 1, 10, 10, 3) 1 1
+Total_time - 316.39 - - - -
</pre></div>
</div>
</div>
@@ -610,10 +610,10 @@ 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 80.95 96.807 (1, 6, 10, 10, 1) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.74 2.08 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.931 1.113 (1, 1, 10, 10, 3) 1 1
-Total_time - 83.62 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 131.3 97.967 (1, 6, 10, 10, 1) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.824 1.361 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.901 0.672 (1, 1, 10, 10, 3) 1 1
+Total_time - 134.025 - - - -
</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 cba056911..87c1481ba 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>00:48.363</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>00:44.855</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:43.939</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>00:03.794</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>00:00.211</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>00:00.210</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 with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
-<li><p><strong>00:00.210</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>00:40.645</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>00:03.636</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>00:00.198</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 with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
+<li><p><strong>00:00.191</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>00:00.186</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>
</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 85ff5a719..4e559c8eb 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>00:09.558</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:06.279</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:07.390</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>00:01.940</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>00:00.229</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>00:04.304</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>00:01.763</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>00:00.212</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 5810a89c6..2bd7cd0f6 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>00:06.021</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:05.851</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:02.221</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>00:01.214</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>00:00.765</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>00:00.758</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>00:00.323</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>00:00.254</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>00:00.252</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>00:00.235</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>00:02.074</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>00:01.350</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>00:00.723</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>00:00.308</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>00:00.237</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>00:00.230</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>00:00.229</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 c27d4238e..7ee9b8f81 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -552,7 +552,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp2ehf9h_2/input0.cc'\nsource_filename = \"/tmp/tmp2ehf9h_2/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpdsdid5jn/input0.cc'\nsource_filename = \"/tmp/tmpdsdid5jn/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/objects.inv b/docs/objects.inv
index c3d511ff7..63461f5f2 100644
Binary files a/docs/objects.inv and b/docs/objects.inv differ
diff --git a/docs/reference/api/doxygen/affine__type_8h__incl.svg b/docs/reference/api/doxygen/affine__type_8h__incl.svg
index f54e2828b..eb11c0375 100644
--- a/docs/reference/api/doxygen/affine__type_8h__incl.svg
+++ b/docs/reference/api/doxygen/affine__type_8h__incl.svg
@@ -31,18 +31,18 @@
<path fill="none" stroke="#191970" d="M1509.5955,-772.2977C1504.9743,-763.9388 1499.2031,-753.4997 1494.3111,-744.6509"/>
<polygon fill="#191970" stroke="#191970" points="1497.3452,-742.9051 1489.4438,-735.8469 1491.2191,-746.292 1497.3452,-742.9051"/>
</g>
-<!-- Node48 -->
-<g id="node49" class="node">
-<title>Node48</title>
-<g id="a_node49"><a xlink:href="ir_2type_8h.html" target="_top" xlink:title="IR/AST nodes for the unified type system in TVM. ">
+<!-- Node46 -->
+<g id="node47" class="node">
+<title>Node46</title>
+<g id="a_node47"><a xlink:href="ir_2type_8h.html" target="_top" xlink:title="IR/AST nodes for the unified type system in TVM. ">
<polygon fill="#ffffff" stroke="#000000" points="1772,-660.5 1772,-679.5 1852,-679.5 1852,-660.5 1772,-660.5"/>
<text text-anchor="middle" x="1812" y="-667.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
</a>
</g>
</g>
-<!-- Node0->Node48 -->
-<g id="edge154" class="edge">
-<title>Node0->Node48</title>
+<!-- Node0->Node46 -->
+<g id="edge152" class="edge">
+<title>Node0->Node46</title>
<path fill="none" stroke="#191970" d="M1555.5339,-772.4992C1613.3634,-749.3871 1723.3368,-705.4351 1778.5765,-683.358"/>
<polygon fill="#191970" stroke="#191970" points="1779.9246,-686.5885 1787.9115,-679.6272 1777.3267,-680.0884 1779.9246,-686.5885"/>
</g>
@@ -71,7 +71,7 @@
</g>
</g>
<!-- Node1->Node3 -->
-<g id="edge147" class="edge">
+<g id="edge145" class="edge">
<title>Node1->Node3</title>
<path fill="none" stroke="#191970" d="M1500.5925,-716.4862C1549.3294,-688.5415 1692.318,-606.5549 1751.2044,-572.7906"/>
<polygon fill="#191970" stroke="#191970" points="1753.2922,-575.6281 1760.2264,-567.6177 1749.8102,-569.5555 1753.2922,-575.6281"/>
@@ -86,78 +86,78 @@
</g>
</g>
<!-- Node1->Node8 -->
-<g id="edge149" class="edge">
+<g id="edge147" class="edge">
<title>Node1->Node8</title>
<path fill="none" stroke="#191970" d="M1481.1965,-716.2066C1475.9113,-696.8521 1465,-652.2612 1465,-614 1465,-614 1465,-614 1465,-502 1465,-462.1521 1611.0209,-220.9926 1653.5242,-151.6407"/>
<polygon fill="#191970" stroke="#191970" points="1656.7644,-153.0526 1659.0144,-142.6995 1650.7991,-149.3898 1656.7644,-153.0526"/>
</g>
-<!-- Node16 -->
-<g id="node17" class="node">
-<title>Node16</title>
+<!-- Node14 -->
+<g id="node15" class="node">
+<title>Node14</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="2637,-62 2637,-81 2681,-81 2681,-62 2637,-62"/>
<text text-anchor="middle" x="2659" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
</g>
-<!-- Node1->Node16 -->
-<g id="edge152" class="edge">
-<title>Node1->Node16</title>
+<!-- Node1->Node14 -->
+<g id="edge150" class="edge">
+<title>Node1->Node14</title>
<path fill="none" stroke="#191970" d="M1523.5587,-724.924C1749.6345,-718.5212 2868,-683.3434 2868,-614 2868,-614 2868,-614 2868,-189 2868,-109.0588 2750.3801,-82.9327 2691.6132,-74.8682"/>
<polygon fill="#191970" stroke="#191970" points="2691.7287,-71.3547 2681.3667,-73.5636 2690.8445,-78.2986 2691.7287,-71.3547"/>
</g>
-<!-- Node17 -->
-<g id="node18" class="node">
-<title>Node17</title>
+<!-- Node15 -->
+<g id="node16" class="node">
+<title>Node15</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="197.5,-62 197.5,-81 266.5,-81 266.5,-62 197.5,-62"/>
<text text-anchor="middle" x="232" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
</g>
-<!-- Node1->Node17 -->
-<g id="edge153" class="edge">
-<title>Node1->Node17</title>
+<!-- Node1->Node15 -->
+<g id="edge151" class="edge">
+<title>Node1->Node15</title>
<path fill="none" stroke="#191970" d="M1444.3612,-725.4997C1237.0884,-722.767 282.5268,-708.705 154,-680 81.3251,-663.7689 0,-688.4653 0,-614 0,-614 0,-614 0,-189 0,-106.0731 119.427,-81.6566 186.9091,-74.4794"/>
<polygon fill="#191970" stroke="#191970" points="187.6542,-77.9234 197.2673,-73.4692 186.9747,-70.9564 187.6542,-77.9234"/>
</g>
-<!-- Node22 -->
-<g id="node23" class="node">
-<title>Node22</title>
+<!-- Node20 -->
+<g id="node21" class="node">
+<title>Node20</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="928,-179.5 928,-198.5 992,-198.5 992,-179.5 928,-179.5"/>
<text text-anchor="middle" x="960" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
</g>
-<!-- Node1->Node22 -->
-<g id="edge150" class="edge">
-<title>Node1->Node22</title>
+<!-- Node1->Node20 -->
+<g id="edge148" class="edge">
+<title>Node1->Node20</title>
<path fill="none" stroke="#191970" d="M1444.4047,-725.2082C1281.3028,-721.7922 660.9493,-707.2433 467,-680 341.6729,-662.3958 190,-740.5575 190,-614 190,-614 190,-614 190,-440.5 190,-321.3905 416.1884,-305.5865 703,-235 777.7362,-216.6069 866.3966,-202.3759 917.6294,-194.8777"/>
<polygon fill="#191970" stroke="#191970" points="918.273,-198.3211 927.6685,-193.4247 917.2702,-191.3933 918.273,-198.3211"/>
</g>
-<!-- Node33 -->
-<g id="node34" class="node">
-<title>Node33</title>
-<g id="a_node34"><a xlink:href="string_8h.html" target="_top" xlink:title="Runtime String container types. ">
+<!-- Node31 -->
+<g id="node32" class="node">
+<title>Node31</title>
+<g id="a_node32"><a xlink:href="string_8h.html" target="_top" xlink:title="Runtime String container types. ">
<polygon fill="#ffffff" stroke="#000000" points="858,-291.5 858,-321.5 984,-321.5 984,-291.5 858,-291.5"/>
<text text-anchor="start" x="866" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
<text text-anchor="middle" x="921" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/string.h</text>
</a>
</g>
</g>
-<!-- Node1->Node33 -->
-<g id="edge148" class="edge">
-<title>Node1->Node33</title>
+<!-- Node1->Node31 -->
+<g id="edge146" class="edge">
+<title>Node1->Node31</title>
<path fill="none" stroke="#191970" d="M1444.1674,-724.4729C1234.2532,-716.1697 266,-674.5862 266,-614 266,-614 266,-614 266,-502 266,-441.8228 677.1922,-354.2752 847.9833,-320.5072"/>
<polygon fill="#191970" stroke="#191970" points="848.7637,-323.9208 857.8986,-318.5539 847.4106,-317.0528 848.7637,-323.9208"/>
</g>
-<!-- Node44 -->
-<g id="node45" class="node">
-<title>Node44</title>
+<!-- Node42 -->
+<g id="node43" class="node">
+<title>Node42</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="332,-364 332,-383 376,-383 376,-364 332,-364"/>
<text text-anchor="middle" x="354" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
</g>
-<!-- Node1->Node44 -->
-<g id="edge151" class="edge">
-<title>Node1->Node44</title>
+<!-- Node1->Node42 -->
+<g id="edge149" class="edge">
+<title>Node1->Node42</title>
<path fill="none" stroke="#191970" d="M1444.4802,-725.8459C1231.5475,-724.8614 228.5529,-718.2037 176,-680 150.7535,-661.6469 152,-645.2125 152,-614 152,-614 152,-614 152,-558 152,-466.8203 265.5938,-408.6386 322.4141,-385.2392"/>
<polygon fill="#191970" stroke="#191970" points="324.016,-388.368 331.9979,-381.401 321.4135,-381.8698 324.016,-388.368"/>
</g>
-<!-- Node1->Node48 -->
-<g id="edge140" class="edge">
-<title>Node1->Node48</title>
+<!-- Node1->Node46 -->
+<g id="edge138" class="edge">
+<title>Node1->Node46</title>
<path fill="none" stroke="#191970" d="M1523.5645,-719.2451C1583.2444,-709.0558 1696.2574,-689.7609 1761.7024,-678.5874"/>
<polygon fill="#191970" stroke="#191970" points="1762.5137,-681.9996 1771.782,-676.8665 1761.3356,-675.0994 1762.5137,-681.9996"/>
</g>
@@ -168,14 +168,14 @@
<polygon fill="#191970" stroke="#191970" points="1798.8514,-571.2268 1788.8812,-567.6427 1794.4402,-576.662 1798.8514,-571.2268"/>
</g>
<!-- Node2->Node8 -->
-<g id="edge138" class="edge">
+<g id="edge136" class="edge">
<title>Node2->Node8</title>
<path fill="none" stroke="#191970" d="M1886.8085,-611.3816C1984.897,-603.3294 2238.3248,-571.9179 2408,-456 2440.2817,-433.9459 2451.5036,-425.6925 2465,-389 2469.7562,-376.0692 2466.7074,-371.6716 2465,-358 2461.1929,-327.5154 2465.0656,-316.5454 2448,-291 2424.8913,-256.4087 2410.215,-253.5905 2373,-235 2293.8625,-195.4673 2268.9206,-195.4437 2182,-179 2024.0115,-149.1116 1835.0905,-138.5404 1735.1128,-134.8783"/>
<polygon fill="#191970" stroke="#191970" points="1734.9553,-131.3706 1724.8378,-134.5149 1734.7079,-138.3662 1734.9553,-131.3706"/>
</g>
-<!-- Node2->Node16 -->
-<g id="edge139" class="edge">
-<title>Node2->Node16</title>
+<!-- Node2->Node14 -->
+<g id="edge137" class="edge">
+<title>Node2->Node14</title>
<path fill="none" stroke="#191970" d="M1886.6559,-609.7322C1951.4926,-602.6708 2082.1987,-587.3905 2192,-568 2478.4604,-517.4121 2792,-597.393 2792,-306.5 2792,-306.5 2792,-306.5 2792,-189 2792,-135.1753 2730.0653,-100.0942 2690.7248,-83.2401"/>
<polygon fill="#191970" stroke="#191970" points="2691.8409,-79.9147 2681.2612,-79.3475 2689.178,-86.3884 2691.8409,-79.9147"/>
</g>
@@ -205,13 +205,13 @@
</g>
</g>
<!-- Node3->Node5 -->
-<g id="edge129" class="edge">
+<g id="edge127" class="edge">
<title>Node3->Node5</title>
<path fill="none" stroke="#191970" d="M1777.2094,-548.3402C1777.81,-520.6387 1779.5525,-440.2663 1780.4506,-398.8429"/>
<polygon fill="#191970" stroke="#191970" points="1783.952,-398.8099 1780.6697,-388.7363 1776.9536,-398.6581 1783.952,-398.8099"/>
</g>
<!-- Node3->Node8 -->
-<g id="edge133" class="edge">
+<g id="edge131" class="edge">
<title>Node3->Node8</title>
<path fill="none" stroke="#191970" d="M1826.6593,-555.699C2004.7907,-545.7045 2592.3888,-496.7671 2425,-291 2338.1162,-184.1958 1905.7739,-147.2304 1734.7758,-136.6589"/>
<polygon fill="#191970" stroke="#191970" points="1734.8219,-133.1554 1724.6284,-136.0437 1734.3983,-140.1426 1734.8219,-133.1554"/>
@@ -227,92 +227,92 @@
</g>
</g>
<!-- Node3->Node9 -->
-<g id="edge131" class="edge">
+<g id="edge129" class="edge">
<title>Node3->Node9</title>
<path fill="none" stroke="#191970" d="M1826.5025,-557.0696C1960.4033,-553.4966 2325.2316,-536.4285 2408,-456 2499.3881,-367.1955 2525.7948,-298.6664 2482,-179 2471.1217,-149.2756 2463.601,-141.6136 2438,-123 2418.0221,-108.4748 2393.4982,-97.6244 2370.5883,-89.7267"/>
<polygon fill="#191970" stroke="#191970" points="2371.4799,-86.3352 2360.8866,-86.5207 2369.2834,-92.9817 2371.4799,-86.3352"/>
</g>
-<!-- Node3->Node16 -->
-<g id="edge134" class="edge">
-<title>Node3->Node16</title>
+<!-- Node3->Node14 -->
+<g id="edge132" class="edge">
+<title>Node3->Node14</title>
<path fill="none" stroke="#191970" d="M1826.8217,-556.0937C1943.2168,-551.2712 2230.1473,-537.147 2323,-512 2527.8361,-456.5249 2754,-518.7152 2754,-306.5 2754,-306.5 2754,-306.5 2754,-189 2754,-144.262 2711.839,-106.9849 2683.6909,-87.0828"/>
<polygon fill="#191970" stroke="#191970" points="2685.3512,-83.979 2675.1155,-81.2441 2681.4116,-89.7652 2685.3512,-83.979"/>
</g>
-<!-- Node3->Node17 -->
-<g id="edge135" class="edge">
-<title>Node3->Node17</title>
+<!-- Node3->Node15 -->
+<g id="edge133" class="edge">
+<title>Node3->Node15</title>
<path fill="none" stroke="#191970" d="M1727.2135,-557.3275C1485.8848,-553.9495 443.5874,-537.9617 301,-512 225.2247,-498.2032 192.9133,-508.9712 137,-456 82.4417,-404.3125 79.9024,-363.4174 100,-291 123.1306,-207.6537 187.3192,-124.2395 216.8176,-88.9795"/>
<polygon fill="#191970" stroke="#191970" points="219.719,-90.97 223.5237,-81.0819 214.3831,-86.4391 219.719,-90.97"/>
</g>
-<!-- Node18 -->
-<g id="node19" class="node">
-<title>Node18</title>
+<!-- Node16 -->
+<g id="node17" class="node">
+<title>Node16</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="604.5,-62 604.5,-81 649.5,-81 649.5,-62 604.5,-62"/>
<text text-anchor="middle" x="627" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
</g>
-<!-- Node3->Node18 -->
-<g id="edge136" class="edge">
-<title>Node3->Node18</title>
+<!-- Node3->Node16 -->
+<g id="edge134" class="edge">
+<title>Node3->Node16</title>
<path fill="none" stroke="#191970" d="M1727.283,-557.2315C1522.4816,-553.9083 743.4629,-539.6754 499,-512 367.6959,-497.1352 325.5497,-514.3755 207,-456 157.4308,-431.5914 114,-428.753 114,-373.5 114,-373.5 114,-373.5 114,-306.5 114,-226.8147 312.8371,-152.2844 391,-123 460.951,-96.7923 547.8148,-82.1221 594.1895,-75.622"/>
<polygon fill="#191970" stroke="#191970" points="594.8263,-79.0677 604.2625,-74.2502 593.8816,-72.1317 594.8263,-79.0677"/>
</g>
-<!-- Node20 -->
-<g id="node21" class="node">
-<title>Node20</title>
+<!-- Node18 -->
+<g id="node19" class="node">
+<title>Node18</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1456.5,-235.5 1456.5,-254.5 1503.5,-254.5 1503.5,-235.5 1456.5,-235.5"/>
<text text-anchor="middle" x="1480" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
</g>
-<!-- Node3->Node20 -->
-<g id="edge137" class="edge">
-<title>Node3->Node20</title>
+<!-- Node3->Node18 -->
+<g id="edge135" class="edge">
+<title>Node3->Node18</title>
<path fill="none" stroke="#191970" d="M1727.2485,-553.115C1664.8536,-544.3812 1558.4986,-520.4624 1496,-456 1445.2951,-403.7019 1463.5373,-305.8371 1474.3407,-264.3657"/>
<polygon fill="#191970" stroke="#191970" points="1477.7297,-265.2415 1477.0042,-254.6716 1470.9798,-263.3869 1477.7297,-265.2415"/>
</g>
-<!-- Node25 -->
-<g id="node26" class="node">
-<title>Node25</title>
-<g id="a_node26"><a xlink:href="runtime_2memory_8h.html" target="_top" xlink:title="Runtime memory management. ">
+<!-- Node23 -->
+<g id="node24" class="node">
+<title>Node23</title>
+<g id="a_node24"><a xlink:href="runtime_2memory_8h.html" target="_top" xlink:title="Runtime memory management. ">
<polygon fill="#ffffff" stroke="#000000" points="374.5,-179.5 374.5,-198.5 503.5,-198.5 503.5,-179.5 374.5,-179.5"/>
<text text-anchor="middle" x="439" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
</a>
</g>
</g>
-<!-- Node3->Node25 -->
-<g id="edge132" class="edge">
-<title>Node3->Node25</title>
+<!-- Node3->Node23 -->
+<g id="edge130" class="edge">
+<title>Node3->Node23</title>
<path fill="none" stroke="#191970" d="M1727.4797,-557.7078C1455.1431,-555.4644 152,-536.2277 152,-373.5 152,-373.5 152,-373.5 152,-306.5 152,-257.1341 192.8013,-256.9878 237,-235 276.6959,-215.2522 324.9993,-203.8567 364.4167,-197.3486"/>
<polygon fill="#191970" stroke="#191970" points="365.1648,-200.774 374.4969,-195.7579 364.0736,-193.8596 365.1648,-200.774"/>
</g>
-<!-- Node29 -->
-<g id="node30" class="node">
-<title>Node29</title>
-<g id="a_node30"><a xlink:href="structural__hash_8h.html" target="_top" xlink:title="tvm/node/structural\l_hash.h">
+<!-- Node27 -->
+<g id="node28" class="node">
+<title>Node27</title>
+<g id="a_node28"><a xlink:href="structural__hash_8h.html" target="_top" xlink:title="tvm/node/structural\l_hash.h">
<polygon fill="#ffffff" stroke="#000000" points="2114.5,-425.5 2114.5,-455.5 2227.5,-455.5 2227.5,-425.5 2114.5,-425.5"/>
<text text-anchor="start" x="2122.5" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
<text text-anchor="middle" x="2171" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_hash.h</text>
</a>
</g>
</g>
-<!-- Node3->Node29 -->
-<g id="edge130" class="edge">
-<title>Node3->Node29</title>
+<!-- Node3->Node27 -->
+<g id="edge128" class="edge">
+<title>Node3->Node27</title>
<path fill="none" stroke="#191970" d="M1810.5398,-548.4414C1842.7456,-539.2287 1892.7486,-524.8356 1936,-512 1996.1146,-494.16 2064.7012,-473.2252 2112.1951,-458.6364"/>
<polygon fill="#191970" stroke="#191970" points="2113.4686,-461.9066 2121.999,-455.623 2111.412,-455.2155 2113.4686,-461.9066"/>
</g>
-<!-- Node46 -->
-<g id="node47" class="node">
-<title>Node46</title>
-<g id="a_node47"><a xlink:href="repr__printer_8h.html" target="_top" xlink:title="Printer class to print repr string of each AST/IR nodes. ">
-<polygon fill="#ffffff" stroke="#000000" points="763.5,-492.5 763.5,-511.5 894.5,-511.5 894.5,-492.5 763.5,-492.5"/>
-<text text-anchor="middle" x="829" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/repr_printer.h</text>
+<!-- Node44 -->
+<g id="node45" class="node">
+<title>Node44</title>
+<g id="a_node45"><a xlink:href="repr__printer_8h.html" target="_top" xlink:title="Printer class to print repr string of each AST/IR nodes. ">
+<polygon fill="#ffffff" stroke="#000000" points="761.5,-492.5 761.5,-511.5 892.5,-511.5 892.5,-492.5 761.5,-492.5"/>
+<text text-anchor="middle" x="827" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/repr_printer.h</text>
</a>
</g>
</g>
-<!-- Node3->Node46 -->
-<g id="edge126" class="edge">
-<title>Node3->Node46</title>
-<path fill="none" stroke="#191970" d="M1727.3662,-555.068C1571.5577,-545.8642 1091.4606,-517.504 904.5589,-506.4634"/>
-<polygon fill="#191970" stroke="#191970" points="904.7332,-502.9677 894.5442,-505.8718 904.3203,-509.9555 904.7332,-502.9677"/>
+<!-- Node3->Node44 -->
+<g id="edge124" class="edge">
+<title>Node3->Node44</title>
+<path fill="none" stroke="#191970" d="M1727.2614,-555.068C1571.1242,-545.8642 1090.0143,-517.504 902.7183,-506.4634"/>
+<polygon fill="#191970" stroke="#191970" points="902.8711,-502.9664 892.6824,-505.8718 902.4591,-509.9543 902.8711,-502.9664"/>
</g>
<!-- Node4->Node5 -->
<g id="edge5" class="edge">
@@ -321,90 +321,90 @@
<polygon fill="#191970" stroke="#191970" points="1799.4917,-395.1111 1791.0555,-388.7016 1793.6534,-398.973 1799.4917,-395.1111"/>
</g>
<!-- Node4->Node8 -->
-<g id="edge91" class="edge">
+<g id="edge89" class="edge">
<title>Node4->Node8</title>
<path fill="none" stroke="#191970" d="M1866.3373,-492.4992C1866.8456,-468.8374 1866.1517,-405.744 1847,-358 1837.8243,-335.1256 1732.5153,-197.2942 1716,-179 1706.387,-168.3516 1694.5452,-157.5519 1684.6927,-149.0951"/>
<polygon fill="#191970" stroke="#191970" points="1686.8806,-146.3618 1676.9772,-142.5972 1682.3714,-151.716 1686.8806,-146.3618"/>
</g>
<!-- Node4->Node9 -->
-<g id="edge87" class="edge">
+<g id="edge85" class="edge">
<title>Node4->Node9</title>
<path fill="none" stroke="#191970" d="M1926.6124,-497.2536C2016.5282,-489.7438 2180.6956,-474.1022 2237,-456 2334.6835,-424.5942 2372.3647,-410.0791 2425,-322 2458.7931,-265.4511 2495.4707,-236.8735 2464,-179 2440.4869,-135.7602 2391.2554,-107.2381 2352.4813,-90.5092"/>
<polygon fill="#191970" stroke="#191970" points="2353.448,-87.1205 2342.8708,-86.5084 2350.7577,-93.5829 2353.448,-87.1205"/>
</g>
-<!-- Node4->Node16 -->
-<g id="edge123" class="edge">
-<title>Node4->Node16</title>
+<!-- Node4->Node14 -->
+<g id="edge121" class="edge">
+<title>Node4->Node14</title>
<path fill="none" stroke="#191970" d="M1926.6756,-498.5576C2027.8425,-492.3021 2226.4483,-477.6803 2293,-456 2463.1938,-400.5567 2518.2905,-384.3671 2642,-255 2684.1928,-210.8776 2691.745,-182.9088 2680,-123 2677.79,-111.7273 2673.0988,-99.8097 2668.7098,-90.2811"/>
<polygon fill="#191970" stroke="#191970" points="2671.7249,-88.4786 2664.1976,-81.0227 2665.4325,-91.5454 2671.7249,-88.4786"/>
</g>
-<!-- Node4->Node17 -->
-<g id="edge124" class="edge">
-<title>Node4->Node17</title>
-<path fill="none" stroke="#191970" d="M1805.271,-496.9707C1649.5731,-484.1512 1223.4008,-449.5331 868,-425 802.3678,-420.4694 334.0892,-421.6945 277,-389 174.4422,-330.266 211.0316,-150.6131 226.5173,-90.9729"/>
-<polygon fill="#191970" stroke="#191970" points="229.9042,-91.8557 229.125,-81.2896 223.145,-90.0355 229.9042,-91.8557"/>
+<!-- Node4->Node15 -->
+<g id="edge122" class="edge">
+<title>Node4->Node15</title>
+<path fill="none" stroke="#191970" d="M1805.4338,-497.001C1649.6583,-484.2166 1222.3254,-449.6064 866,-425 800.5892,-420.483 333.8934,-421.5902 277,-389 174.4483,-330.2554 211.0336,-150.6095 226.5179,-90.972"/>
+<polygon fill="#191970" stroke="#191970" points="229.9046,-91.8553 229.1252,-81.2891 223.1454,-90.0351 229.9046,-91.8553"/>
</g>
-<!-- Node4->Node20 -->
-<g id="edge125" class="edge">
-<title>Node4->Node20</title>
+<!-- Node4->Node18 -->
+<g id="edge123" class="edge">
+<title>Node4->Node18</title>
<path fill="none" stroke="#191970" d="M1805.2293,-497.69C1711.9988,-490.5162 1542.2664,-475.0177 1522,-456 1468.1098,-405.4302 1472.4328,-306.9879 1477.1761,-264.8814"/>
<polygon fill="#191970" stroke="#191970" points="1480.6686,-265.1601 1478.4468,-254.8009 1473.7236,-264.2846 1480.6686,-265.1601"/>
</g>
-<!-- Node4->Node25 -->
-<g id="edge89" class="edge">
-<title>Node4->Node25</title>
+<!-- Node4->Node23 -->
+<g id="edge87" class="edge">
+<title>Node4->Node23</title>
<path fill="none" stroke="#191970" d="M1805.4482,-498.1676C1731.647,-492.6521 1603.4819,-480.3422 1496,-456 1453.507,-446.3763 1445.8843,-432.6945 1403,-425 1284.8206,-403.7957 424.8266,-452.6187 323,-389 259.0813,-349.0652 205.5052,-294.318 252,-235 266.123,-216.9819 318.6525,-205.0772 364.4544,-197.9017"/>
<polygon fill="#191970" stroke="#191970" points="365.045,-201.3522 374.4101,-196.398 363.9995,-194.4308 365.045,-201.3522"/>
</g>
-<!-- Node28 -->
-<g id="node29" class="node">
-<title>Node28</title>
-<g id="a_node29"><a xlink:href="data__type_8h.html" target="_top" xlink:title="tvm/runtime/data_type.h">
+<!-- Node26 -->
+<g id="node27" class="node">
+<title>Node26</title>
+<g id="a_node27"><a xlink:href="data__type_8h.html" target="_top" xlink:title="tvm/runtime/data_type.h">
<polygon fill="#ffffff" stroke="#000000" points="2156,-297 2156,-316 2294,-316 2294,-297 2156,-297"/>
<text text-anchor="middle" x="2225" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/data_type.h</text>
</a>
</g>
</g>
-<!-- Node4->Node28 -->
-<g id="edge88" class="edge">
-<title>Node4->Node28</title>
+<!-- Node4->Node26 -->
+<g id="edge86" class="edge">
+<title>Node4->Node26</title>
<path fill="none" stroke="#191970" d="M1883.8029,-492.3051C1942.3126,-460.4426 2128.9011,-358.8324 2198.495,-320.9338"/>
<polygon fill="#191970" stroke="#191970" points="2200.4047,-323.8792 2207.513,-316.0229 2197.0569,-317.7316 2200.4047,-323.8792"/>
</g>
-<!-- Node4->Node29 -->
-<g id="edge46" class="edge">
-<title>Node4->Node29</title>
+<!-- Node4->Node27 -->
+<g id="edge44" class="edge">
+<title>Node4->Node27</title>
<path fill="none" stroke="#191970" d="M1916.2719,-492.4313C1963.6993,-483.3298 2036.7582,-469.1177 2100,-456 2101.467,-455.6957 2102.9507,-455.3863 2104.4463,-455.0729"/>
<polygon fill="#191970" stroke="#191970" points="2105.3823,-458.4524 2114.4414,-452.9585 2103.9335,-451.604 2105.3823,-458.4524"/>
</g>
-<!-- Node30 -->
-<g id="node31" class="node">
-<title>Node30</title>
-<g id="a_node31"><a xlink:href="ndarray_8h.html" target="_top" xlink:title="A device-independent managed NDArray abstraction. ">
+<!-- Node28 -->
+<g id="node29" class="node">
+<title>Node28</title>
+<g id="a_node29"><a xlink:href="ndarray_8h.html" target="_top" xlink:title="A device-independent managed NDArray abstraction. ">
<polygon fill="#ffffff" stroke="#000000" points="1894.5,-364 1894.5,-383 2019.5,-383 2019.5,-364 1894.5,-364"/>
<text text-anchor="middle" x="1957" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/ndarray.h</text>
</a>
</g>
</g>
-<!-- Node4->Node30 -->
-<g id="edge90" class="edge">
-<title>Node4->Node30</title>
+<!-- Node4->Node28 -->
+<g id="edge88" class="edge">
+<title>Node4->Node28</title>
<path fill="none" stroke="#191970" d="M1872.8398,-492.3416C1887.9913,-470.9463 1924.5859,-419.2715 1944.1533,-391.6407"/>
<polygon fill="#191970" stroke="#191970" points="1947.1578,-393.4541 1950.0808,-383.2705 1941.4452,-389.4086 1947.1578,-393.4541"/>
</g>
-<!-- Node41 -->
-<g id="node42" class="node">
-<title>Node41</title>
-<g id="a_node42"><a xlink:href="packed__func_8h.html" target="_top" xlink:title="Type-erased function used across TVM API. ">
+<!-- Node39 -->
+<g id="node40" class="node">
+<title>Node39</title>
+<g id="a_node40"><a xlink:href="packed__func_8h.html" target="_top" xlink:title="Type-erased function used across TVM API. ">
<polygon fill="#ffffff" stroke="#000000" points="1531,-425.5 1531,-455.5 1647,-455.5 1647,-425.5 1531,-425.5"/>
<text text-anchor="start" x="1539" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/packed</text>
<text text-anchor="middle" x="1589" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_func.h</text>
</a>
</g>
</g>
-<!-- Node4->Node41 -->
-<g id="edge92" class="edge">
-<title>Node4->Node41</title>
+<!-- Node4->Node39 -->
+<g id="edge90" class="edge">
+<title>Node4->Node39</title>
<path fill="none" stroke="#191970" d="M1823.0227,-492.4581C1778.6242,-482.6007 1708.6912,-467.074 1656.8237,-455.5583"/>
<polygon fill="#191970" stroke="#191970" points="1657.5529,-452.1351 1647.032,-453.3844 1656.0357,-458.9687 1657.5529,-452.1351"/>
</g>
@@ -423,31 +423,31 @@
<path fill="none" stroke="#191970" d="M1724.465,-363.8954C1709.1076,-361.6089 1692.4514,-359.4223 1677,-358 1229.7169,-316.8287 1111.9885,-383.1756 667,-322 660.5337,-321.111 653.7683,-319.8476 647.159,-318.421"/>
<polygon fill="#191970" stroke="#191970" points="647.4969,-314.9074 636.9666,-316.0752 645.9268,-321.729 647.4969,-314.9074"/>
</g>
-<!-- Node5->Node16 -->
-<g id="edge45" class="edge">
-<title>Node5->Node16</title>
+<!-- Node5->Node14 -->
+<g id="edge43" class="edge">
+<title>Node5->Node14</title>
<path fill="none" stroke="#191970" d="M1837.7643,-364.8881C1889.3135,-356.3653 1967.1259,-341.7378 2033,-322 2063.8456,-312.7577 2504.2438,-134.2978 2627.2795,-84.3766"/>
<polygon fill="#191970" stroke="#191970" points="2628.9373,-87.4811 2636.8874,-80.4777 2626.3052,-80.9948 2628.9373,-87.4811"/>
</g>
-<!-- Node21 -->
-<g id="node22" class="node">
-<title>Node21</title>
-<g id="a_node22"><a xlink:href="array_8h.html" target="_top" xlink:title="Runtime Array container types. ">
+<!-- Node19 -->
+<g id="node20" class="node">
+<title>Node19</title>
+<g id="a_node20"><a xlink:href="array_8h.html" target="_top" xlink:title="Runtime Array container types. ">
<polygon fill="#ffffff" stroke="#000000" points="1116,-291.5 1116,-321.5 1242,-321.5 1242,-291.5 1116,-291.5"/>
<text text-anchor="start" x="1124" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
<text text-anchor="middle" x="1179" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/array.h</text>
</a>
</g>
</g>
-<!-- Node5->Node21 -->
-<g id="edge23" class="edge">
-<title>Node5->Node21</title>
+<!-- Node5->Node19 -->
+<g id="edge21" class="edge">
+<title>Node5->Node19</title>
<path fill="none" stroke="#191970" d="M1724.4126,-364.3982C1709.0552,-362.1107 1692.4122,-359.7979 1677,-358 1491.4969,-336.3605 1440.6534,-349.5811 1252.0942,-322.042"/>
<polygon fill="#191970" stroke="#191970" points="1252.5371,-318.5695 1242.1325,-320.5682 1251.5125,-325.4941 1252.5371,-318.5695"/>
</g>
-<!-- Node5->Node28 -->
-<g id="edge40" class="edge">
-<title>Node5->Node28</title>
+<!-- Node5->Node26 -->
+<g id="edge38" class="edge">
+<title>Node5->Node26</title>
<path fill="none" stroke="#191970" d="M1837.6839,-364.9463C1917.7096,-352.8704 2063.8544,-330.817 2151.9375,-317.5252"/>
<polygon fill="#191970" stroke="#191970" points="2152.49,-320.9815 2161.8558,-316.0285 2151.4454,-314.0599 2152.49,-320.9815"/>
</g>
@@ -469,21 +469,21 @@
<path fill="none" stroke="#191970" d="M637.6559,-296.9268C647.1069,-294.6402 657.3907,-292.4457 667,-291 926.3476,-251.9816 1005.1689,-331.5986 1256,-255 1272.7672,-249.8796 1273.8914,-241.9189 1290,-235 1372.8507,-199.4144 1397.9762,-202.6099 1485,-179 1529.388,-166.9573 1580.1734,-154.0586 1616.4577,-144.9937"/>
<polygon fill="#191970" stroke="#191970" points="1617.4334,-148.3577 1626.2898,-142.5425 1615.7401,-141.5655 1617.4334,-148.3577"/>
</g>
-<!-- Node6->Node17 -->
-<g id="edge20" class="edge">
-<title>Node6->Node17</title>
+<!-- Node6->Node15 -->
+<g id="edge18" class="edge">
+<title>Node6->Node15</title>
<path fill="none" stroke="#191970" d="M548.4264,-301.1166C499.6283,-294.7844 427.0657,-281.5598 369,-255 331.1696,-237.696 320.5127,-231.0569 294,-199 266.1496,-165.3256 247.1089,-117.0362 238.0536,-90.5991"/>
<polygon fill="#191970" stroke="#191970" points="241.3457,-89.4066 234.8824,-81.0116 234.6998,-91.6048 241.3457,-89.4066"/>
</g>
-<!-- Node6->Node18 -->
-<g id="edge21" class="edge">
-<title>Node6->Node18</title>
+<!-- Node6->Node16 -->
+<g id="edge19" class="edge">
+<title>Node6->Node16</title>
<path fill="none" stroke="#191970" d="M548.3759,-300.4266C467.2453,-288.742 328.1896,-257.0005 365,-179 381.7093,-143.5932 398.1686,-140.8777 433,-123 485.9335,-95.8312 554.5563,-81.9943 594.3933,-75.7914"/>
<polygon fill="#191970" stroke="#191970" points="594.9768,-79.2431 604.3516,-74.3072 593.9449,-72.3196 594.9768,-79.2431"/>
</g>
-<!-- Node6->Node20 -->
-<g id="edge22" class="edge">
-<title>Node6->Node20</title>
+<!-- Node6->Node18 -->
+<g id="edge20" class="edge">
+<title>Node6->Node18</title>
<path fill="none" stroke="#191970" d="M637.3183,-296.9784C646.8604,-294.6627 657.2743,-292.4407 667,-291 686.589,-288.0982 1295.005,-255.0205 1446.2297,-246.8273"/>
<polygon fill="#191970" stroke="#191970" points="1446.5173,-250.317 1456.3134,-246.2812 1446.1387,-243.3272 1446.5173,-250.317"/>
</g>
@@ -493,816 +493,792 @@
<path fill="none" stroke="#191970" d="M1724.5511,-127.2142C1840.5164,-115.9475 2096.3726,-91.0894 2223.4141,-78.7465"/>
<polygon fill="#191970" stroke="#191970" points="2223.832,-82.2225 2233.4467,-77.7718 2223.1551,-75.2553 2223.832,-82.2225"/>
</g>
-<!-- Node15 -->
-<g id="node16" class="node">
-<title>Node15</title>
+<!-- Node13 -->
+<g id="node14" class="node">
+<title>Node13</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1233.5,-62 1233.5,-81 1358.5,-81 1358.5,-62 1233.5,-62"/>
<text text-anchor="middle" x="1296" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/logging.h</text>
</g>
-<!-- Node8->Node15 -->
-<g id="edge15" class="edge">
-<title>Node8->Node15</title>
+<!-- Node8->Node13 -->
+<g id="edge13" class="edge">
+<title>Node8->Node13</title>
<path fill="none" stroke="#191970" d="M1607.7486,-123.4581C1541.6656,-112.4443 1433.1184,-94.3531 1363.1203,-82.6867"/>
<polygon fill="#191970" stroke="#191970" points="1363.5214,-79.2054 1353.0821,-81.0137 1362.3706,-86.1101 1363.5214,-79.2054"/>
</g>
-<!-- Node8->Node16 -->
-<g id="edge16" class="edge">
-<title>Node8->Node16</title>
+<!-- Node8->Node14 -->
+<g id="edge14" class="edge">
+<title>Node8->Node14</title>
<path fill="none" stroke="#191970" d="M1724.5711,-129.3143C1911.7345,-117.7342 2483.9773,-82.3289 2626.5601,-73.5071"/>
<polygon fill="#191970" stroke="#191970" points="2627.0499,-76.9836 2636.8146,-72.8726 2626.6175,-69.9969 2627.0499,-76.9836"/>
</g>
-<!-- Node8->Node17 -->
-<g id="edge17" class="edge">
-<title>Node8->Node17</title>
+<!-- Node8->Node15 -->
+<g id="edge15" class="edge">
+<title>Node8->Node15</title>
<path fill="none" stroke="#191970" d="M1605.2502,-130.4357C1366.7939,-120.2019 486.0419,-82.4027 276.8066,-73.423"/>
<polygon fill="#191970" stroke="#191970" points="276.6495,-69.9131 266.5086,-72.981 276.3493,-76.9067 276.6495,-69.9131"/>
</g>
-<!-- Node8->Node18 -->
-<g id="edge18" class="edge">
-<title>Node8->Node18</title>
+<!-- Node8->Node16 -->
+<g id="edge16" class="edge">
+<title>Node8->Node16</title>
<path fill="none" stroke="#191970" d="M1605.4644,-129.4726C1412.3517,-118.031 807.3194,-82.1837 659.7181,-73.4385"/>
<polygon fill="#191970" stroke="#191970" points="659.6995,-69.9313 649.5099,-72.8337 659.2854,-76.9191 659.6995,-69.9313"/>
</g>
-<!-- Node19 -->
-<g id="node20" class="node">
-<title>Node19</title>
+<!-- Node17 -->
+<g id="node18" class="node">
+<title>Node17</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1814,-62 1814,-81 1864,-81 1864,-62 1814,-62"/>
<text text-anchor="middle" x="1839" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">atomic</text>
</g>
-<!-- Node8->Node19 -->
-<g id="edge19" class="edge">
-<title>Node8->Node19</title>
+<!-- Node8->Node17 -->
+<g id="edge17" class="edge">
+<title>Node8->Node17</title>
<path fill="none" stroke="#191970" d="M1692.1875,-123.3906C1722.418,-112.7057 1771.323,-95.4203 1804.2947,-83.7665"/>
<polygon fill="#191970" stroke="#191970" points="1805.7154,-86.9766 1813.9775,-80.3442 1803.3827,-80.3768 1805.7154,-86.9766"/>
</g>
<!-- Node10 -->
<g id="node11" class="node">
<title>Node10</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2085.5,-.5 2085.5,-19.5 2178.5,-19.5 2178.5,-.5 2085.5,-.5"/>
-<text text-anchor="middle" x="2132" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dlpack/dlpack.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2158.5,-.5 2158.5,-19.5 2251.5,-19.5 2251.5,-.5 2158.5,-.5"/>
+<text text-anchor="middle" x="2205" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dlpack/dlpack.h</text>
</g>
<!-- Node9->Node10 -->
<g id="edge10" class="edge">
<title>Node9->Node10</title>
-<path fill="none" stroke="#191970" d="M2257.3918,-56.4554C2229.8158,-46.239 2193.7058,-32.8609 2167.3218,-23.0861"/>
-<polygon fill="#191970" stroke="#191970" points="2168.3172,-19.7224 2157.7241,-19.5303 2165.8853,-26.2864 2168.3172,-19.7224"/>
+<path fill="none" stroke="#191970" d="M2275.0112,-56.2977C2260.8034,-46.9022 2242.6215,-34.8787 2228.3475,-25.4395"/>
+<polygon fill="#191970" stroke="#191970" points="2229.8845,-22.2598 2219.6128,-19.6633 2226.0233,-28.0986 2229.8845,-22.2598"/>
</g>
<!-- Node11 -->
<g id="node12" class="node">
<title>Node11</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2197,-.5 2197,-19.5 2253,-19.5 2253,-.5 2197,-.5"/>
-<text text-anchor="middle" x="2225" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stddef.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2270,-.5 2270,-19.5 2326,-19.5 2326,-.5 2270,-.5"/>
+<text text-anchor="middle" x="2298" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stddef.h</text>
</g>
<!-- Node9->Node11 -->
<g id="edge11" class="edge">
<title>Node9->Node11</title>
-<path fill="none" stroke="#191970" d="M2279.9551,-56.2977C2269.1886,-47.2274 2255.5149,-35.7077 2244.5005,-26.4285"/>
-<polygon fill="#191970" stroke="#191970" points="2246.5911,-23.6132 2236.6882,-19.8469 2242.0809,-28.9667 2246.5911,-23.6132"/>
+<path fill="none" stroke="#191970" d="M2298,-56.2977C2298,-48.3834 2298,-38.6043 2298,-30.0759"/>
+<polygon fill="#191970" stroke="#191970" points="2301.5001,-29.8469 2298,-19.8469 2294.5001,-29.847 2301.5001,-29.8469"/>
</g>
<!-- Node12 -->
<g id="node13" class="node">
<title>Node12</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2271.5,-.5 2271.5,-19.5 2324.5,-19.5 2324.5,-.5 2271.5,-.5"/>
-<text text-anchor="middle" x="2298" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stdint.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2344.5,-.5 2344.5,-19.5 2397.5,-19.5 2397.5,-.5 2344.5,-.5"/>
+<text text-anchor="middle" x="2371" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stdint.h</text>
</g>
<!-- Node9->Node12 -->
<g id="edge12" class="edge">
<title>Node9->Node12</title>
-<path fill="none" stroke="#191970" d="M2298,-56.2977C2298,-48.3834 2298,-38.6043 2298,-30.0759"/>
-<polygon fill="#191970" stroke="#191970" points="2301.5001,-29.8469 2298,-19.8469 2294.5001,-29.847 2301.5001,-29.8469"/>
-</g>
-<!-- Node13 -->
-<g id="node14" class="node">
-<title>Node13</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2343,-.5 2343,-19.5 2393,-19.5 2393,-.5 2343,-.5"/>
-<text text-anchor="middle" x="2368" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stdio.h</text>
-</g>
-<!-- Node9->Node13 -->
-<g id="edge13" class="edge">
-<title>Node9->Node13</title>
-<path fill="none" stroke="#191970" d="M2315.3034,-56.2977C2325.5262,-47.3163 2338.4823,-35.9334 2348.9896,-26.702"/>
-<polygon fill="#191970" stroke="#191970" points="2351.5897,-29.0766 2356.7921,-19.8469 2346.9695,-23.8178 2351.5897,-29.0766"/>
-</g>
-<!-- Node14 -->
-<g id="node15" class="node">
-<title>Node14</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2411,-.5 2411,-19.5 2483,-19.5 2483,-.5 2411,-.5"/>
-<text text-anchor="middle" x="2447" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">sys/types.h</text>
-</g>
-<!-- Node9->Node14 -->
-<g id="edge14" class="edge">
-<title>Node9->Node14</title>
-<path fill="none" stroke="#191970" d="M2334.4495,-56.4554C2358.8744,-46.374 2390.7581,-33.2139 2414.3521,-23.4755"/>
-<polygon fill="#191970" stroke="#191970" points="2416.0021,-26.5809 2423.9103,-19.5303 2413.3313,-20.1104 2416.0021,-26.5809"/>
+<path fill="none" stroke="#191970" d="M2316.0449,-56.2977C2326.8114,-47.2274 2340.4851,-35.7077 2351.4995,-26.4285"/>
+<polygon fill="#191970" stroke="#191970" points="2353.9191,-28.9667 2359.3118,-19.8469 2349.4089,-23.6132 2353.9191,-28.9667"/>
</g>
-<!-- Node21->Node18 -->
-<g id="edge26" class="edge">
-<title>Node21->Node18</title>
+<!-- Node19->Node16 -->
+<g id="edge24" class="edge">
+<title>Node19->Node16</title>
<path fill="none" stroke="#191970" d="M1226.99,-291.3757C1257.8536,-278.9531 1289.3722,-259.4473 1273,-235 1202.6228,-129.9112 781.2106,-85.0637 659.7448,-74.2144"/>
<polygon fill="#191970" stroke="#191970" points="659.8518,-70.7105 649.5844,-73.3243 659.2408,-77.6838 659.8518,-70.7105"/>
</g>
-<!-- Node21->Node20 -->
-<g id="edge27" class="edge">
-<title>Node21->Node20</title>
+<!-- Node19->Node18 -->
+<g id="edge25" class="edge">
+<title>Node19->Node18</title>
<path fill="none" stroke="#191970" d="M1242.2056,-292.8416C1245.1778,-292.2156 1248.1208,-291.5991 1251,-291 1320.2819,-276.5848 1401.5209,-260.4471 1446.2208,-251.6339"/>
<polygon fill="#191970" stroke="#191970" points="1447.1069,-255.0267 1456.242,-249.6601 1445.7541,-248.1586 1447.1069,-255.0267"/>
</g>
-<!-- Node21->Node22 -->
-<g id="edge24" class="edge">
-<title>Node21->Node22</title>
+<!-- Node19->Node20 -->
+<g id="edge22" class="edge">
+<title>Node19->Node20</title>
<path fill="none" stroke="#191970" d="M1214.9796,-291.4585C1243.0221,-277.6345 1274.3646,-256.1394 1256,-235 1220.3007,-193.9068 1072.6552,-208.182 1002.3528,-198.8051"/>
<polygon fill="#191970" stroke="#191970" points="1002.5925,-195.2987 992.163,-197.1634 1001.4789,-202.2095 1002.5925,-195.2987"/>
</g>
-<!-- Node23 -->
-<g id="node24" class="node">
-<title>Node23</title>
+<!-- Node21 -->
+<g id="node22" class="node">
+<title>Node21</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1522,-235.5 1522,-254.5 1580,-254.5 1580,-235.5 1522,-235.5"/>
<text text-anchor="middle" x="1551" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">memory</text>
</g>
-<!-- Node21->Node23 -->
-<g id="edge25" class="edge">
-<title>Node21->Node23</title>
+<!-- Node19->Node21 -->
+<g id="edge23" class="edge">
+<title>Node19->Node21</title>
<path fill="none" stroke="#191970" d="M1242.1631,-292.6203C1245.1466,-292.0535 1248.1037,-291.51 1251,-291 1363.5925,-271.1745 1395.976,-277.6154 1511.6652,-255.0764"/>
<polygon fill="#191970" stroke="#191970" points="1512.493,-258.4805 1521.6232,-253.1056 1511.134,-251.6137 1512.493,-258.4805"/>
</g>
-<!-- Node24 -->
-<g id="node25" class="node">
-<title>Node24</title>
-<g id="a_node25"><a xlink:href="runtime_2container_2base_8h.html" target="_top" xlink:title="Base utilities for common POD(plain old data) container types. ">
+<!-- Node22 -->
+<g id="node23" class="node">
+<title>Node22</title>
+<g id="a_node23"><a xlink:href="runtime_2container_2base_8h.html" target="_top" xlink:title="Base utilities for common POD(plain old data) container types. ">
<polygon fill="#ffffff" stroke="#000000" points="1192.5,-235.5 1192.5,-254.5 1247.5,-254.5 1247.5,-235.5 1192.5,-235.5"/>
<text text-anchor="middle" x="1220" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">./base.h</text>
</a>
</g>
</g>
-<!-- Node21->Node24 -->
-<g id="edge28" class="edge">
-<title>Node21->Node24</title>
+<!-- Node19->Node22 -->
+<g id="edge26" class="edge">
+<title>Node19->Node22</title>
<path fill="none" stroke="#191970" d="M1189.1348,-291.2977C1194.7668,-282.8498 1201.8151,-272.2773 1207.754,-263.369"/>
<polygon fill="#191970" stroke="#191970" points="1210.8005,-265.1089 1213.4354,-254.8469 1204.9762,-261.226 1210.8005,-265.1089"/>
</g>
-<!-- Node24->Node7 -->
-<g id="edge29" class="edge">
-<title>Node24->Node7</title>
+<!-- Node22->Node7 -->
+<g id="edge27" class="edge">
+<title>Node22->Node7</title>
<path fill="none" stroke="#191970" d="M1192.0714,-237.9249C1187.3906,-236.8616 1182.5704,-235.8413 1178,-235 1122.4429,-224.7739 965.6792,-205.514 881.7097,-195.4678"/>
<polygon fill="#191970" stroke="#191970" points="881.9398,-191.9705 871.5953,-194.2601 881.1098,-198.9211 881.9398,-191.9705"/>
</g>
-<!-- Node24->Node8 -->
-<g id="edge36" class="edge">
-<title>Node24->Node8</title>
+<!-- Node22->Node8 -->
+<g id="edge34" class="edge">
+<title>Node22->Node8</title>
<path fill="none" stroke="#191970" d="M1235.9293,-235.4348C1260.9815,-220.8487 1311.5337,-193.2424 1358,-179 1437.1946,-154.7261 1531.9527,-143.0184 1595.0418,-137.5376"/>
<polygon fill="#191970" stroke="#191970" points="1595.6034,-141.0029 1605.2757,-136.679 1595.0181,-134.0274 1595.6034,-141.0029"/>
</g>
-<!-- Node24->Node15 -->
-<g id="edge30" class="edge">
-<title>Node24->Node15</title>
+<!-- Node22->Node13 -->
+<g id="edge28" class="edge">
+<title>Node22->Node13</title>
<path fill="none" stroke="#191970" d="M1224.1943,-235.4248C1236.3115,-207.7625 1271.5304,-127.3614 1287.7034,-90.4403"/>
<polygon fill="#191970" stroke="#191970" points="1290.9184,-91.8236 1291.7249,-81.2595 1284.5066,-89.015 1290.9184,-91.8236"/>
</g>
-<!-- Node24->Node18 -->
-<g id="edge39" class="edge">
-<title>Node24->Node18</title>
+<!-- Node22->Node16 -->
+<g id="edge37" class="edge">
+<title>Node22->Node16</title>
<path fill="none" stroke="#191970" d="M1206.1194,-235.4791C1184.7006,-221.2037 1141.8869,-194.2997 1102,-179 942.5368,-117.8335 738.9838,-86.264 660.2038,-75.667"/>
<polygon fill="#191970" stroke="#191970" points="660.3056,-72.1501 649.9333,-74.31 659.3887,-79.0898 660.3056,-72.1501"/>
</g>
-<!-- Node24->Node22 -->
-<g id="edge37" class="edge">
-<title>Node24->Node22</title>
+<!-- Node22->Node20 -->
+<g id="edge35" class="edge">
+<title>Node22->Node20</title>
<path fill="none" stroke="#191970" d="M1192.4674,-238.307C1187.6516,-237.1756 1182.6844,-236.0338 1178,-235 1102.671,-218.3765 1081.0153,-216.6316 1002.1147,-199.1857"/>
<polygon fill="#191970" stroke="#191970" points="1002.7232,-195.7356 992.2015,-196.9783 1001.2016,-202.5682 1002.7232,-195.7356"/>
</g>
-<!-- Node24->Node25 -->
-<g id="edge31" class="edge">
-<title>Node24->Node25</title>
+<!-- Node22->Node23 -->
+<g id="edge29" class="edge">
+<title>Node22->Node23</title>
<path fill="none" stroke="#191970" d="M1192.1162,-237.6558C1187.43,-236.6249 1182.5964,-235.6851 1178,-235 1051.5674,-216.156 675.5557,-198.8278 514.0639,-192.0336"/>
<polygon fill="#191970" stroke="#191970" points="513.8796,-188.5229 503.7419,-191.6014 513.5866,-195.5168 513.8796,-188.5229"/>
</g>
-<!-- Node27 -->
-<g id="node28" class="node">
-<title>Node27</title>
+<!-- Node25 -->
+<g id="node26" class="node">
+<title>Node25</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1010.5,-179.5 1010.5,-198.5 1093.5,-198.5 1093.5,-179.5 1010.5,-179.5"/>
<text text-anchor="middle" x="1052" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">initializer_list</text>
</g>
-<!-- Node24->Node27 -->
-<g id="edge38" class="edge">
-<title>Node24->Node27</title>
+<!-- Node22->Node25 -->
+<g id="edge36" class="edge">
+<title>Node22->Node25</title>
<path fill="none" stroke="#191970" d="M1192.2582,-235.7527C1164.4226,-226.4742 1121.3332,-212.1111 1090.2591,-201.753"/>
<polygon fill="#191970" stroke="#191970" points="1091.2205,-198.3842 1080.6268,-198.5423 1089.0068,-205.025 1091.2205,-198.3842"/>
</g>
-<!-- Node25->Node8 -->
-<g id="edge32" class="edge">
-<title>Node25->Node8</title>
+<!-- Node23->Node8 -->
+<g id="edge30" class="edge">
+<title>Node23->Node8</title>
<path fill="none" stroke="#191970" d="M503.6186,-186.0484C713.8786,-176.4444 1377.4882,-146.1327 1595.1201,-136.1919"/>
<polygon fill="#191970" stroke="#191970" points="1595.3129,-139.6868 1605.1428,-135.7341 1594.9935,-132.6941 1595.3129,-139.6868"/>
</g>
-<!-- Node25->Node17 -->
-<g id="edge34" class="edge">
-<title>Node25->Node17</title>
+<!-- Node23->Node15 -->
+<g id="edge32" class="edge">
+<title>Node23->Node15</title>
<path fill="none" stroke="#191970" d="M422.0603,-179.3845C385.8075,-158.8062 300.5215,-110.3951 257.6091,-86.0365"/>
<polygon fill="#191970" stroke="#191970" points="259.3078,-82.9763 248.8834,-81.0836 255.8523,-89.0639 259.3078,-82.9763"/>
</g>
-<!-- Node25->Node18 -->
-<g id="edge35" class="edge">
-<title>Node25->Node18</title>
+<!-- Node23->Node16 -->
+<g id="edge33" class="edge">
+<title>Node23->Node16</title>
<path fill="none" stroke="#191970" d="M436.7749,-179.2164C434.1795,-165.0908 431.7718,-138.9349 445,-123 464.0123,-100.0974 547.3685,-83.8054 594.4025,-76.2468"/>
<polygon fill="#191970" stroke="#191970" points="595.0679,-79.6853 604.4072,-74.6826 593.9866,-72.7694 595.0679,-79.6853"/>
</g>
-<!-- Node26 -->
-<g id="node27" class="node">
-<title>Node26</title>
+<!-- Node24 -->
+<g id="node25" class="node">
+<title>Node24</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="454.5,-123.5 454.5,-142.5 503.5,-142.5 503.5,-123.5 454.5,-123.5"/>
<text text-anchor="middle" x="479" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstdlib</text>
</g>
-<!-- Node25->Node26 -->
-<g id="edge33" class="edge">
-<title>Node25->Node26</title>
+<!-- Node23->Node24 -->
+<g id="edge31" class="edge">
+<title>Node23->Node24</title>
<path fill="none" stroke="#191970" d="M445.9675,-179.2455C451.5311,-171.4564 459.478,-160.3308 466.1829,-150.9439"/>
<polygon fill="#191970" stroke="#191970" points="469.148,-152.8144 472.1124,-142.6427 463.4519,-148.7457 469.148,-152.8144"/>
</g>
-<!-- Node28->Node9 -->
-<g id="edge41" class="edge">
-<title>Node28->Node9</title>
+<!-- Node26->Node9 -->
+<g id="edge39" class="edge">
+<title>Node26->Node9</title>
<path fill="none" stroke="#191970" d="M2256.6339,-296.8306C2276.3624,-288.976 2300.1772,-275.7374 2312,-255 2340.7842,-204.5119 2321.5955,-133.1282 2308.0751,-96.0331"/>
<polygon fill="#191970" stroke="#191970" points="2311.2848,-94.6265 2304.4566,-86.5255 2304.7426,-97.1164 2311.2848,-94.6265"/>
</g>
-<!-- Node28->Node15 -->
-<g id="edge42" class="edge">
-<title>Node28->Node15</title>
+<!-- Node26->Node13 -->
+<g id="edge40" class="edge">
+<title>Node26->Node13</title>
<path fill="none" stroke="#191970" d="M2199.8093,-296.9865C2163.4363,-283.2575 2094.0862,-257.1102 2035,-235 1901.3146,-184.9745 1872.761,-156.4621 1734,-123 1607.8102,-92.5695 1456.6612,-79.8575 1368.9563,-74.742"/>
<polygon fill="#191970" stroke="#191970" points="1369.0243,-71.2403 1358.8424,-74.1695 1368.6286,-78.2291 1369.0243,-71.2403"/>
</g>
-<!-- Node28->Node16 -->
-<g id="edge43" class="edge">
-<title>Node28->Node16</title>
+<!-- Node26->Node14 -->
+<g id="edge41" class="edge">
+<title>Node26->Node14</title>
<path fill="none" stroke="#191970" d="M2245.2879,-296.9402C2266.5538,-286.8287 2300.8612,-270.2585 2330,-255 2444.7436,-194.9147 2578.8497,-117.9757 2633.6904,-86.2181"/>
<polygon fill="#191970" stroke="#191970" points="2635.4978,-89.216 2642.3933,-81.1721 2631.9866,-83.1602 2635.4978,-89.216"/>
</g>
-<!-- Node28->Node17 -->
-<g id="edge44" class="edge">
-<title>Node28->Node17</title>
+<!-- Node26->Node15 -->
+<g id="edge42" class="edge">
+<title>Node26->Node15</title>
<path fill="none" stroke="#191970" d="M2199.091,-296.9126C2151.0174,-279.266 2049.5768,-242.7355 2013,-235 1665.7744,-161.5667 518.4996,-88.906 276.9023,-74.1941"/>
<polygon fill="#191970" stroke="#191970" points="276.9273,-70.6893 266.7335,-73.5766 276.5029,-77.6764 276.9273,-70.6893"/>
</g>
-<!-- Node29->Node6 -->
-<g id="edge47" class="edge">
-<title>Node29->Node6</title>
+<!-- Node27->Node6 -->
+<g id="edge45" class="edge">
+<title>Node27->Node6</title>
<path fill="none" stroke="#191970" d="M2114.1816,-438.6985C1890.0761,-431.5091 1074.056,-404.5113 959,-389 843.5771,-373.4393 710.6286,-337.772 644.6427,-318.8366"/>
<polygon fill="#191970" stroke="#191970" points="645.6,-315.47 635.0214,-316.0577 643.6575,-322.1951 645.6,-315.47"/>
</g>
-<!-- Node29->Node16 -->
-<g id="edge86" class="edge">
-<title>Node29->Node16</title>
+<!-- Node27->Node14 -->
+<g id="edge84" class="edge">
+<title>Node27->Node14</title>
<path fill="none" stroke="#191970" d="M2227.7349,-437.055C2281.1548,-431.9977 2362.2623,-419.5541 2426,-389 2544.9458,-331.9807 2595.721,-317.3337 2654,-199 2671.1367,-164.2044 2667.1712,-117.3633 2662.9528,-91.2304"/>
<polygon fill="#191970" stroke="#191970" points="2666.345,-90.3183 2661.1309,-81.0953 2659.4554,-91.5568 2666.345,-90.3183"/>
</g>
-<!-- Node29->Node28 -->
-<g id="edge48" class="edge">
-<title>Node29->Node28</title>
+<!-- Node27->Node26 -->
+<g id="edge46" class="edge">
+<title>Node27->Node26</title>
<path fill="none" stroke="#191970" d="M2177.0895,-425.389C2186.9758,-400.8564 2206.4994,-352.4088 2217.3224,-325.5517"/>
<polygon fill="#191970" stroke="#191970" points="2220.6382,-326.6876 2221.1297,-316.1042 2214.1455,-324.0711 2220.6382,-326.6876"/>
</g>
-<!-- Node29->Node30 -->
-<g id="edge49" class="edge">
-<title>Node29->Node30</title>
+<!-- Node27->Node28 -->
+<g id="edge47" class="edge">
+<title>Node27->Node28</title>
<path fill="none" stroke="#191970" d="M2122.9742,-425.4639C2085.0142,-413.5792 2032.6531,-397.1858 1996.9597,-386.0108"/>
<polygon fill="#191970" stroke="#191970" points="1997.9491,-382.6531 1987.3602,-383.0053 1995.8576,-389.3333 1997.9491,-382.6531"/>
</g>
-<!-- Node40 -->
-<g id="node41" class="node">
-<title>Node40</title>
+<!-- Node38 -->
+<g id="node39" class="node">
+<title>Node38</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="2350.5,-297 2350.5,-316 2415.5,-316 2415.5,-297 2350.5,-297"/>
<text text-anchor="middle" x="2383" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">functional</text>
</g>
-<!-- Node29->Node40 -->
-<g id="edge85" class="edge">
-<title>Node29->Node40</title>
+<!-- Node27->Node38 -->
+<g id="edge83" class="edge">
+<title>Node27->Node38</title>
<path fill="none" stroke="#191970" d="M2227.949,-429.7856C2273.2594,-420.2519 2331.9521,-405.3217 2350,-389 2368.0873,-372.6426 2376.42,-345.0273 2380.1482,-326.2957"/>
<polygon fill="#191970" stroke="#191970" points="2383.6382,-326.6519 2381.8855,-316.203 2376.7397,-325.4643 2383.6382,-326.6519"/>
</g>
-<!-- Node30->Node8 -->
-<g id="edge75" class="edge">
-<title>Node30->Node8</title>
+<!-- Node28->Node8 -->
+<g id="edge73" class="edge">
+<title>Node28->Node8</title>
<path fill="none" stroke="#191970" d="M1934.7139,-363.8751C1917.5128,-355.3176 1894.4501,-341.1843 1881,-322 1858.1041,-289.3428 1882.1077,-267.5077 1859,-235 1832.7893,-198.127 1813.8961,-200.3287 1774,-179 1750.438,-166.4036 1722.6785,-154.7317 1701.0705,-146.2969"/>
<polygon fill="#191970" stroke="#191970" points="1702.2024,-142.9823 1691.6126,-142.6576 1699.6885,-149.5154 1702.2024,-142.9823"/>
</g>
-<!-- Node30->Node9 -->
-<g id="edge50" class="edge">
-<title>Node30->Node9</title>
+<!-- Node28->Node9 -->
+<g id="edge48" class="edge">
+<title>Node28->Node9</title>
<path fill="none" stroke="#191970" d="M1987.5586,-363.9705C2035.1114,-347.8993 2127.6169,-311.5217 2188,-255 2237.7849,-208.3987 2272.7375,-134.2606 2288.5267,-96.0506"/>
<polygon fill="#191970" stroke="#191970" points="2291.7983,-97.2961 2292.2974,-86.7131 2285.3075,-94.675 2291.7983,-97.2961"/>
</g>
-<!-- Node30->Node18 -->
-<g id="edge83" class="edge">
-<title>Node30->Node18</title>
+<!-- Node28->Node16 -->
+<g id="edge81" class="edge">
+<title>Node28->Node16</title>
<path fill="none" stroke="#191970" d="M1897.4721,-363.961C1881.1494,-361.6598 1863.4256,-359.4496 1847,-358 1736.4686,-348.2456 952.8412,-361.1069 849,-322 828.6026,-314.3183 830.3406,-301.05 811,-291 758.5611,-263.751 725.245,-296.3244 683,-255 659.2849,-231.8017 638.5698,-133.6012 630.5448,-91.1662"/>
<polygon fill="#191970" stroke="#191970" points="633.9654,-90.4155 628.7043,-81.2192 627.0822,-91.6891 633.9654,-90.4155"/>
</g>
-<!-- Node30->Node19 -->
-<g id="edge81" class="edge">
-<title>Node30->Node19</title>
+<!-- Node28->Node17 -->
+<g id="edge79" class="edge">
+<title>Node28->Node17</title>
<path fill="none" stroke="#191970" d="M1982.4821,-363.8529C2000.1723,-355.6533 2022.3961,-341.9948 2033,-322 2051.5888,-286.9489 2034.7351,-268.192 2013,-235 1970.9219,-170.7418 1898.5972,-113.8466 1861.8569,-87.321"/>
<polygon fill="#191970" stroke="#191970" points="1863.4992,-84.1939 1853.3229,-81.2455 1859.4395,-89.8964 1863.4992,-84.1939"/>
</g>
-<!-- Node30->Node20 -->
-<g id="edge84" class="edge">
-<title>Node30->Node20</title>
+<!-- Node28->Node18 -->
+<g id="edge82" class="edge">
+<title>Node28->Node18</title>
<path fill="none" stroke="#191970" d="M1894.498,-364.4216C1801.3835,-350.8164 1635.7424,-326.2718 1623,-322 1577.6037,-306.7814 1529.5149,-277.8744 1502.2783,-260.1337"/>
<polygon fill="#191970" stroke="#191970" points="1504.1477,-257.1739 1493.875,-254.5812 1500.2888,-263.0141 1504.1477,-257.1739"/>
</g>
-<!-- Node30->Node28 -->
-<g id="edge74" class="edge">
-<title>Node30->Node28</title>
+<!-- Node28->Node26 -->
+<g id="edge72" class="edge">
+<title>Node28->Node26</title>
<path fill="none" stroke="#191970" d="M1995.1133,-363.9717C2042.8021,-352.0495 2124.5167,-331.6208 2176.5849,-318.6038"/>
<polygon fill="#191970" stroke="#191970" points="2177.7318,-321.9248 2186.5843,-316.1039 2176.034,-315.1339 2177.7318,-321.9248"/>
</g>
-<!-- Node31 -->
-<g id="node32" class="node">
-<title>Node31</title>
-<g id="a_node32"><a xlink:href="optional_8h.html" target="_top" xlink:title="Runtime Optional container types. ">
+<!-- Node29 -->
+<g id="node30" class="node">
+<title>Node29</title>
+<g id="a_node30"><a xlink:href="optional_8h.html" target="_top" xlink:title="Runtime Optional container types. ">
<polygon fill="#ffffff" stroke="#000000" points="676,-291.5 676,-321.5 802,-321.5 802,-291.5 676,-291.5"/>
<text text-anchor="start" x="684" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
<text text-anchor="middle" x="739" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/optional.h</text>
</a>
</g>
</g>
-<!-- Node30->Node31 -->
-<g id="edge51" class="edge">
-<title>Node30->Node31</title>
+<!-- Node28->Node29 -->
+<g id="edge49" class="edge">
+<title>Node28->Node29</title>
<path fill="none" stroke="#191970" d="M1897.7594,-363.9887C1881.3579,-361.673 1863.5225,-359.4485 1847,-358 1390.2504,-317.9572 1271.3789,-375.4193 816,-322 814.8827,-321.8689 813.7574,-321.7302 812.6258,-321.5845"/>
<polygon fill="#191970" stroke="#191970" points="812.7142,-318.0612 802.3174,-320.1002 811.7165,-324.9897 812.7142,-318.0612"/>
</g>
-<!-- Node32 -->
-<g id="node33" class="node">
-<title>Node32</title>
-<g id="a_node33"><a xlink:href="shape__tuple_8h.html" target="_top" xlink:title="Runtime ShapeTuple container types. ">
+<!-- Node30 -->
+<g id="node31" class="node">
+<title>Node30</title>
+<g id="a_node31"><a xlink:href="shape__tuple_8h.html" target="_top" xlink:title="Runtime ShapeTuple container types. ">
<polygon fill="#ffffff" stroke="#000000" points="1260,-291.5 1260,-321.5 1386,-321.5 1386,-291.5 1260,-291.5"/>
<text text-anchor="start" x="1268" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
<text text-anchor="middle" x="1323" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/shape_tuple.h</text>
</a>
</g>
</g>
-<!-- Node30->Node32 -->
-<g id="edge54" class="edge">
-<title>Node30->Node32</title>
+<!-- Node28->Node30 -->
+<g id="edge52" class="edge">
+<title>Node28->Node30</title>
<path fill="none" stroke="#191970" d="M1894.2355,-364.1918C1878.8067,-362.0318 1862.3196,-359.8313 1847,-358 1686.4163,-338.8035 1497.5154,-321.5427 1396.1376,-312.7109"/>
<polygon fill="#191970" stroke="#191970" points="1396.2724,-309.2095 1386.007,-311.8311 1395.6667,-316.1833 1396.2724,-309.2095"/>
</g>
-<!-- Node30->Node33 -->
-<g id="edge58" class="edge">
-<title>Node30->Node33</title>
+<!-- Node28->Node31 -->
+<g id="edge56" class="edge">
+<title>Node28->Node31</title>
<path fill="none" stroke="#191970" d="M1896.5992,-363.9771C1880.5155,-361.7186 1863.1282,-359.5254 1847,-358 1519.1851,-326.9955 1435.5636,-343.6757 1107,-322 1069.9329,-319.5546 1028.7641,-316.1911 994.6006,-313.2221"/>
<polygon fill="#191970" stroke="#191970" points="994.5024,-309.7002 984.2353,-312.3148 993.892,-316.6735 994.5024,-309.7002"/>
</g>
-<!-- Node37 -->
-<g id="node38" class="node">
-<title>Node37</title>
-<g id="a_node38"><a xlink:href="serializer_8h.html" target="_top" xlink:title="Serializer extension to support TVM data types Include this file to enable serialization of DLDataTyp...">
+<!-- Node35 -->
+<g id="node36" class="node">
+<title>Node35</title>
+<g id="a_node36"><a xlink:href="serializer_8h.html" target="_top" xlink:title="Serializer extension to support TVM data types Include this file to enable serialization of DLDataTyp...">
<polygon fill="#ffffff" stroke="#000000" points="1890,-297 1890,-316 2024,-316 2024,-297 1890,-297"/>
<text text-anchor="middle" x="1957" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/serializer.h</text>
</a>
</g>
</g>
-<!-- Node30->Node37 -->
-<g id="edge76" class="edge">
-<title>Node30->Node37</title>
+<!-- Node28->Node35 -->
+<g id="edge74" class="edge">
+<title>Node28->Node35</title>
<path fill="none" stroke="#191970" d="M1952.0994,-363.9005C1950.2106,-354.149 1949.7686,-338.7597 1950.7734,-326.3695"/>
<polygon fill="#191970" stroke="#191970" points="1954.2923,-326.4478 1952.1029,-316.0817 1947.35,-325.5506 1954.2923,-326.4478"/>
</g>
-<!-- Node30->Node40 -->
-<g id="edge82" class="edge">
-<title>Node30->Node40</title>
+<!-- Node28->Node38 -->
+<g id="edge80" class="edge">
+<title>Node28->Node38</title>
<path fill="none" stroke="#191970" d="M2019.6367,-365.0144C2089.0457,-355.4165 2204.2307,-338.9106 2303,-322 2315.2012,-319.911 2328.4078,-317.4392 2340.507,-315.0904"/>
<polygon fill="#191970" stroke="#191970" points="2341.2241,-318.5165 2350.3636,-313.1574 2339.8769,-311.6474 2341.2241,-318.5165"/>
</g>
-<!-- Node31->Node18 -->
-<g id="edge52" class="edge">
-<title>Node31->Node18</title>
+<!-- Node29->Node16 -->
+<g id="edge50" class="edge">
+<title>Node29->Node16</title>
<path fill="none" stroke="#191970" d="M675.7064,-292.988C672.4249,-292.3103 669.1743,-291.6441 666,-291 624.7405,-282.6276 506.7922,-287.475 480,-255 474.3432,-248.1434 474.8441,-242.2408 480,-235 508.3715,-195.1559 549.5513,-233.7243 584,-199 613.3539,-169.4112 622.6782,-119.1351 625.6338,-91.3797"/>
<polygon fill="#191970" stroke="#191970" points="629.1303,-91.5734 626.5285,-81.303 622.1577,-90.9542 629.1303,-91.5734"/>
</g>
-<!-- Node31->Node24 -->
-<g id="edge53" class="edge">
-<title>Node31->Node24</title>
+<!-- Node29->Node22 -->
+<g id="edge51" class="edge">
+<title>Node29->Node22</title>
<path fill="none" stroke="#191970" d="M802.0859,-297.4058C817.4302,-295.2504 833.7985,-292.9984 849,-291 979.957,-273.7844 1012.9655,-271.6156 1144,-255 1156.3556,-253.4333 1169.8343,-251.6756 1181.9637,-250.076"/>
<polygon fill="#191970" stroke="#191970" points="1182.759,-253.5014 1192.2134,-248.7198 1181.8407,-246.5619 1182.759,-253.5014"/>
</g>
-<!-- Node32->Node18 -->
-<g id="edge55" class="edge">
-<title>Node32->Node18</title>
+<!-- Node30->Node16 -->
+<g id="edge53" class="edge">
+<title>Node30->Node16</title>
<path fill="none" stroke="#191970" d="M1318.0569,-291.2884C1312.6655,-276.1056 1303.0331,-252.5678 1290,-235 1267.6791,-204.9129 1259.3638,-197.869 1227,-179 1160.462,-140.2065 1138.1086,-140.0571 1063,-123 914.8861,-89.3634 733.1805,-76.8881 659.9221,-73.0108"/>
<polygon fill="#191970" stroke="#191970" points="659.6819,-69.4941 649.5165,-72.4799 659.3252,-76.485 659.6819,-69.4941"/>
</g>
-<!-- Node32->Node20 -->
-<g id="edge56" class="edge">
-<title>Node32->Node20</title>
+<!-- Node30->Node18 -->
+<g id="edge54" class="edge">
+<title>Node30->Node18</title>
<path fill="none" stroke="#191970" d="M1361.4065,-291.4554C1387.5329,-281.2212 1421.7586,-267.8143 1446.7235,-258.0351"/>
<polygon fill="#191970" stroke="#191970" points="1448.1005,-261.2547 1456.1351,-254.3484 1445.5474,-254.7369 1448.1005,-261.2547"/>
</g>
-<!-- Node32->Node24 -->
-<g id="edge57" class="edge">
-<title>Node32->Node24</title>
+<!-- Node30->Node22 -->
+<g id="edge55" class="edge">
+<title>Node30->Node22</title>
<path fill="none" stroke="#191970" d="M1297.8034,-291.4554C1281.6728,-281.8241 1260.8362,-269.3828 1244.7864,-259.7996"/>
<polygon fill="#191970" stroke="#191970" points="1246.3416,-256.6518 1235.9613,-254.5303 1242.753,-262.662 1246.3416,-256.6518"/>
</g>
-<!-- Node33->Node7 -->
-<g id="edge59" class="edge">
-<title>Node33->Node7</title>
+<!-- Node31->Node7 -->
+<g id="edge57" class="edge">
+<title>Node31->Node7</title>
<path fill="none" stroke="#191970" d="M908.7111,-291.1389C891.2365,-269.2957 859.2979,-229.3724 840.954,-206.4424"/>
<polygon fill="#191970" stroke="#191970" points="843.6276,-204.1817 834.6476,-198.5595 838.1615,-208.5546 843.6276,-204.1817"/>
</g>
-<!-- Node33->Node8 -->
-<g id="edge63" class="edge">
-<title>Node33->Node8</title>
+<!-- Node31->Node8 -->
+<g id="edge61" class="edge">
+<title>Node31->Node8</title>
<path fill="none" stroke="#191970" d="M984.2937,-301.082C1058.2855,-293.964 1184.1099,-279.4142 1290,-255 1418.5496,-225.3614 1566.0399,-171.2206 1631.1266,-146.2553"/>
<polygon fill="#191970" stroke="#191970" points="1632.7778,-149.37 1640.8496,-142.5073 1630.2599,-142.8384 1632.7778,-149.37"/>
</g>
-<!-- Node33->Node15 -->
-<g id="edge61" class="edge">
-<title>Node33->Node15</title>
+<!-- Node31->Node13 -->
+<g id="edge59" class="edge">
+<title>Node31->Node13</title>
<path fill="none" stroke="#191970" d="M915.0521,-291.3881C905.8034,-265.0948 891.7061,-210.713 919,-179 954.0365,-138.2908 1106.9121,-159.5775 1158,-143 1200.7034,-129.1432 1246.191,-103.0959 1272.8054,-86.5574"/>
<polygon fill="#191970" stroke="#191970" points="1274.8049,-89.4343 1281.3966,-81.1396 1271.071,-83.5133 1274.8049,-89.4343"/>
</g>
-<!-- Node33->Node16 -->
-<g id="edge69" class="edge">
-<title>Node33->Node16</title>
+<!-- Node31->Node14 -->
+<g id="edge67" class="edge">
+<title>Node31->Node14</title>
<path fill="none" stroke="#191970" d="M984.2264,-300.5455C1020.1908,-297.3306 1066.0868,-293.5164 1107,-291 1308.1122,-278.6306 1818.9243,-309.1639 2013,-255 2129.8799,-222.3804 2136.6333,-160.6225 2252,-123 2321.8164,-100.232 2543.2389,-80.728 2626.7259,-74.0027"/>
<polygon fill="#191970" stroke="#191970" points="2627.1947,-77.4765 2636.8849,-73.1928 2626.6383,-70.4986 2627.1947,-77.4765"/>
</g>
-<!-- Node33->Node17 -->
-<g id="edge72" class="edge">
-<title>Node33->Node17</title>
+<!-- Node31->Node15 -->
+<g id="edge70" class="edge">
+<title>Node31->Node15</title>
<path fill="none" stroke="#191970" d="M857.6715,-297.0866C777.744,-285.0974 644.8049,-264.7661 596,-255 492.4131,-234.2719 457.6544,-249.7451 365,-199 313.3578,-170.7165 267.2164,-117.1851 245.3662,-89.3388"/>
<polygon fill="#191970" stroke="#191970" points="248.0234,-87.0529 239.1493,-81.2649 242.4771,-91.3236 248.0234,-87.0529"/>
</g>
-<!-- Node33->Node18 -->
-<g id="edge71" class="edge">
-<title>Node33->Node18</title>
+<!-- Node31->Node16 -->
+<g id="edge69" class="edge">
+<title>Node31->Node16</title>
<path fill="none" stroke="#191970" d="M857.9011,-293.1727C795.6156,-279.7772 708.8303,-260.3658 703,-255 657.5997,-213.2169 686.4143,-178.7611 660,-123 654.4879,-111.3639 646.7944,-99.1974 640.2224,-89.6132"/>
<polygon fill="#191970" stroke="#191970" points="642.9959,-87.4729 634.3728,-81.3171 637.275,-91.5067 642.9959,-87.4729"/>
</g>
-<!-- Node33->Node20 -->
-<g id="edge73" class="edge">
-<title>Node33->Node20</title>
+<!-- Node31->Node18 -->
+<g id="edge71" class="edge">
+<title>Node31->Node18</title>
<path fill="none" stroke="#191970" d="M984.0378,-299.5647C1102.69,-286.5109 1355.4421,-258.7036 1446.4606,-248.6899"/>
<polygon fill="#191970" stroke="#191970" points="1446.9357,-252.1589 1456.493,-247.5862 1446.1702,-245.2009 1446.9357,-252.1589"/>
</g>
-<!-- Node33->Node22 -->
-<g id="edge64" class="edge">
-<title>Node33->Node22</title>
+<!-- Node31->Node20 -->
+<g id="edge62" class="edge">
+<title>Node31->Node20</title>
<path fill="none" stroke="#191970" d="M926.0986,-291.1389C933.157,-269.873 945.9035,-231.4702 953.5955,-208.2956"/>
<polygon fill="#191970" stroke="#191970" points="956.9986,-209.1529 956.8271,-198.5595 950.355,-206.9478 956.9986,-209.1529"/>
</g>
-<!-- Node33->Node23 -->
-<g id="edge68" class="edge">
-<title>Node33->Node23</title>
+<!-- Node31->Node21 -->
+<g id="edge66" class="edge">
+<title>Node31->Node21</title>
<path fill="none" stroke="#191970" d="M984.1264,-301.1506C1122.9344,-289.3703 1446.9397,-261.7718 1511.3849,-254.5721"/>
<polygon fill="#191970" stroke="#191970" points="1512.1785,-257.999 1521.6507,-253.2525 1511.286,-251.0561 1512.1785,-257.999"/>
</g>
-<!-- Node33->Node24 -->
-<g id="edge60" class="edge">
-<title>Node33->Node24</title>
+<!-- Node31->Node22 -->
+<g id="edge58" class="edge">
+<title>Node31->Node22</title>
<path fill="none" stroke="#191970" d="M984.0449,-293.5326C1044.0286,-281.1948 1132.3039,-263.0378 1182.1341,-252.7885"/>
<polygon fill="#191970" stroke="#191970" points="1183.0699,-256.1693 1192.1597,-250.7263 1181.6596,-249.3129 1183.0699,-256.1693"/>
</g>
-<!-- Node33->Node25 -->
-<g id="edge62" class="edge">
-<title>Node33->Node25</title>
+<!-- Node31->Node23 -->
+<g id="edge60" class="edge">
+<title>Node31->Node23</title>
<path fill="none" stroke="#191970" d="M857.7805,-294.9067C806.8462,-285.2356 733.5123,-270.5651 670,-255 602.6325,-238.4901 525.3112,-215.5791 479.396,-201.5502"/>
<polygon fill="#191970" stroke="#191970" points="480.3992,-198.1971 469.8125,-198.6116 478.3471,-204.8895 480.3992,-198.1971"/>
</g>
-<!-- Node33->Node27 -->
-<g id="edge67" class="edge">
-<title>Node33->Node27</title>
+<!-- Node31->Node25 -->
+<g id="edge65" class="edge">
+<title>Node31->Node25</title>
<path fill="none" stroke="#191970" d="M937.8578,-291.3795C962.5504,-269.2315 1008.3724,-228.1316 1033.804,-205.3209"/>
<polygon fill="#191970" stroke="#191970" points="1036.1846,-207.8872 1041.2919,-198.6046 1031.5106,-202.6762 1036.1846,-207.8872"/>
</g>
-<!-- Node34 -->
-<g id="node35" class="node">
-<title>Node34</title>
+<!-- Node32 -->
+<g id="node33" class="node">
+<title>Node32</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="780.5,-235.5 780.5,-254.5 833.5,-254.5 833.5,-235.5 780.5,-235.5"/>
<text text-anchor="middle" x="807" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstddef</text>
</g>
-<!-- Node33->Node34 -->
-<g id="edge65" class="edge">
-<title>Node33->Node34</title>
+<!-- Node31->Node32 -->
+<g id="edge63" class="edge">
+<title>Node31->Node32</title>
<path fill="none" stroke="#191970" d="M893.1125,-291.4554C875.0924,-281.734 851.7659,-269.15 833.9368,-259.5317"/>
<polygon fill="#191970" stroke="#191970" points="835.1287,-256.1979 824.666,-254.5303 831.8052,-262.3586 835.1287,-256.1979"/>
</g>
-<!-- Node35 -->
-<g id="node36" class="node">
-<title>Node35</title>
+<!-- Node33 -->
+<g id="node34" class="node">
+<title>Node33</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="712,-235.5 712,-254.5 762,-254.5 762,-235.5 712,-235.5"/>
<text text-anchor="middle" x="737" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstring</text>
</g>
-<!-- Node33->Node35 -->
-<g id="edge66" class="edge">
-<title>Node33->Node35</title>
+<!-- Node31->Node33 -->
+<g id="edge64" class="edge">
+<title>Node31->Node33</title>
<path fill="none" stroke="#191970" d="M875.9885,-291.4554C843.9158,-280.7354 801.4283,-266.5344 771.9086,-256.6678"/>
<polygon fill="#191970" stroke="#191970" points="772.64,-253.222 762.0463,-253.3714 770.421,-259.861 772.64,-253.222"/>
</g>
-<!-- Node36 -->
-<g id="node37" class="node">
-<title>Node36</title>
+<!-- Node34 -->
+<g id="node35" class="node">
+<title>Node34</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1042.5,-235.5 1042.5,-254.5 1135.5,-254.5 1135.5,-235.5 1042.5,-235.5"/>
<text text-anchor="middle" x="1089" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">unordered_map</text>
</g>
-<!-- Node33->Node36 -->
-<g id="edge70" class="edge">
-<title>Node33->Node36</title>
+<!-- Node31->Node34 -->
+<g id="edge68" class="edge">
+<title>Node31->Node34</title>
<path fill="none" stroke="#191970" d="M962.0974,-291.4554C990.0057,-281.239 1026.5508,-267.8609 1053.2527,-258.0861"/>
<polygon fill="#191970" stroke="#191970" points="1054.7786,-261.2547 1062.966,-254.5303 1052.3722,-254.6813 1054.7786,-261.2547"/>
</g>
-<!-- Node37->Node9 -->
-<g id="edge79" class="edge">
-<title>Node37->Node9</title>
+<!-- Node35->Node9 -->
+<g id="edge77" class="edge">
+<title>Node35->Node9</title>
<path fill="none" stroke="#191970" d="M1968.0166,-296.9187C1979.5185,-286.7896 1997.9447,-270.2052 2013,-255 2068.5751,-198.8714 2065.4572,-165.5566 2132,-123 2159.5206,-105.3995 2193.4523,-93.4751 2223.4506,-85.5736"/>
<polygon fill="#191970" stroke="#191970" points="2224.6199,-88.8882 2233.4583,-83.0458 2222.9056,-82.1014 2224.6199,-88.8882"/>
</g>
-<!-- Node37->Node30 -->
-<g id="edge80" class="edge">
-<title>Node37->Node30</title>
+<!-- Node35->Node28 -->
+<g id="edge78" class="edge">
+<title>Node35->Node28</title>
<path fill="none" stroke="#191970" d="M1961.8971,-316.0817C1963.7877,-325.8263 1964.2315,-341.214 1963.2285,-353.6079"/>
<polygon fill="#191970" stroke="#191970" points="1959.709,-353.5349 1961.9006,-363.9005 1966.6514,-354.4306 1959.709,-353.5349"/>
</g>
-<!-- Node38 -->
-<g id="node39" class="node">
-<title>Node38</title>
+<!-- Node36 -->
+<g id="node37" class="node">
+<title>Node36</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1788,-235.5 1788,-254.5 1850,-254.5 1850,-235.5 1788,-235.5"/>
<text text-anchor="middle" x="1819" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/io.h</text>
</g>
-<!-- Node37->Node38 -->
-<g id="edge77" class="edge">
-<title>Node37->Node38</title>
+<!-- Node35->Node36 -->
+<g id="edge75" class="edge">
+<title>Node35->Node36</title>
<path fill="none" stroke="#191970" d="M1935.4375,-296.8906C1912.4488,-286.6457 1875.844,-270.3326 1849.8176,-258.7339"/>
<polygon fill="#191970" stroke="#191970" points="1850.9698,-255.4156 1840.4111,-254.5419 1848.1204,-261.8095 1850.9698,-255.4156"/>
</g>
-<!-- Node39 -->
-<g id="node40" class="node">
-<title>Node39</title>
+<!-- Node37 -->
+<g id="node38" class="node">
+<title>Node37</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1906.5,-235.5 1906.5,-254.5 2003.5,-254.5 2003.5,-235.5 1906.5,-235.5"/>
<text text-anchor="middle" x="1955" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/serializer.h</text>
</g>
-<!-- Node37->Node39 -->
-<g id="edge78" class="edge">
-<title>Node37->Node39</title>
+<!-- Node35->Node37 -->
+<g id="edge76" class="edge">
+<title>Node35->Node37</title>
<path fill="none" stroke="#191970" d="M1956.6875,-296.8906C1956.4103,-288.3657 1955.9964,-275.6392 1955.6479,-264.9235"/>
<polygon fill="#191970" stroke="#191970" points="1959.1402,-264.6258 1955.3169,-254.7449 1952.1439,-264.8534 1959.1402,-264.6258"/>
</g>
-<!-- Node41->Node8 -->
-<g id="edge114" class="edge">
-<title>Node41->Node8</title>
+<!-- Node39->Node8 -->
+<g id="edge112" class="edge">
+<title>Node39->Node8</title>
<path fill="none" stroke="#191970" d="M1616.9712,-425.4718C1634.5826,-415.685 1657.5761,-402.3042 1677,-389 1695.2678,-376.4877 1698.2815,-371.2788 1716,-358 1738.202,-341.3612 1754.2356,-346.6343 1767,-322 1773.3387,-309.7669 1770.8952,-304.2157 1767,-291 1764.6927,-283.1717 1703.7536,-191.1678 1677.1884,-151.2686"/>
<polygon fill="#191970" stroke="#191970" points="1679.968,-149.1282 1671.51,-142.7475 1674.1429,-153.01 1679.968,-149.1282"/>
</g>
-<!-- Node41->Node9 -->
-<g id="edge93" class="edge">
-<title>Node41->Node9</title>
+<!-- Node39->Node9 -->
+<g id="edge91" class="edge">
+<title>Node39->Node9</title>
<path fill="none" stroke="#191970" d="M1647.091,-438.6646C1799.5796,-432.6689 2206.1922,-408.7255 2303,-322 2369.6203,-262.318 2364.5617,-206.6902 2333,-123 2329.2629,-113.0905 2323.2603,-103.2233 2317.2734,-94.8172"/>
<polygon fill="#191970" stroke="#191970" points="2319.94,-92.5395 2311.1358,-86.6456 2314.3429,-96.7434 2319.94,-92.5395"/>
</g>
-<!-- Node41->Node15 -->
-<g id="edge102" class="edge">
-<title>Node41->Node15</title>
+<!-- Node39->Node13 -->
+<g id="edge100" class="edge">
+<title>Node39->Node13</title>
<path fill="none" stroke="#191970" d="M1530.8986,-432.4454C1509.8583,-429.7535 1485.8906,-426.9465 1464,-425 1168.7329,-398.7451 1090.9435,-434.3443 798,-389 680.499,-370.8122 611.7898,-416.0156 539,-322 449.9823,-207.0243 1016.8395,-112.0605 1223.0706,-81.6907"/>
<polygon fill="#191970" stroke="#191970" points="1223.8143,-85.1193 1233.2021,-80.2081 1222.8006,-78.193 1223.8143,-85.1193"/>
</g>
-<!-- Node41->Node16 -->
-<g id="edge118" class="edge">
-<title>Node41->Node16</title>
+<!-- Node39->Node14 -->
+<g id="edge116" class="edge">
+<title>Node39->Node14</title>
<path fill="none" stroke="#191970" d="M1647.1704,-439.8308C1796.3121,-437.5635 2189.2753,-427.9594 2313,-389 2479.4771,-336.5783 2609.4859,-149.7963 2647.5982,-90.0337"/>
<polygon fill="#191970" stroke="#191970" points="2650.7425,-91.6081 2653.0994,-81.2787 2644.8154,-87.8838 2650.7425,-91.6081"/>
</g>
-<!-- Node41->Node17 -->
-<g id="edge120" class="edge">
-<title>Node41->Node17</title>
+<!-- Node39->Node15 -->
+<g id="edge118" class="edge">
+<title>Node39->Node15</title>
<path fill="none" stroke="#191970" d="M1530.9156,-432.2452C1509.8769,-429.5338 1485.9059,-426.7656 1464,-425 1357.6034,-416.4244 603.6262,-427.3198 504,-389 381.2049,-341.7686 360.0188,-301.0652 277,-199 259.1833,-177.0957 256.6633,-169.5303 247,-143 240.8429,-126.0958 236.9038,-105.9734 234.615,-91.3888"/>
<polygon fill="#191970" stroke="#191970" points="238.0744,-90.8553 233.1771,-81.4602 231.1466,-91.8587 238.0744,-90.8553"/>
</g>
-<!-- Node41->Node18 -->
-<g id="edge121" class="edge">
-<title>Node41->Node18</title>
+<!-- Node39->Node16 -->
+<g id="edge119" class="edge">
+<title>Node39->Node16</title>
<path fill="none" stroke="#191970" d="M1530.9128,-432.2798C1509.8738,-429.5718 1485.9033,-426.7969 1464,-425 1365.1403,-416.8899 665.509,-422.0919 572,-389 492.4629,-360.8526 422.1841,-309.9116 461,-235 473.7644,-210.3657 491.2474,-217.4148 512,-199 552.0897,-163.4265 592.5941,-114.9283 613.1548,-89.2087"/>
<polygon fill="#191970" stroke="#191970" points="616.0965,-91.1318 619.5623,-81.1198 610.6094,-86.7853 616.0965,-91.1318"/>
</g>
-<!-- Node41->Node20 -->
-<g id="edge122" class="edge">
-<title>Node41->Node20</title>
+<!-- Node39->Node18 -->
+<g id="edge120" class="edge">
+<title>Node39->Node18</title>
<path fill="none" stroke="#191970" d="M1580.6279,-425.4841C1561.042,-390.3551 1512.2606,-302.8618 1490.624,-264.0549"/>
<polygon fill="#191970" stroke="#191970" points="1493.4407,-261.9196 1485.514,-254.8898 1487.3268,-265.3284 1493.4407,-261.9196"/>
</g>
-<!-- Node41->Node21 -->
-<g id="edge94" class="edge">
-<title>Node41->Node21</title>
+<!-- Node39->Node19 -->
+<g id="edge92" class="edge">
+<title>Node39->Node19</title>
<path fill="none" stroke="#191970" d="M1542.7648,-425.389C1466.8958,-400.5928 1316.273,-351.3648 1234.6665,-324.6934"/>
<polygon fill="#191970" stroke="#191970" points="1235.6588,-321.3356 1225.0663,-321.5558 1233.4842,-327.9893 1235.6588,-321.3356"/>
</g>
-<!-- Node41->Node23 -->
-<g id="edge117" class="edge">
-<title>Node41->Node23</title>
+<!-- Node39->Node21 -->
+<g id="edge115" class="edge">
+<title>Node39->Node21</title>
<path fill="none" stroke="#191970" d="M1586.0813,-425.4841C1579.2967,-390.5793 1562.4635,-303.9766 1554.8494,-264.804"/>
<polygon fill="#191970" stroke="#191970" points="1558.2661,-264.0383 1552.9223,-254.8898 1551.3947,-265.374 1558.2661,-264.0383"/>
</g>
-<!-- Node41->Node28 -->
-<g id="edge101" class="edge">
-<title>Node41->Node28</title>
+<!-- Node39->Node26 -->
+<g id="edge99" class="edge">
+<title>Node39->Node26</title>
<path fill="none" stroke="#191970" d="M1647.2624,-437.3213C1750.4097,-431.0939 1960.7156,-415.5302 2029,-389 2050.9605,-380.4678 2051.0448,-368.7677 2072,-358 2105.4036,-340.8357 2145.7571,-327.5301 2176.633,-318.7464"/>
<polygon fill="#191970" stroke="#191970" points="2177.6088,-322.1079 2186.3038,-316.0542 2175.7315,-315.3643 2177.6088,-322.1079"/>
</g>
-<!-- Node41->Node30 -->
-<g id="edge113" class="edge">
-<title>Node41->Node30</title>
+<!-- Node39->Node28 -->
+<g id="edge111" class="edge">
+<title>Node39->Node28</title>
<path fill="none" stroke="#191970" d="M1647.3107,-429.8837C1714.8229,-417.592 1825.5788,-397.4272 1894.7298,-384.8372"/>
<polygon fill="#191970" stroke="#191970" points="1895.4091,-388.2712 1904.6204,-383.0365 1894.1552,-381.3844 1895.4091,-388.2712"/>
</g>
-<!-- Node41->Node40 -->
-<g id="edge115" class="edge">
-<title>Node41->Node40</title>
+<!-- Node39->Node38 -->
+<g id="edge113" class="edge">
+<title>Node39->Node38</title>
<path fill="none" stroke="#191970" d="M1647.2121,-438.0108C1802.0379,-431.1247 2216.1321,-410.8785 2274,-389 2311.6563,-374.763 2347.2666,-343.1711 2367.0091,-323.4744"/>
<polygon fill="#191970" stroke="#191970" points="2369.7427,-325.6848 2374.2295,-316.0869 2364.7367,-320.792 2369.7427,-325.6848"/>
</g>
-<!-- Node42 -->
-<g id="node43" class="node">
-<title>Node42</title>
-<g id="a_node43"><a xlink:href="map_8h.html" target="_top" xlink:title="Runtime Map container types. ">
+<!-- Node40 -->
+<g id="node41" class="node">
+<title>Node40</title>
+<g id="a_node41"><a xlink:href="map_8h.html" target="_top" xlink:title="Runtime Map container types. ">
<polygon fill="#ffffff" stroke="#000000" points="968,-358.5 968,-388.5 1094,-388.5 1094,-358.5 968,-358.5"/>
<text text-anchor="start" x="976" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
<text text-anchor="middle" x="1031" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/map.h</text>
</a>
</g>
</g>
-<!-- Node41->Node42 -->
-<g id="edge95" class="edge">
-<title>Node41->Node42</title>
+<!-- Node39->Node40 -->
+<g id="edge93" class="edge">
+<title>Node39->Node40</title>
<path fill="none" stroke="#191970" d="M1530.8633,-433.5194C1428.5441,-421.2338 1216.5234,-395.7761 1104.323,-382.304"/>
<polygon fill="#191970" stroke="#191970" points="1104.459,-378.7953 1094.113,-381.0781 1103.6244,-385.7454 1104.459,-378.7953"/>
</g>
-<!-- Node43 -->
-<g id="node44" class="node">
-<title>Node43</title>
-<g id="a_node44"><a xlink:href="runtime_2module_8h.html" target="_top" xlink:title="Runtime container of the functions generated by TVM, This is used to support dynamically link...">
+<!-- Node41 -->
+<g id="node42" class="node">
+<title>Node41</title>
+<g id="a_node42"><a xlink:href="runtime_2module_8h.html" target="_top" xlink:title="Runtime container of the functions generated by TVM, This is used to support dynamically link...">
<polygon fill="#ffffff" stroke="#000000" points="1632,-297 1632,-316 1758,-316 1758,-297 1632,-297"/>
<text text-anchor="middle" x="1695" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/module.h</text>
</a>
</g>
</g>
-<!-- Node41->Node43 -->
-<g id="edge103" class="edge">
-<title>Node41->Node43</title>
+<!-- Node39->Node41 -->
+<g id="edge101" class="edge">
+<title>Node39->Node41</title>
<path fill="none" stroke="#191970" d="M1584.9126,-425.4163C1585.674,-407.7345 1593.2757,-378.2747 1609,-358 1621.391,-342.0232 1640.1818,-329.3183 1657.1642,-320.5248"/>
<polygon fill="#191970" stroke="#191970" points="1658.9027,-323.5712 1666.3608,-316.046 1655.8378,-317.2778 1658.9027,-323.5712"/>
</g>
-<!-- Node41->Node44 -->
-<g id="edge116" class="edge">
-<title>Node41->Node44</title>
+<!-- Node39->Node42 -->
+<g id="edge114" class="edge">
+<title>Node39->Node42</title>
<path fill="none" stroke="#191970" d="M1530.7391,-433.4455C1503.5151,-430.3939 1470.6389,-427.0571 1441,-425 1207.8683,-408.8193 618.0394,-440.0906 390,-389 387.3523,-388.4068 384.6559,-387.6223 381.9904,-386.7206"/>
<polygon fill="#191970" stroke="#191970" points="383.202,-383.4365 372.6134,-383.0732 380.6643,-389.9603 383.202,-383.4365"/>
</g>
-<!-- Node45 -->
-<g id="node46" class="node">
-<title>Node45</title>
+<!-- Node43 -->
+<g id="node44" class="node">
+<title>Node43</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1626.5,-364 1626.5,-383 1667.5,-383 1667.5,-364 1626.5,-364"/>
<text text-anchor="middle" x="1647" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tuple</text>
</g>
-<!-- Node41->Node45 -->
-<g id="edge119" class="edge">
-<title>Node41->Node45</title>
+<!-- Node39->Node43 -->
+<g id="edge117" class="edge">
+<title>Node39->Node43</title>
<path fill="none" stroke="#191970" d="M1602.1611,-425.2967C1611.0223,-415.0605 1622.7295,-401.5366 1631.9593,-390.8746"/>
<polygon fill="#191970" stroke="#191970" points="1634.8297,-392.9064 1638.7285,-383.055 1629.5373,-388.3248 1634.8297,-392.9064"/>
</g>
-<!-- Node42->Node18 -->
-<g id="edge98" class="edge">
-<title>Node42->Node18</title>
+<!-- Node40->Node16 -->
+<g id="edge96" class="edge">
+<title>Node40->Node16</title>
<path fill="none" stroke="#191970" d="M967.6292,-372.0921C852.3384,-368.679 613.327,-357.4328 539,-322 494.9458,-300.9987 446.117,-276.3751 472,-235 491.397,-203.993 518.7166,-223.3577 546,-199 581.0587,-167.7008 606.7271,-117.7405 618.9715,-90.5848"/>
<polygon fill="#191970" stroke="#191970" points="622.3309,-91.6378 623.1286,-81.073 615.9167,-88.8345 622.3309,-91.6378"/>
</g>
-<!-- Node42->Node22 -->
-<g id="edge96" class="edge">
-<title>Node42->Node22</title>
+<!-- Node40->Node20 -->
+<g id="edge94" class="edge">
+<title>Node40->Node20</title>
<path fill="none" stroke="#191970" d="M1025.1898,-358.4016C1012.3487,-325.0329 981.5717,-245.0561 967.3878,-208.1979"/>
<polygon fill="#191970" stroke="#191970" points="970.5637,-206.7053 963.7057,-198.6295 964.0307,-209.2194 970.5637,-206.7053"/>
</g>
-<!-- Node42->Node24 -->
-<g id="edge99" class="edge">
-<title>Node42->Node24</title>
+<!-- Node40->Node22 -->
+<g id="edge97" class="edge">
+<title>Node40->Node22</title>
<path fill="none" stroke="#191970" d="M1043.6041,-358.295C1061.2282,-337.311 1092.6792,-300.9345 1107,-291 1130.0435,-275.0145 1159.4787,-263.2594 1182.5295,-255.635"/>
<polygon fill="#191970" stroke="#191970" points="1183.8366,-258.8919 1192.302,-252.5212 1181.7114,-252.2223 1183.8366,-258.8919"/>
</g>
-<!-- Node42->Node31 -->
-<g id="edge100" class="edge">
-<title>Node42->Node31</title>
+<!-- Node40->Node29 -->
+<g id="edge98" class="edge">
+<title>Node40->Node29</title>
<path fill="none" stroke="#191970" d="M967.9998,-359.0445C922.2169,-348.5395 860.1285,-334.2932 812.1514,-323.2847"/>
<polygon fill="#191970" stroke="#191970" points="812.6682,-319.8124 802.1387,-320.9873 811.1027,-326.6351 812.6682,-319.8124"/>
</g>
-<!-- Node42->Node36 -->
-<g id="edge97" class="edge">
-<title>Node42->Node36</title>
+<!-- Node40->Node34 -->
+<g id="edge95" class="edge">
+<title>Node40->Node34</title>
<path fill="none" stroke="#191970" d="M1037.8811,-358.2548C1048.4832,-334.7656 1068.796,-289.7623 1080.3684,-264.1235"/>
<polygon fill="#191970" stroke="#191970" points="1083.6817,-265.2903 1084.6056,-254.7358 1077.3015,-262.4105 1083.6817,-265.2903"/>
</g>
-<!-- Node43->Node8 -->
-<g id="edge107" class="edge">
-<title>Node43->Node8</title>
+<!-- Node41->Node8 -->
+<g id="edge105" class="edge">
+<title>Node41->Node8</title>
<path fill="none" stroke="#191970" d="M1688.7265,-296.711C1682.6752,-286.6834 1673.8794,-270.4216 1670,-255 1661.1604,-219.8605 1661.896,-177.2564 1663.361,-152.8216"/>
<polygon fill="#191970" stroke="#191970" points="1666.8639,-152.9029 1664.0769,-142.6812 1659.8813,-152.4099 1666.8639,-152.9029"/>
</g>
-<!-- Node43->Node9 -->
-<g id="edge105" class="edge">
-<title>Node43->Node9</title>
+<!-- Node41->Node9 -->
+<g id="edge103" class="edge">
+<title>Node41->Node9</title>
<path fill="none" stroke="#191970" d="M1703.9391,-296.8688C1718.4038,-281.7851 1748.4192,-252.5078 1779,-235 1924.9385,-151.4488 2118.8598,-105.0385 2223.4734,-84.515"/>
<polygon fill="#191970" stroke="#191970" points="2224.1618,-87.9468 2233.3139,-82.6093 2222.8309,-81.0745 2224.1618,-87.9468"/>
</g>
-<!-- Node43->Node16 -->
-<g id="edge109" class="edge">
-<title>Node43->Node16</title>
+<!-- Node41->Node14 -->
+<g id="edge107" class="edge">
+<title>Node41->Node14</title>
<path fill="none" stroke="#191970" d="M1758.1336,-300.6732C1881.0917,-289.1638 2142.2678,-263.8622 2160,-255 2232.0009,-219.0155 2212.4762,-159.9238 2284,-123 2343.6042,-92.2296 2546.6195,-77.7846 2626.4145,-73.1954"/>
<polygon fill="#191970" stroke="#191970" points="2627.0397,-76.666 2636.8284,-72.6126 2626.6484,-69.677 2627.0397,-76.666"/>
</g>
-<!-- Node43->Node20 -->
-<g id="edge111" class="edge">
-<title>Node43->Node20</title>
+<!-- Node41->Node18 -->
+<g id="edge109" class="edge">
+<title>Node41->Node18</title>
<path fill="none" stroke="#191970" d="M1660.3934,-296.9125C1625.7439,-287.2678 1570.6518,-271.8107 1513.6803,-255.1629"/>
<polygon fill="#191970" stroke="#191970" points="1514.358,-251.7144 1503.7774,-252.2625 1512.3905,-258.4322 1514.358,-251.7144"/>
</g>
-<!-- Node43->Node23 -->
-<g id="edge108" class="edge">
-<title>Node43->Node23</title>
+<!-- Node41->Node21 -->
+<g id="edge106" class="edge">
+<title>Node41->Node21</title>
<path fill="none" stroke="#191970" d="M1672.5,-296.8906C1648.4062,-286.6005 1609.9787,-270.1888 1582.7992,-258.5809"/>
<polygon fill="#191970" stroke="#191970" points="1583.9131,-255.2508 1573.342,-254.5419 1581.1637,-261.6883 1583.9131,-255.2508"/>
</g>
-<!-- Node43->Node25 -->
-<g id="edge106" class="edge">
-<title>Node43->Node25</title>
+<!-- Node41->Node23 -->
+<g id="edge104" class="edge">
+<title>Node41->Node23</title>
<path fill="none" stroke="#191970" d="M1631.8031,-303.0863C1514.0586,-296.241 1265.5761,-279.3767 1183,-255 1164.3174,-249.4848 1162.7765,-240.1865 1144,-235 1084.5122,-218.5681 682.8095,-199.5412 513.7494,-192.156"/>
<polygon fill="#191970" stroke="#191970" points="513.8273,-188.6561 503.6846,-191.7181 513.523,-195.6495 513.8273,-188.6561"/>
</g>
-<!-- Node43->Node36 -->
-<g id="edge110" class="edge">
-<title>Node43->Node36</title>
+<!-- Node41->Node34 -->
+<g id="edge108" class="edge">
+<title>Node41->Node34</title>
<path fill="none" stroke="#191970" d="M1631.9043,-301.9445C1530.2227,-294.2872 1323.9056,-277.4446 1150,-255 1148.5365,-254.8111 1147.0543,-254.6137 1145.5595,-254.4092"/>
<polygon fill="#191970" stroke="#191970" points="1145.9754,-250.933 1135.5786,-252.9718 1144.9776,-257.8615 1145.9754,-250.933"/>
</g>
-<!-- Node43->Node38 -->
-<g id="edge104" class="edge">
-<title>Node43->Node38</title>
+<!-- Node41->Node36 -->
+<g id="edge102" class="edge">
+<title>Node41->Node36</title>
<path fill="none" stroke="#191970" d="M1714.375,-296.8906C1734.8495,-286.7359 1767.3442,-270.6196 1790.6885,-259.0416"/>
<polygon fill="#191970" stroke="#191970" points="1792.3575,-262.1207 1799.7611,-254.5419 1789.2472,-255.8496 1792.3575,-262.1207"/>
</g>
-<!-- Node43->Node41 -->
-<g id="edge112" class="edge">
-<title>Node43->Node41</title>
+<!-- Node41->Node39 -->
+<g id="edge110" class="edge">
+<title>Node41->Node39</title>
<path fill="none" stroke="#191970" d="M1680.8117,-316.046C1665.3608,-324.725 1641.6856,-339.0645 1627,-358 1613.8555,-374.9484 1606.387,-398.3152 1600.8203,-415.855"/>
<polygon fill="#191970" stroke="#191970" points="1597.4752,-414.8246 1597.734,-425.4163 1604.1367,-416.975 1597.4752,-414.8246"/>
</g>
-<!-- Node46->Node6 -->
-<g id="edge127" class="edge">
-<title>Node46->Node6</title>
-<path fill="none" stroke="#191970" d="M763.4654,-497.2753C639.355,-486.8196 377.5268,-457.0468 323,-389 314.3845,-378.2482 314.1827,-368.5869 323,-358 336.6792,-341.5754 459.7178,-323.833 538.0781,-314.0531"/>
+<!-- Node44->Node6 -->
+<g id="edge125" class="edge">
+<title>Node44->Node6</title>
+<path fill="none" stroke="#191970" d="M761.2419,-497.1931C637.423,-486.6289 377.2228,-456.7261 323,-389 314.389,-378.2446 314.1827,-368.5869 323,-358 336.6792,-341.5754 459.7178,-323.833 538.0781,-314.0531"/>
<polygon fill="#191970" stroke="#191970" points="538.853,-317.4841 548.3486,-312.7848 537.995,-310.5369 538.853,-317.4841"/>
</g>
-<!-- Node47 -->
-<g id="node48" class="node">
-<title>Node47</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="799,-431 799,-450 859,-450 859,-431 799,-431"/>
-<text text-anchor="middle" x="829" y="-438" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">iostream</text>
+<!-- Node45 -->
+<g id="node46" class="node">
+<title>Node45</title>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="797,-431 797,-450 857,-450 857,-431 797,-431"/>
+<text text-anchor="middle" x="827" y="-438" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">iostream</text>
</g>
-<!-- Node46->Node47 -->
-<g id="edge128" class="edge">
-<title>Node46->Node47</title>
-<path fill="none" stroke="#191970" d="M829,-492.3906C829,-483.8657 829,-471.1392 829,-460.4235"/>
-<polygon fill="#191970" stroke="#191970" points="832.5001,-460.2448 829,-450.2449 825.5001,-460.2449 832.5001,-460.2448"/>
+<!-- Node44->Node45 -->
+<g id="edge126" class="edge">
+<title>Node44->Node45</title>
+<path fill="none" stroke="#191970" d="M827,-492.3906C827,-483.8657 827,-471.1392 827,-460.4235"/>
+<polygon fill="#191970" stroke="#191970" points="830.5001,-460.2448 827,-450.2449 823.5001,-460.2449 830.5001,-460.2448"/>
</g>
-<!-- Node48->Node2 -->
-<g id="edge141" class="edge">
-<title>Node48->Node2</title>
+<!-- Node46->Node2 -->
+<g id="edge139" class="edge">
+<title>Node46->Node2</title>
<path fill="none" stroke="#191970" d="M1817.9223,-660.2455C1822.6027,-652.5367 1829.2673,-641.5598 1834.9289,-632.2348"/>
<polygon fill="#191970" stroke="#191970" points="1837.9475,-634.007 1840.1455,-623.6427 1831.9639,-630.3741 1837.9475,-634.007"/>
</g>
-<!-- Node48->Node3 -->
-<g id="edge142" class="edge">
-<title>Node48->Node3</title>
+<!-- Node46->Node3 -->
+<g id="edge140" class="edge">
+<title>Node46->Node3</title>
<path fill="none" stroke="#191970" d="M1808.7214,-660.1487C1805.6416,-650.8474 1800.9316,-636.5003 1797,-624 1792.1168,-608.4742 1786.7803,-590.8271 1782.8488,-577.6894"/>
<polygon fill="#191970" stroke="#191970" points="1786.1294,-576.4426 1779.9175,-567.8598 1779.4213,-578.443 1786.1294,-576.4426"/>
</g>
-<!-- Node48->Node8 -->
-<g id="edge145" class="edge">
-<title>Node48->Node8</title>
+<!-- Node46->Node8 -->
+<g id="edge143" class="edge">
+<title>Node46->Node8</title>
<path fill="none" stroke="#191970" d="M1852.2442,-669.702C2036.6547,-667.8162 2792,-654.2191 2792,-558 2792,-558 2792,-558 2792,-502 2792,-220.0509 2521.8812,-237.1798 2246,-179 2150.2314,-158.8036 1867.4065,-142.8622 1734.9355,-136.2786"/>
<polygon fill="#191970" stroke="#191970" points="1735.0545,-132.7803 1724.8942,-135.7832 1734.7095,-139.7718 1735.0545,-132.7803"/>
</g>
-<!-- Node48->Node16 -->
-<g id="edge146" class="edge">
-<title>Node48->Node16</title>
+<!-- Node46->Node14 -->
+<g id="edge144" class="edge">
+<title>Node46->Node14</title>
<path fill="none" stroke="#191970" d="M1852.1441,-669.8013C2000.9157,-668.7666 2520.4799,-662.4207 2681,-624 2751.4385,-607.1404 2830,-630.4281 2830,-558 2830,-558 2830,-558 2830,-189 2830,-122.2473 2740.3907,-90.5655 2690.9084,-78.1594"/>
<polygon fill="#191970" stroke="#191970" points="2691.5442,-74.713 2681.0052,-75.799 2689.9212,-81.5223 2691.5442,-74.713"/>
</g>
-<!-- Node48->Node21 -->
-<g id="edge143" class="edge">
-<title>Node48->Node21</title>
+<!-- Node46->Node19 -->
+<g id="edge141" class="edge">
+<title>Node46->Node19</title>
<path fill="none" stroke="#191970" d="M1795.3942,-660.4641C1711.3532,-612.2036 1332.297,-394.5308 1214.2254,-326.7282"/>
<polygon fill="#191970" stroke="#191970" points="1215.6329,-323.5004 1205.2181,-321.5557 1212.147,-329.5708 1215.6329,-323.5004"/>
</g>
-<!-- Node48->Node28 -->
-<g id="edge144" class="edge">
-<title>Node48->Node28</title>
+<!-- Node46->Node26 -->
+<g id="edge142" class="edge">
+<title>Node46->Node26</title>
<path fill="none" stroke="#191970" d="M1852.003,-663.8935C1940.2116,-648.0286 2150.4446,-596.4921 2237,-456 2262.1679,-415.1488 2244.8855,-355.6468 2233.2254,-325.511"/>
<polygon fill="#191970" stroke="#191970" points="2236.4282,-324.0957 2229.4143,-316.1549 2229.9454,-326.7365 2236.4282,-324.0957"/>
</g>
diff --git a/docs/reference/api/doxygen/algorithm_8h__incl.svg b/docs/reference/api/doxygen/algorithm_8h__incl.svg
index f459b2aad..0440c263e 100644
--- a/docs/reference/api/doxygen/algorithm_8h__incl.svg
+++ b/docs/reference/api/doxygen/algorithm_8h__incl.svg
@@ -31,45 +31,45 @@
<path fill="none" stroke="#191970" d="M3612.2454,-1007.42C3623.4215,-985.5326 3642,-943.3768 3642,-905 3642,-905 3642,-905 3642,-849 3642,-733.9816 3530.5639,-747.6996 3420,-716 3352.2266,-696.5688 2869.5667,-677.5763 2710.6163,-671.7871"/>
<polygon fill="#191970" stroke="#191970" points="2710.3016,-668.2735 2700.1814,-671.4091 2710.0481,-675.2689 2710.3016,-668.2735"/>
</g>
-<!-- Node18 -->
+<!-- Node16 -->
<g id="node13" class="node">
-<title>Node18</title>
+<title>Node16</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="3257,-6 3257,-25 3301,-25 3301,-6 3257,-6"/>
<text text-anchor="middle" x="3279" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
</g>
-<!-- Node0->Node18 -->
+<!-- Node0->Node16 -->
<g id="edge224" class="edge">
-<title>Node0->Node18</title>
+<title>Node0->Node16</title>
<path fill="none" stroke="#191970" d="M3668.0464,-1010.9689C3756.6958,-993.1537 3906,-955.6333 3906,-905 3906,-905 3906,-905 3906,-133 3906,-95.6561 3886.8365,-84.7858 3854,-67 3806.1369,-41.075 3425.4349,-22.0636 3311.2572,-16.9012"/>
<polygon fill="#191970" stroke="#191970" points="3311.2063,-13.3955 3301.0599,-16.4451 3310.8934,-20.3885 3311.2063,-13.3955"/>
</g>
-<!-- Node51 -->
+<!-- Node49 -->
<g id="node35" class="node">
-<title>Node51</title>
+<title>Node49</title>
<g id="a_node35"><a xlink:href="relay_2base_8h.html" target="_top" xlink:title="Base classes for the Relay IR. ">
<polygon fill="#ffffff" stroke="#000000" points="3286.5,-839.5 3286.5,-858.5 3385.5,-858.5 3385.5,-839.5 3286.5,-839.5"/>
<text text-anchor="middle" x="3336" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/relay/base.h</text>
</a>
</g>
</g>
-<!-- Node0->Node51 -->
+<!-- Node0->Node49 -->
<g id="edge128" class="edge">
-<title>Node0->Node51</title>
+<title>Node0->Node49</title>
<path fill="none" stroke="#191970" d="M3580.6856,-1007.4065C3529.9733,-974.576 3410.1311,-896.9916 3359.3691,-864.1289"/>
<polygon fill="#191970" stroke="#191970" points="3361.0234,-861.0305 3350.7269,-858.534 3357.2193,-866.9066 3361.0234,-861.0305"/>
</g>
-<!-- Node55 -->
+<!-- Node53 -->
<g id="node39" class="node">
-<title>Node55</title>
+<title>Node53</title>
<g id="a_node39"><a xlink:href="relay_2expr_8h.html" target="_top" xlink:title="Relay expression language. ">
<polygon fill="#ffffff" stroke="#000000" points="3145.5,-951.5 3145.5,-970.5 3242.5,-970.5 3242.5,-951.5 3145.5,-951.5"/>
<text text-anchor="middle" x="3194" y="-958.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/relay/expr.h</text>
</a>
</g>
</g>
-<!-- Node0->Node55 -->
+<!-- Node0->Node53 -->
<g id="edge159" class="edge">
-<title>Node0->Node55</title>
+<title>Node0->Node53</title>
<path fill="none" stroke="#191970" d="M3539.9375,-1012.8906C3461.1568,-1001.0735 3328.5536,-981.183 3252.4414,-969.7662"/>
<polygon fill="#191970" stroke="#191970" points="3252.9497,-966.3034 3242.5412,-968.2812 3251.9113,-973.2259 3252.9497,-966.3034"/>
</g>
@@ -116,51 +116,51 @@
<path fill="none" stroke="#191970" d="M2619.826,-661.0662C2577.9981,-651.3778 2515.621,-635.7285 2494,-624 2449.1635,-599.6779 2447.6825,-580.0495 2408,-548 2330.8461,-485.6868 2308.7334,-468.399 2215,-436 2040.3876,-375.6451 1986.8078,-387.1701 1807.5532,-345.2937"/>
<polygon fill="#191970" stroke="#191970" points="1808.3457,-341.8846 1797.8091,-342.9938 1806.7376,-348.6974 1808.3457,-341.8846"/>
</g>
-<!-- Node1->Node18 -->
+<!-- Node1->Node16 -->
<g id="edge123" class="edge">
-<title>Node1->Node18</title>
+<title>Node1->Node16</title>
<path fill="none" stroke="#191970" d="M2696.283,-660.4016C2745.2289,-647.4753 2835.1403,-623.8074 2912,-604 2975.0477,-587.7521 2995.1398,-595.8296 3054,-568 3165.2217,-515.4135 3195.4027,-494.6471 3274,-400 3302.213,-366.0258 3297.7024,-349.8761 3322,-313 3333.1525,-296.0741 3337.6886,-293.1026 3350,-277 3377.8584,-240.5629 3412,-240.3667 3412,-194.5 3412,-194.5 3412,-194.5 3412,-133 3412,-79.1753 3350.0653,-44.0942 3310.7248,-27.2401"/>
<polygon fill="#191970" stroke="#191970" points="3311.8409,-23.9147 3301.2612,-23.3475 3309.178,-30.3884 3311.8409,-23.9147"/>
</g>
-<!-- Node19 -->
+<!-- Node17 -->
<g id="node14" class="node">
-<title>Node19</title>
+<title>Node17</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1078.5,-6 1078.5,-25 1147.5,-25 1147.5,-6 1078.5,-6"/>
<text text-anchor="middle" x="1113" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
</g>
-<!-- Node1->Node19 -->
+<!-- Node1->Node17 -->
<g id="edge124" class="edge">
-<title>Node1->Node19</title>
+<title>Node1->Node17</title>
<path fill="none" stroke="#191970" d="M2619.7443,-662.3281C2516.5303,-642.7126 2233.9777,-589.3615 1998,-548 1753.9814,-505.2291 1058.9705,-535.663 902,-344 823.5082,-248.1605 833.5673,-140.8872 933,-67 972.8755,-37.369 1029.2985,-24.7622 1068.2961,-19.4136"/>
<polygon fill="#191970" stroke="#191970" points="1068.8276,-22.8744 1078.3091,-18.1466 1067.9488,-15.9297 1068.8276,-22.8744"/>
</g>
-<!-- Node20 -->
+<!-- Node18 -->
<g id="node15" class="node">
-<title>Node20</title>
+<title>Node18</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="435.5,-6 435.5,-25 480.5,-25 480.5,-6 435.5,-6"/>
<text text-anchor="middle" x="458" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
</g>
-<!-- Node1->Node20 -->
+<!-- Node1->Node18 -->
<g id="edge126" class="edge">
-<title>Node1->Node20</title>
+<title>Node1->Node18</title>
<path fill="none" stroke="#191970" d="M2619.633,-667.9346C2499.0645,-661.7311 2131.0491,-642.5442 1826,-624 1460.3741,-601.7733 218,-627.8008 218,-261.5 218,-261.5 218,-261.5 218,-133 218,-95.6561 238.8749,-87.6348 270,-67 318.0568,-35.1401 385.4199,-22.8767 425.1044,-18.2374"/>
<polygon fill="#191970" stroke="#191970" points="425.7163,-21.6919 435.2855,-17.1444 424.969,-14.7319 425.7163,-21.6919"/>
</g>
-<!-- Node22 -->
+<!-- Node20 -->
<g id="node16" class="node">
-<title>Node22</title>
+<title>Node20</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="2704.5,-123.5 2704.5,-142.5 2751.5,-142.5 2751.5,-123.5 2704.5,-123.5"/>
<text text-anchor="middle" x="2728" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
</g>
-<!-- Node1->Node22 -->
+<!-- Node1->Node20 -->
<g id="edge127" class="edge">
-<title>Node1->Node22</title>
+<title>Node1->Node20</title>
<path fill="none" stroke="#191970" d="M2700.2614,-662.7875C2728.3378,-656.2402 2765.5401,-644.3949 2793,-624 2867.3672,-568.7663 3000.5781,-366.625 3024,-277 3053.6876,-163.3988 2845.054,-139.3903 2761.5232,-134.3384"/>
<polygon fill="#191970" stroke="#191970" points="2761.6974,-130.8427 2751.5204,-133.7886 2761.3133,-137.8322 2761.6974,-130.8427"/>
</g>
-<!-- Node31 -->
+<!-- Node29 -->
<g id="node21" class="node">
-<title>Node31</title>
+<title>Node29</title>
<g id="a_node21"><a xlink:href="structural__hash_8h.html" target="_top" xlink:title="tvm/node/structural\l_hash.h">
<polygon fill="#ffffff" stroke="#000000" points="2037.5,-313.5 2037.5,-343.5 2150.5,-343.5 2150.5,-313.5 2037.5,-313.5"/>
<text text-anchor="start" x="2045.5" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
@@ -168,39 +168,39 @@
</a>
</g>
</g>
-<!-- Node1->Node31 -->
+<!-- Node1->Node29 -->
<g id="edge120" class="edge">
-<title>Node1->Node31</title>
+<title>Node1->Node29</title>
<path fill="none" stroke="#191970" d="M2656.8482,-660.171C2644.7723,-624.1372 2597.8708,-498.1478 2514,-436 2464.4528,-399.2858 2438.0432,-420.8292 2380,-400 2359.7815,-392.7444 2356.4029,-386.7196 2336,-380 2277.9825,-360.8924 2209.749,-347.1218 2160.5508,-338.6548"/>
<polygon fill="#191970" stroke="#191970" points="2161.1072,-335.1993 2150.6631,-336.9804 2159.9383,-342.101 2161.1072,-335.1993"/>
</g>
-<!-- Node38 -->
+<!-- Node36 -->
<g id="node25" class="node">
-<title>Node38</title>
+<title>Node36</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1310.5,-123.5 1310.5,-142.5 1403.5,-142.5 1403.5,-123.5 1310.5,-123.5"/>
<text text-anchor="middle" x="1357" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">unordered_map</text>
</g>
-<!-- Node1->Node38 -->
+<!-- Node1->Node36 -->
<g id="edge125" class="edge">
-<title>Node1->Node38</title>
+<title>Node1->Node36</title>
<path fill="none" stroke="#191970" d="M2643.0889,-660.4269C2594.9069,-633.2384 2452.6158,-553.5433 2332,-492 2280.6489,-465.7985 2269.9439,-453.4535 2215,-436 2211.3745,-434.8483 1678.5365,-345.4014 1675,-344 1549.1683,-294.1368 1421.6632,-189.4922 1375.2833,-149.2484"/>
<polygon fill="#191970" stroke="#191970" points="1377.4159,-146.4637 1367.5859,-142.5111 1372.8055,-151.731 1377.4159,-146.4637"/>
</g>
-<!-- Node42 -->
+<!-- Node40 -->
<g id="node26" class="node">
-<title>Node42</title>
+<title>Node40</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1726.5,-185 1726.5,-204 1791.5,-204 1791.5,-185 1726.5,-185"/>
<text text-anchor="middle" x="1759" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">functional</text>
</g>
-<!-- Node1->Node42 -->
+<!-- Node1->Node40 -->
<g id="edge122" class="edge">
-<title>Node1->Node42</title>
+<title>Node1->Node40</title>
<path fill="none" stroke="#191970" d="M2644.748,-660.3753C2601.2798,-633.0509 2472.8174,-553.0329 2363,-492 2315.3942,-465.5423 2304.2996,-456.8871 2254,-436 2227.4711,-424.9838 1779.3219,-299.9148 1762,-277 1748.409,-259.0207 1750.3831,-232 1753.9283,-213.8298"/>
<polygon fill="#191970" stroke="#191970" points="1757.3465,-214.5819 1756.1523,-204.0546 1750.521,-213.0289 1757.3465,-214.5819"/>
</g>
-<!-- Node43 -->
+<!-- Node41 -->
<g id="node27" class="node">
-<title>Node43</title>
+<title>Node41</title>
<g id="a_node27"><a xlink:href="packed__func_8h.html" target="_top" xlink:title="Type-erased function used across TVM API. ">
<polygon fill="#ffffff" stroke="#000000" points="2290,-313.5 2290,-343.5 2406,-343.5 2406,-313.5 2290,-313.5"/>
<text text-anchor="start" x="2298" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/packed</text>
@@ -208,9 +208,9 @@
</a>
</g>
</g>
-<!-- Node1->Node43 -->
+<!-- Node1->Node41 -->
<g id="edge121" class="edge">
-<title>Node1->Node43</title>
+<title>Node1->Node41</title>
<path fill="none" stroke="#191970" d="M2659.2037,-660.1611C2657.1802,-634.2803 2652,-562.1514 2652,-502 2652,-502 2652,-502 2652,-446 2652,-395.7565 2505.7004,-358.6762 2416.0834,-340.6756"/>
<polygon fill="#191970" stroke="#191970" points="2416.7436,-337.2384 2406.2546,-338.7319 2415.3856,-344.1054 2416.7436,-337.2384"/>
</g>
@@ -259,33 +259,33 @@
<path fill="none" stroke="#191970" d="M1835.2036,-609.8891C1715.3055,-597.1289 1359.2157,-556.3823 1252,-512 1143.1589,-466.9448 1115.2815,-439.9895 1047,-344 1020.6708,-306.9867 995.7521,-285.601 1018,-246 1032.019,-221.0465 1102.6875,-192.3598 1128,-179 1181.4789,-150.7741 1194.6363,-142.1337 1252,-123 1297.3453,-107.8751 1349.9656,-96.2204 1390.6732,-88.4266"/>
<polygon fill="#191970" stroke="#191970" points="1391.5746,-91.8184 1400.7543,-86.5285 1390.2793,-84.9393 1391.5746,-91.8184"/>
</g>
-<!-- Node3->Node18 -->
+<!-- Node3->Node16 -->
<g id="edge117" class="edge">
-<title>Node3->Node18</title>
+<title>Node3->Node16</title>
<path fill="none" stroke="#191970" d="M1914.5302,-609.4231C1932.3711,-607.4975 1953.7243,-605.3866 1973,-604 2068.1713,-597.1539 2743.704,-606.5208 2831,-568 2906.6187,-534.6321 2899.1144,-489.6378 2962,-436 3015.3031,-390.5355 3034.4029,-386.6288 3090,-344 3127.348,-315.3636 3136.5902,-308.001 3172,-277 3212.3365,-241.6858 3260,-248.1109 3260,-194.5 3260,-194.5 3260,-194.5 3260,-133 3260,-98.1936 3268.2253,-58.3226 3273.8521,-35.1197"/>
<polygon fill="#191970" stroke="#191970" points="3277.3113,-35.7152 3276.3581,-25.1633 3270.523,-34.0065 3277.3113,-35.7152"/>
</g>
-<!-- Node3->Node19 -->
+<!-- Node3->Node17 -->
<g id="edge118" class="edge">
-<title>Node3->Node19</title>
+<title>Node3->Node17</title>
<path fill="none" stroke="#191970" d="M1835.0672,-613.2078C1695.1428,-610.1948 1227.5913,-598.0642 1081,-568 835.1199,-517.5728 722.2339,-494.5233 597,-277 562.5633,-217.1856 534.1227,-177.0851 577,-123 592.4482,-103.5137 941.933,-43.8694 1068.0936,-22.8918"/>
<polygon fill="#191970" stroke="#191970" points="1068.8997,-26.306 1078.1918,-21.216 1067.7537,-19.4004 1068.8997,-26.306"/>
</g>
-<!-- Node24 -->
+<!-- Node22 -->
<g id="node18" class="node">
-<title>Node24</title>
+<title>Node22</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="586,-123.5 586,-142.5 650,-142.5 650,-123.5 586,-123.5"/>
<text text-anchor="middle" x="618" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
</g>
-<!-- Node3->Node24 -->
+<!-- Node3->Node22 -->
<g id="edge115" class="edge">
-<title>Node3->Node24</title>
+<title>Node3->Node22</title>
<path fill="none" stroke="#191970" d="M1835.3157,-613.3752C1704.6147,-611.017 1288.6037,-601.0376 1159,-568 921.2932,-507.4056 839.6166,-471.3993 690,-277 659.1625,-236.9324 635.956,-181.591 624.9931,-152.5662"/>
<polygon fill="#191970" stroke="#191970" points="628.1425,-150.9903 621.4006,-142.8174 621.5743,-153.4108 628.1425,-150.9903"/>
</g>
-<!-- Node35 -->
+<!-- Node33 -->
<g id="node23" class="node">
-<title>Node35</title>
+<title>Node33</title>
<g id="a_node23"><a xlink:href="string_8h.html" target="_top" xlink:title="Runtime String container types. ">
<polygon fill="#ffffff" stroke="#ff0000" points="922,-179.5 922,-209.5 1048,-209.5 1048,-179.5 922,-179.5"/>
<text text-anchor="start" x="930" y="-197.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
@@ -293,36 +293,36 @@
</a>
</g>
</g>
-<!-- Node3->Node35 -->
+<!-- Node3->Node33 -->
<g id="edge113" class="edge">
-<title>Node3->Node35</title>
+<title>Node3->Node33</title>
<path fill="none" stroke="#191970" d="M1835.2196,-613.7299C1711.8116,-612.4964 1336.8773,-605.6644 1222,-568 1089.1321,-524.4371 1041.0172,-511.012 956,-400 913.1364,-344.0305 886.8463,-309.2402 918,-246 924.1481,-233.5198 934.9346,-223.219 946.0797,-215.1955"/>
<polygon fill="#191970" stroke="#191970" points="948.2285,-217.9693 954.6014,-209.5054 944.3413,-212.1477 948.2285,-217.9693"/>
</g>
-<!-- Node46 -->
+<!-- Node44 -->
<g id="node30" class="node">
-<title>Node46</title>
+<title>Node44</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="2922,-252 2922,-271 2966,-271 2966,-252 2922,-252"/>
<text text-anchor="middle" x="2944" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
</g>
-<!-- Node3->Node46 -->
+<!-- Node3->Node44 -->
<g id="edge116" class="edge">
-<title>Node3->Node46</title>
+<title>Node3->Node44</title>
<path fill="none" stroke="#191970" d="M1914.7808,-605.2019C2092.8696,-565.7867 2808.6034,-407.0679 2818,-400 2853.2921,-373.4541 2836.037,-345.4413 2866,-313 2879.9895,-297.8534 2899.3073,-285.0554 2915.2179,-276.006"/>
<polygon fill="#191970" stroke="#191970" points="2917.0404,-278.9983 2924.126,-271.1215 2913.6749,-272.8605 2917.0404,-278.9983"/>
</g>
-<!-- Node50 -->
+<!-- Node48 -->
<g id="node34" class="node">
-<title>Node50</title>
+<title>Node48</title>
<g id="a_node34"><a xlink:href="ir_2type_8h.html" target="_top" xlink:title="IR/AST nodes for the unified type system in TVM. ">
<polygon fill="#ffffff" stroke="#000000" points="1909,-548.5 1909,-567.5 1989,-567.5 1989,-548.5 1909,-548.5"/>
<text text-anchor="middle" x="1949" y="-555.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
</a>
</g>
</g>
-<!-- Node3->Node50 -->
+<!-- Node3->Node48 -->
<g id="edge105" class="edge">
-<title>Node3->Node50</title>
+<title>Node3->Node48</title>
<path fill="none" stroke="#191970" d="M1887.8898,-604.2455C1898.9254,-595.8943 1915.0292,-583.7076 1927.9353,-573.9408"/>
<polygon fill="#191970" stroke="#191970" points="1930.3959,-576.4681 1936.2579,-567.6427 1926.1717,-570.8862 1930.3959,-576.4681"/>
</g>
@@ -338,9 +338,9 @@
<path fill="none" stroke="#191970" d="M2056.3056,-500.983C1931.2966,-497.0282 1556.0575,-478.7044 1469,-400 1381.2173,-320.64 1430.6091,-152.8097 1450.7329,-96.1475"/>
<polygon fill="#191970" stroke="#191970" points="1454.0656,-97.2245 1454.2144,-86.6307 1447.4917,-94.8195 1454.0656,-97.2245"/>
</g>
-<!-- Node4->Node18 -->
+<!-- Node4->Node16 -->
<g id="edge104" class="edge">
-<title>Node4->Node18</title>
+<title>Node4->Node16</title>
<path fill="none" stroke="#191970" d="M2137.6989,-494.6031C2208.5657,-481.7229 2347.3878,-456.4897 2350,-456 2479.0226,-431.8141 2519.8434,-452.86 2640,-400 2867.4144,-299.9545 2865.4387,-175.3825 3089,-67 3141.2849,-41.6523 3207.9239,-27.258 3246.7905,-20.4482"/>
<polygon fill="#191970" stroke="#191970" points="3247.7045,-23.8434 3256.9814,-18.7258 3246.5379,-16.9413 3247.7045,-23.8434"/>
</g>
@@ -387,63 +387,63 @@
<path fill="none" stroke="#191970" d="M2079.5841,-444.1312C2193.2643,-439.453 2469.756,-425.727 2559,-400 2648.2634,-374.2674 2683.7561,-358.686 2728,-277 2761.3374,-215.4503 2802.2959,-179.5191 2761,-123 2729.7893,-80.284 2587.9462,-46.3731 2498.5214,-28.7635"/>
<polygon fill="#191970" stroke="#191970" points="2499.1784,-25.3257 2488.6941,-26.8522 2497.842,-32.197 2499.1784,-25.3257"/>
</g>
-<!-- Node5->Node18 -->
+<!-- Node5->Node16 -->
<g id="edge99" class="edge">
-<title>Node5->Node18</title>
+<title>Node5->Node16</title>
<path fill="none" stroke="#191970" d="M2079.6866,-444.6552C2200.2502,-440.9686 2505.5472,-428.98 2603,-400 2652.9447,-385.1477 2661.2349,-370.6673 2706,-344 2900.0216,-228.4185 2929.3121,-166.6046 3132,-67 3170.1582,-48.2483 3216.8939,-33.1041 3247.2778,-24.2094"/>
<polygon fill="#191970" stroke="#191970" points="3248.2761,-27.5642 3256.919,-21.4363 3246.3411,-20.8369 3248.2761,-27.5642"/>
</g>
-<!-- Node5->Node19 -->
+<!-- Node5->Node17 -->
<g id="edge100" class="edge">
-<title>Node5->Node19</title>
+<title>Node5->Node17</title>
<path fill="none" stroke="#191970" d="M1980.4053,-445.5405C1853.6861,-443.8486 1520.7194,-436.0052 1417,-400 1351.6708,-377.3216 1346.7114,-349.5404 1288,-313 1237.1258,-281.3373 1227.7019,-265.3808 1171,-246 1116.2226,-227.277 951,-253.6705 913,-210 880.6055,-172.7714 892.9038,-127.0412 956,-67 986.4147,-38.0579 1033.3692,-25.3759 1068.01,-19.8216"/>
<polygon fill="#191970" stroke="#191970" points="1068.9429,-23.2227 1078.3333,-18.3166 1067.933,-16.2959 1068.9429,-23.2227"/>
</g>
-<!-- Node5->Node20 -->
+<!-- Node5->Node18 -->
<g id="edge101" class="edge">
-<title>Node5->Node20</title>
+<title>Node5->Node18</title>
<path fill="none" stroke="#191970" d="M1980.1182,-444.005C1847.4052,-438.4598 1487.8249,-421.8221 1371,-400 1089.7836,-347.4706 1004.9212,-350.9973 756,-210 705.6111,-181.4581 707.609,-154.4777 659,-123 625.9325,-101.5865 612.9837,-105.1146 578,-87 543.5889,-69.1819 505.3835,-45.751 481.5268,-30.6558"/>
<polygon fill="#191970" stroke="#191970" points="483.1426,-27.5355 472.8273,-25.1178 479.3835,-33.4405 483.1426,-27.5355"/>
</g>
-<!-- Node5->Node22 -->
+<!-- Node5->Node20 -->
<g id="edge102" class="edge">
-<title>Node5->Node22</title>
+<title>Node5->Node20</title>
<path fill="none" stroke="#191970" d="M2079.6312,-444.4005C2187.2681,-440.4361 2438.9199,-428.3519 2519,-400 2612.0755,-367.0471 2641.1545,-353.1537 2704,-277 2732.8323,-242.0622 2733.9639,-224.2559 2732,-179 2731.6273,-170.4107 2730.8648,-160.9445 2730.1045,-152.8439"/>
<polygon fill="#191970" stroke="#191970" points="2733.5834,-152.4567 2729.106,-142.8544 2726.6181,-153.1529 2733.5834,-152.4567"/>
</g>
-<!-- Node5->Node31 -->
+<!-- Node5->Node29 -->
<g id="edge95" class="edge">
-<title>Node5->Node31</title>
+<title>Node5->Node29</title>
<path fill="none" stroke="#191970" d="M2040.4849,-436.3056C2049.4064,-427.5607 2062.0098,-413.9617 2070,-400 2078.2404,-385.6009 2084.203,-367.9404 2088.1238,-353.7741"/>
<polygon fill="#191970" stroke="#191970" points="2091.6051,-354.2924 2090.7281,-343.7339 2084.8294,-352.5348 2091.6051,-354.2924"/>
</g>
-<!-- Node27 -->
+<!-- Node25 -->
<g id="node24" class="node">
-<title>Node27</title>
+<title>Node25</title>
<g id="a_node24"><a xlink:href="runtime_2memory_8h.html" target="_top" xlink:title="Runtime memory management. ">
<polygon fill="#ffffff" stroke="#ff0000" points="1048.5,-123.5 1048.5,-142.5 1177.5,-142.5 1177.5,-123.5 1048.5,-123.5"/>
<text text-anchor="middle" x="1113" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
</a>
</g>
</g>
-<!-- Node5->Node27 -->
+<!-- Node5->Node25 -->
<g id="edge97" class="edge">
-<title>Node5->Node27</title>
+<title>Node5->Node25</title>
<path fill="none" stroke="#191970" d="M1980.0751,-444.0312C1848.2094,-438.5514 1497.4585,-422.0627 1450,-400 1400.6976,-377.0801 1405.3501,-348.3018 1364,-313 1285.2119,-245.7363 1182.395,-177.4843 1136.6855,-148.0421"/>
<polygon fill="#191970" stroke="#191970" points="1138.3853,-144.9743 1128.078,-142.5226 1134.6067,-150.8669 1138.3853,-144.9743"/>
</g>
-<!-- Node48 -->
+<!-- Node46 -->
<g id="node32" class="node">
-<title>Node48</title>
+<title>Node46</title>
<g id="a_node32"><a xlink:href="repr__printer_8h.html" target="_top" xlink:title="Printer class to print repr string of each AST/IR nodes. ">
<polygon fill="#ffffff" stroke="#000000" points="1494.5,-380.5 1494.5,-399.5 1625.5,-399.5 1625.5,-380.5 1494.5,-380.5"/>
<text text-anchor="middle" x="1560" y="-387.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/repr_printer.h</text>
</a>
</g>
</g>
-<!-- Node5->Node48 -->
+<!-- Node5->Node46 -->
<g id="edge91" class="edge">
-<title>Node5->Node48</title>
+<title>Node5->Node46</title>
<path fill="none" stroke="#191970" d="M1980.1467,-440.06C1897.7754,-430.2456 1732.7548,-410.5836 1635.848,-399.0372"/>
<polygon fill="#191970" stroke="#191970" points="1636.1661,-395.5504 1625.8222,-397.8426 1635.3378,-402.5013 1636.1661,-395.5504"/>
</g>
@@ -465,69 +465,69 @@
<path fill="none" stroke="#191970" d="M2453.2152,-380.2777C2467.802,-345.3888 2514.2129,-223.2459 2495,-123 2490.0253,-97.0439 2489.0781,-89.3668 2475,-67 2468.2604,-56.2924 2458.9476,-46.0514 2450.1226,-37.5623"/>
<polygon fill="#191970" stroke="#191970" points="2452.2902,-34.7999 2442.5697,-30.5854 2447.5404,-39.9418 2452.2902,-34.7999"/>
</g>
-<!-- Node6->Node18 -->
+<!-- Node6->Node16 -->
<g id="edge88" class="edge">
-<title>Node6->Node18</title>
+<title>Node6->Node16</title>
<path fill="none" stroke="#191970" d="M2485.4455,-380.4574C2514.4029,-372.3215 2555.608,-359.4968 2590,-344 2806.1949,-246.584 2826.1496,-155.859 3046,-67 3114.5943,-39.2756 3200.8767,-25.1698 3246.786,-19.1741"/>
<polygon fill="#191970" stroke="#191970" points="3247.2681,-22.6411 3256.7519,-17.918 3246.3927,-15.696 3247.2681,-22.6411"/>
</g>
-<!-- Node6->Node19 -->
+<!-- Node6->Node17 -->
<g id="edge89" class="edge">
-<title>Node6->Node19</title>
+<title>Node6->Node17</title>
<path fill="none" stroke="#191970" d="M2388.0905,-389.3248C2261.4519,-387.2295 1973.9684,-378.6821 1883,-344 1861.6541,-335.8618 1862.2577,-323.5594 1842,-313 1720.34,-249.5842 1638.7285,-315.4818 1551,-210 1509.7437,-160.3948 1572.7265,-111.4904 1526,-67 1499.6553,-41.9161 1259.2075,-24.4286 1157.7913,-18.1077"/>
<polygon fill="#191970" stroke="#191970" points="1157.7581,-14.5991 1147.5622,-17.4787 1157.3285,-21.5859 1157.7581,-14.5991"/>
</g>
-<!-- Node6->Node22 -->
+<!-- Node6->Node20 -->
<g id="edge90" class="edge">
-<title>Node6->Node22</title>
+<title>Node6->Node20</title>
<path fill="none" stroke="#191970" d="M2473.228,-380.4167C2494.425,-371.743 2525.8296,-358.1861 2552,-344 2574.3264,-331.8976 2577.6907,-325.134 2600,-313 2634.2513,-294.3708 2653.5387,-305.6356 2680,-277 2712.6688,-241.6469 2723.1553,-183.4306 2726.4824,-152.9775"/>
<polygon fill="#191970" stroke="#191970" points="2729.9905,-153.0538 2727.4384,-142.771 2723.021,-152.401 2729.9905,-153.0538"/>
</g>
-<!-- Node30 -->
+<!-- Node28 -->
<g id="node20" class="node">
-<title>Node30</title>
+<title>Node28</title>
<g id="a_node20"><a xlink:href="data__type_8h.html" target="_top" xlink:title="tvm/runtime/data_type.h">
<polygon fill="#ffffff" stroke="#000000" points="2334,-185 2334,-204 2472,-204 2472,-185 2334,-185"/>
<text text-anchor="middle" x="2403" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/data_type.h</text>
</a>
</g>
</g>
-<!-- Node6->Node30 -->
+<!-- Node6->Node28 -->
<g id="edge56" class="edge">
-<title>Node6->Node30</title>
+<title>Node6->Node28</title>
<path fill="none" stroke="#191970" d="M2448.8834,-380.3138C2448.3532,-356.7517 2445.4711,-294.796 2430,-246 2426.3387,-234.4522 2420.1773,-222.4146 2414.664,-212.8837"/>
<polygon fill="#191970" stroke="#191970" points="2417.647,-211.0526 2409.4835,-204.2993 2411.6538,-214.6695 2417.647,-211.0526"/>
</g>
-<!-- Node6->Node31 -->
+<!-- Node6->Node29 -->
<g id="edge29" class="edge">
-<title>Node6->Node31</title>
+<title>Node6->Node29</title>
<path fill="none" stroke="#191970" d="M2393.9208,-380.4581C2331.031,-369.5631 2228.1649,-351.7427 2160.7665,-340.0666"/>
<polygon fill="#191970" stroke="#191970" points="2161.1282,-336.5772 2150.6775,-338.3188 2159.9333,-343.4745 2161.1282,-336.5772"/>
</g>
-<!-- Node32 -->
+<!-- Node30 -->
<g id="node22" class="node">
-<title>Node32</title>
+<title>Node30</title>
<g id="a_node22"><a xlink:href="ndarray_8h.html" target="_top" xlink:title="A device-independent managed NDArray abstraction. ">
<polygon fill="#ffffff" stroke="#ff0000" points="1864.5,-252 1864.5,-271 1989.5,-271 1989.5,-252 1864.5,-252"/>
<text text-anchor="middle" x="1927" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/ndarray.h</text>
</a>
</g>
</g>
-<!-- Node6->Node32 -->
+<!-- Node6->Node30 -->
<g id="edge58" class="edge">
-<title>Node6->Node32</title>
+<title>Node6->Node30</title>
<path fill="none" stroke="#191970" d="M2413.9521,-380.4601C2348.2213,-362.6021 2208.6828,-324.8598 2160,-313 2099.814,-298.3378 2030.4448,-283.2338 1983.1536,-273.2063"/>
<polygon fill="#191970" stroke="#191970" points="1983.6203,-269.7276 1973.1125,-271.0831 1982.1721,-276.5762 1983.6203,-269.7276"/>
</g>
-<!-- Node6->Node27 -->
+<!-- Node6->Node25 -->
<g id="edge57" class="edge">
-<title>Node6->Node27</title>
+<title>Node6->Node25</title>
<path fill="none" stroke="#191970" d="M2388.376,-385.7697C2238.6604,-375.2666 1862.5673,-348.5193 1850,-344 1827.4896,-335.9051 1828.2294,-321.8373 1806,-313 1701.6839,-271.529 1664.2851,-306.5975 1556,-277 1445.6535,-246.8391 1427.0915,-213.4249 1318,-179 1271.5651,-164.347 1217.8147,-152.4676 1177.0047,-144.4536"/>
<polygon fill="#191970" stroke="#191970" points="1177.6685,-141.0172 1167.1853,-142.5507 1176.3367,-147.8894 1177.6685,-141.0172"/>
</g>
-<!-- Node6->Node43 -->
+<!-- Node6->Node41 -->
<g id="edge60" class="edge">
-<title>Node6->Node43</title>
+<title>Node6->Node41</title>
<path fill="none" stroke="#191970" d="M2433.2188,-380.3906C2419.4231,-371.9903 2398.9274,-359.5103 2381.4934,-348.8945"/>
<polygon fill="#191970" stroke="#191970" points="2383.0689,-345.756 2372.7074,-343.5446 2379.4283,-351.7349 2383.0689,-345.756"/>
</g>
@@ -546,15 +546,15 @@
<path fill="none" stroke="#191970" d="M1685.0144,-313.4614C1665.5718,-305.5931 1645.3244,-293.9464 1632,-277 1617.8528,-259.0071 1614.6499,-232.1773 1614.2971,-214.0419"/>
<polygon fill="#191970" stroke="#191970" points="1617.797,-214.0567 1614.3733,-204.0303 1610.7972,-214.0033 1617.797,-214.0567"/>
</g>
-<!-- Node7->Node18 -->
+<!-- Node7->Node16 -->
<g id="edge28" class="edge">
-<title>Node7->Node18</title>
+<title>Node7->Node16</title>
<path fill="none" stroke="#191970" d="M1797.8644,-326.0784C1892.3015,-321.3123 2087.1187,-308.3157 2249,-277 2354.3713,-256.6161 2381.0314,-249.0499 2481,-210 2613.3738,-158.2919 2630.4968,-106.5402 2767,-67 2857.877,-40.6761 3148.8693,-22.6572 3246.6224,-17.2207"/>
<polygon fill="#191970" stroke="#191970" points="3246.9876,-20.706 3256.7805,-16.6626 3246.6035,-13.7166 3246.9876,-20.706"/>
</g>
-<!-- Node23 -->
+<!-- Node21 -->
<g id="node17" class="node">
-<title>Node23</title>
+<title>Node21</title>
<g id="a_node17"><a xlink:href="array_8h.html" target="_top" xlink:title="Runtime Array container types. ">
<polygon fill="#ffffff" stroke="#ff0000" points="1036,-246.5 1036,-276.5 1162,-276.5 1162,-246.5 1036,-246.5"/>
<text text-anchor="start" x="1044" y="-264.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
@@ -562,15 +562,15 @@
</a>
</g>
</g>
-<!-- Node7->Node23 -->
+<!-- Node7->Node21 -->
<g id="edge18" class="edge">
-<title>Node7->Node23</title>
+<title>Node7->Node21</title>
<path fill="none" stroke="#191970" d="M1684.3476,-322.2362C1658.3473,-319.382 1627.1134,-315.9822 1599,-313 1447.3005,-296.9082 1269.6292,-278.7788 1172.1948,-268.8982"/>
<polygon fill="#191970" stroke="#191970" points="1172.4586,-265.4071 1162.1565,-267.8807 1171.7525,-272.3714 1172.4586,-265.4071"/>
</g>
-<!-- Node7->Node30 -->
+<!-- Node7->Node28 -->
<g id="edge23" class="edge">
-<title>Node7->Node30</title>
+<title>Node7->Node28</title>
<path fill="none" stroke="#191970" d="M1797.5862,-318.0544C1849.782,-308.2075 1929.3447,-292.662 1998,-277 2052.1242,-264.6529 2064.7519,-257.7905 2119,-246 2207.6844,-226.7251 2230.7264,-226.3302 2320,-210 2327.1377,-208.6944 2334.6183,-207.3141 2342.0423,-205.9369"/>
<polygon fill="#191970" stroke="#191970" points="2342.9017,-209.3372 2352.0935,-204.0682 2341.6221,-202.4552 2342.9017,-209.3372"/>
</g>
@@ -580,21 +580,21 @@
<path fill="none" stroke="#191970" d="M1602.7605,-184.8889C1585.0673,-171.0414 1551.2554,-144.7509 1522,-123 1508.2155,-112.7514 1492.5608,-101.4931 1480.1846,-92.6795"/>
<polygon fill="#191970" stroke="#191970" points="1482.0056,-89.68 1471.8259,-86.7434 1477.9525,-95.3872 1482.0056,-89.68"/>
</g>
-<!-- Node8->Node19 -->
+<!-- Node8->Node17 -->
<g id="edge15" class="edge">
-<title>Node8->Node19</title>
+<title>Node8->Node17</title>
<path fill="none" stroke="#191970" d="M1634.8588,-184.9486C1658.9782,-171.9893 1694.6215,-147.616 1679,-123 1647.6801,-73.6466 1616.2199,-83.0005 1560,-67 1485.061,-45.6719 1255.4964,-26.3122 1157.7845,-18.8035"/>
<polygon fill="#191970" stroke="#191970" points="1157.8796,-15.3007 1147.6425,-18.0304 1157.3475,-22.2805 1157.8796,-15.3007"/>
</g>
-<!-- Node8->Node20 -->
+<!-- Node8->Node18 -->
<g id="edge16" class="edge">
-<title>Node8->Node20</title>
+<title>Node8->Node18</title>
<path fill="none" stroke="#191970" d="M1593.2568,-184.996C1556.6193,-169.4051 1479.923,-138.5766 1412,-123 1057.5768,-41.7212 614.3866,-20.9025 491.129,-16.5241"/>
<polygon fill="#191970" stroke="#191970" points="490.9686,-13.0167 480.8544,-16.1714 490.7284,-20.0125 490.9686,-13.0167"/>
</g>
-<!-- Node8->Node22 -->
+<!-- Node8->Node20 -->
<g id="edge17" class="edge">
-<title>Node8->Node22</title>
+<title>Node8->Node20</title>
<path fill="none" stroke="#191970" d="M1669.6551,-185.1095C1684.9365,-182.7799 1701.581,-180.5181 1717,-179 2099.2863,-141.3624 2565.9808,-134.4612 2694.0942,-133.2484"/>
<polygon fill="#191970" stroke="#191970" points="2694.4544,-136.7454 2704.4228,-133.1563 2694.3919,-129.7457 2694.4544,-136.7454"/>
</g>
@@ -604,315 +604,315 @@
<path fill="none" stroke="#191970" d="M1517.7949,-73.3874C1643.4136,-65.7727 1943.9593,-47.422 2196,-31 2247.0543,-27.6735 2304.5271,-23.7653 2348.9973,-20.7055"/>
<polygon fill="#191970" stroke="#191970" points="2349.4076,-24.1856 2359.1434,-20.0066 2348.9265,-17.2022 2349.4076,-24.1856"/>
</g>
-<!-- Node17 -->
+<!-- Node15 -->
<g id="node12" class="node">
-<title>Node17</title>
+<title>Node15</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="2061.5,-6 2061.5,-25 2186.5,-25 2186.5,-6 2061.5,-6"/>
<text text-anchor="middle" x="2124" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/logging.h</text>
</g>
-<!-- Node10->Node17 -->
+<!-- Node10->Node15 -->
<g id="edge11" class="edge">
-<title>Node10->Node17</title>
+<title>Node10->Node15</title>
<path fill="none" stroke="#191970" d="M1517.6892,-71.4882C1639.9795,-60.1956 1918.6157,-34.4657 2051.2384,-22.219"/>
<polygon fill="#191970" stroke="#191970" points="2051.6653,-25.6945 2061.301,-21.2898 2051.0215,-18.7242 2051.6653,-25.6945"/>
</g>
-<!-- Node10->Node18 -->
+<!-- Node10->Node16 -->
<g id="edge12" class="edge">
-<title>Node10->Node18</title>
+<title>Node10->Node16</title>
<path fill="none" stroke="#191970" d="M1517.5526,-74.9887C1803.5752,-65.329 3031.3086,-23.8652 3246.6846,-16.5914"/>
<polygon fill="#191970" stroke="#191970" points="3246.8413,-20.0882 3256.7175,-16.2525 3246.605,-13.0922 3246.8413,-20.0882"/>
</g>
-<!-- Node10->Node19 -->
+<!-- Node10->Node17 -->
<g id="edge13" class="edge">
-<title>Node10->Node19</title>
+<title>Node10->Node17</title>
<path fill="none" stroke="#191970" d="M1404.4723,-67.4581C1336.7026,-55.3774 1221.1665,-34.7818 1157.8201,-23.4897"/>
<polygon fill="#191970" stroke="#191970" points="1158.2021,-20.0027 1147.743,-21.6933 1156.9736,-26.894 1158.2021,-20.0027"/>
</g>
-<!-- Node10->Node20 -->
+<!-- Node10->Node18 -->
<g id="edge14" class="edge">
-<title>Node10->Node20</title>
+<title>Node10->Node18</title>
<path fill="none" stroke="#191970" d="M1398.2722,-68.5179C1393.4487,-67.9582 1388.646,-67.4424 1384,-67 1035.4375,-33.8077 610.8716,-19.845 490.9408,-16.3933"/>
<polygon fill="#191970" stroke="#191970" points="490.7061,-12.8853 480.6109,-16.1005 490.5077,-19.8825 490.7061,-12.8853"/>
</g>
-<!-- Node23->Node20 -->
+<!-- Node21->Node18 -->
<g id="edge21" class="edge">
-<title>Node23->Node20</title>
+<title>Node21->Node18</title>
<path fill="none" stroke="#191970" d="M1035.8251,-254.1055C961.9279,-244.8249 844.53,-227.936 804,-210 746.3606,-184.4925 745.4532,-156.402 692,-123 644.1282,-93.0858 629.2873,-90.5835 578,-67 548.4915,-53.4311 514.4295,-38.9207 490.1515,-28.7719"/>
<polygon fill="#191970" stroke="#191970" points="491.2883,-25.4539 480.7113,-24.8393 488.5965,-31.9157 491.2883,-25.4539"/>
</g>
-<!-- Node23->Node22 -->
+<!-- Node21->Node20 -->
<g id="edge22" class="edge">
-<title>Node23->Node22</title>
+<title>Node21->Node20</title>
<path fill="none" stroke="#191970" d="M1162.1615,-259.8856C1395.587,-253.7624 2198.6837,-231.2411 2249,-210 2267.5388,-202.1738 2263.696,-187.3607 2282,-179 2355.5567,-145.4017 2603.5026,-136.0401 2694.3107,-133.7053"/>
<polygon fill="#191970" stroke="#191970" points="2694.4261,-137.2036 2704.3375,-133.4599 2694.2548,-130.2057 2694.4261,-137.2036"/>
</g>
-<!-- Node23->Node24 -->
+<!-- Node21->Node22 -->
<g id="edge19" class="edge">
-<title>Node23->Node24</title>
+<title>Node21->Node22</title>
<path fill="none" stroke="#191970" d="M1035.9034,-257.8751C967.8547,-252.5646 857.0329,-239.8228 766,-210 719.262,-194.6884 669.4116,-165.8118 641.1408,-148.102"/>
<polygon fill="#191970" stroke="#191970" points="642.734,-144.9676 632.4163,-142.56 638.9806,-150.8763 642.734,-144.9676"/>
</g>
-<!-- Node25 -->
+<!-- Node23 -->
<g id="node19" class="node">
-<title>Node25</title>
+<title>Node23</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1612,-123.5 1612,-142.5 1670,-142.5 1670,-123.5 1612,-123.5"/>
<text text-anchor="middle" x="1641" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">memory</text>
</g>
-<!-- Node23->Node25 -->
+<!-- Node21->Node23 -->
<g id="edge20" class="edge">
-<title>Node23->Node25</title>
+<title>Node21->Node23</title>
<path fill="none" stroke="#191970" d="M1162.235,-246.5079C1275.4149,-219.6747 1509.4244,-164.1946 1602.0384,-142.2372"/>
<polygon fill="#191970" stroke="#191970" points="1602.9211,-145.625 1611.844,-139.9125 1601.3063,-138.8138 1602.9211,-145.625"/>
</g>
-<!-- Node30->Node11 -->
+<!-- Node28->Node11 -->
<g id="edge24" class="edge">
-<title>Node30->Node11</title>
+<title>Node28->Node11</title>
<path fill="none" stroke="#191970" d="M2404.1291,-184.8762C2407.2896,-157.9361 2416.3026,-81.1108 2421.0276,-40.8362"/>
<polygon fill="#191970" stroke="#191970" points="2424.5221,-41.0867 2422.2113,-30.7469 2417.5698,-40.271 2424.5221,-41.0867"/>
</g>
-<!-- Node30->Node17 -->
+<!-- Node28->Node15 -->
<g id="edge25" class="edge">
-<title>Node30->Node17</title>
+<title>Node28->Node15</title>
<path fill="none" stroke="#191970" d="M2387.9997,-184.8762C2342.1108,-155.4349 2203.3598,-66.4154 2147.5872,-30.633"/>
<polygon fill="#191970" stroke="#191970" points="2149.3811,-27.6256 2139.0744,-25.1714 2145.6011,-33.5172 2149.3811,-27.6256"/>
</g>
-<!-- Node30->Node18 -->
+<!-- Node28->Node16 -->
<g id="edge26" class="edge">
-<title>Node30->Node18</title>
+<title>Node28->Node16</title>
<path fill="none" stroke="#191970" d="M2415.841,-184.7104C2449.8186,-159.4637 2545.8049,-92.3558 2638,-67 2754.666,-34.9141 3133.312,-20.2418 3246.6557,-16.4976"/>
<polygon fill="#191970" stroke="#191970" points="3246.8973,-19.9917 3256.7785,-16.1691 3246.6703,-12.9953 3246.8973,-19.9917"/>
</g>
-<!-- Node30->Node19 -->
+<!-- Node28->Node17 -->
<g id="edge27" class="edge">
-<title>Node30->Node19</title>
+<title>Node28->Node17</title>
<path fill="none" stroke="#191970" d="M2380.2189,-184.9444C2314.8086,-157.675 2126.9612,-80.5753 2062,-67 1885.8836,-30.1961 1321.1118,-18.7234 1157.6904,-16.1324"/>
<polygon fill="#191970" stroke="#191970" points="1157.5681,-12.6302 1147.5149,-15.9744 1157.4593,-19.6294 1157.5681,-12.6302"/>
</g>
-<!-- Node31->Node8 -->
+<!-- Node29->Node8 -->
<g id="edge30" class="edge">
-<title>Node31->Node8</title>
+<title>Node29->Node8</title>
<path fill="none" stroke="#191970" d="M2037.312,-325.8522C1970.0146,-321.3232 1855.1707,-309.1901 1762,-277 1714.5334,-260.6005 1664.6512,-229.1173 1636.9199,-210.1409"/>
<polygon fill="#191970" stroke="#191970" points="1638.6102,-207.0536 1628.4,-204.2248 1634.6176,-212.8033 1638.6102,-207.0536"/>
</g>
-<!-- Node31->Node18 -->
+<!-- Node29->Node16 -->
<g id="edge54" class="edge">
-<title>Node31->Node18</title>
+<title>Node29->Node16</title>
<path fill="none" stroke="#191970" d="M2111.5207,-313.2196C2138.5768,-289.6543 2187.6763,-247.028 2190,-246 2309.1784,-193.2779 2364.9352,-269.2627 2481,-210 2528.0523,-185.9751 2519.5883,-154.5017 2562,-123 2584.6584,-106.1703 2648.8045,-74.553 2676,-67 2785.1188,-36.6944 3138.0477,-20.9098 3246.7979,-16.6801"/>
<polygon fill="#191970" stroke="#191970" points="3246.9729,-20.176 3256.8317,-16.2959 3246.705,-13.1812 3246.9729,-20.176"/>
</g>
-<!-- Node31->Node30 -->
+<!-- Node29->Node28 -->
<g id="edge31" class="edge">
-<title>Node31->Node30</title>
+<title>Node29->Node28</title>
<path fill="none" stroke="#191970" d="M2101.0752,-313.348C2110.7207,-294.4534 2129.8369,-262.6203 2156,-246 2163.9293,-240.9629 2272.1639,-219.5474 2342.7003,-205.9682"/>
<polygon fill="#191970" stroke="#191970" points="2343.6573,-209.3484 2352.8173,-204.0245 2342.3365,-202.4742 2343.6573,-209.3484"/>
</g>
-<!-- Node31->Node32 -->
+<!-- Node29->Node30 -->
<g id="edge32" class="edge">
-<title>Node31->Node32</title>
+<title>Node29->Node30</title>
<path fill="none" stroke="#191970" d="M2056.5219,-313.4639C2027.618,-301.8677 1988.0146,-285.9789 1960.2343,-274.8335"/>
<polygon fill="#191970" stroke="#191970" points="1961.5223,-271.5791 1950.9381,-271.1039 1958.9158,-278.0758 1961.5223,-271.5791"/>
</g>
-<!-- Node31->Node42 -->
+<!-- Node29->Node40 -->
<g id="edge53" class="edge">
-<title>Node31->Node42</title>
+<title>Node29->Node40</title>
<path fill="none" stroke="#191970" d="M2037.3292,-322.3428C1953.1068,-312.7092 1802.7552,-293.4693 1784,-277 1766.1285,-261.3067 1760.8113,-233.3052 1759.3486,-214.3243"/>
<polygon fill="#191970" stroke="#191970" points="1762.8339,-213.9295 1758.8625,-204.1072 1755.8418,-214.2622 1762.8339,-213.9295"/>
</g>
-<!-- Node32->Node10 -->
+<!-- Node30->Node10 -->
<g id="edge49" class="edge">
-<title>Node32->Node10</title>
+<title>Node30->Node10</title>
<path fill="none" stroke="#191970" d="M1923.0711,-251.874C1915.6218,-234.7808 1897.8433,-199.0461 1872,-179 1818.6491,-137.6168 1631.1058,-103.4175 1527.7122,-87.178"/>
<polygon fill="#191970" stroke="#191970" points="1528.0905,-83.6948 1517.6713,-85.6158 1527.0143,-90.6116 1528.0905,-83.6948"/>
</g>
-<!-- Node32->Node11 -->
+<!-- Node30->Node11 -->
<g id="edge33" class="edge">
-<title>Node32->Node11</title>
+<title>Node30->Node11</title>
<path fill="none" stroke="#191970" d="M1946.483,-251.8565C2020.2868,-215.3258 2283.588,-84.9997 2384.3688,-35.1163"/>
<polygon fill="#191970" stroke="#191970" points="2386.193,-38.1187 2393.6026,-30.5458 2383.0877,-31.8451 2386.193,-38.1187"/>
</g>
-<!-- Node32->Node20 -->
+<!-- Node30->Node18 -->
<g id="edge51" class="edge">
-<title>Node32->Node20</title>
+<title>Node30->Node18</title>
<path fill="none" stroke="#191970" d="M1916.4868,-251.7973C1895.9559,-233.55 1848.2993,-194.5614 1800,-179 1638.8569,-127.082 1206.9682,-164.1952 1039,-143 916.9785,-127.6026 887.4336,-116.3587 768,-87 736.5793,-79.2763 729.3911,-74.8431 698,-67 624.3556,-48.5998 537.0016,-30.8722 490.6704,-21.7895"/>
<polygon fill="#191970" stroke="#191970" points="491.1012,-18.3076 480.6161,-19.8281 489.7609,-25.1781 491.1012,-18.3076"/>
</g>
-<!-- Node32->Node22 -->
+<!-- Node30->Node20 -->
<g id="edge52" class="edge">
-<title>Node32->Node22</title>
+<title>Node30->Node20</title>
<path fill="none" stroke="#191970" d="M1965.1258,-251.9643C1974.5265,-249.8112 1984.596,-247.671 1994,-246 2121.0068,-223.4328 2161.9443,-257.1871 2282,-210 2303.9269,-201.3818 2303.0273,-187.5008 2325,-179 2392.3122,-152.9584 2610.2096,-139.128 2694.3839,-134.6498"/>
<polygon fill="#191970" stroke="#191970" points="2694.6239,-138.1421 2704.4282,-134.1263 2694.2595,-131.1516 2694.6239,-138.1421"/>
</g>
-<!-- Node32->Node30 -->
+<!-- Node30->Node28 -->
<g id="edge48" class="edge">
-<title>Node32->Node30</title>
+<title>Node30->Node28</title>
<path fill="none" stroke="#191970" d="M1989.661,-252.9943C2006.6425,-250.7056 2025.0236,-248.2432 2042,-246 2165.5136,-229.6791 2196.9393,-229.443 2320,-210 2327.9262,-208.7477 2336.2579,-207.3222 2344.4592,-205.8549"/>
<polygon fill="#191970" stroke="#191970" points="2345.1503,-209.2867 2354.3622,-204.0531 2343.8972,-202.3997 2345.1503,-209.2867"/>
</g>
-<!-- Node32->Node35 -->
+<!-- Node30->Node33 -->
<g id="edge34" class="edge">
-<title>Node32->Node35</title>
+<title>Node30->Node33</title>
<path fill="none" stroke="#191970" d="M1864.4632,-257.0521C1696.7753,-245.1252 1237.7095,-212.474 1058.2626,-199.7108"/>
<polygon fill="#191970" stroke="#191970" points="1058.3554,-196.2087 1048.1322,-198.9903 1057.8587,-203.191 1058.3554,-196.2087"/>
</g>
-<!-- Node32->Node42 -->
+<!-- Node30->Node40 -->
<g id="edge50" class="edge">
-<title>Node32->Node42</title>
+<title>Node30->Node40</title>
<path fill="none" stroke="#191970" d="M1902.9297,-251.9005C1873.987,-240.3579 1825.2379,-220.9163 1792.5983,-207.8993"/>
<polygon fill="#191970" stroke="#191970" points="1793.6108,-204.5351 1783.0257,-204.0817 1791.0177,-211.0371 1793.6108,-204.5351"/>
</g>
-<!-- Node35->Node10 -->
+<!-- Node33->Node10 -->
<g id="edge40" class="edge">
-<title>Node35->Node10</title>
+<title>Node33->Node10</title>
<path fill="none" stroke="#191970" d="M991.852,-179.2362C1000.3149,-162.3494 1016.4127,-135.8323 1039,-123 1068.6172,-106.1739 1276.7869,-89.4908 1388.1222,-81.643"/>
<polygon fill="#191970" stroke="#191970" points="1388.5264,-85.1234 1398.2578,-80.9341 1388.038,-78.1404 1388.5264,-85.1234"/>
</g>
-<!-- Node35->Node17 -->
+<!-- Node33->Node15 -->
<g id="edge35" class="edge">
-<title>Node35->Node17</title>
+<title>Node33->Node15</title>
<path fill="none" stroke="#191970" d="M986.531,-179.315C989.0567,-162.759 995.6915,-136.7889 1013,-123 1145.1466,-17.7249 1221.0465,-85.363 1389,-67 1630.9745,-40.5439 1919.3317,-24.9633 2051.4158,-18.708"/>
<polygon fill="#191970" stroke="#191970" points="2051.5926,-22.2036 2061.4173,-18.2381 2051.264,-15.2114 2051.5926,-22.2036"/>
</g>
-<!-- Node35->Node18 -->
+<!-- Node33->Node16 -->
<g id="edge43" class="edge">
-<title>Node35->Node18</title>
+<title>Node33->Node16</title>
<path fill="none" stroke="#191970" d="M1038.7287,-179.4334C1101.2009,-162.5967 1207.6857,-135.9984 1301,-123 1699.5979,-67.4764 3022.6114,-23.6255 3246.6372,-16.5115"/>
<polygon fill="#191970" stroke="#191970" points="3246.8664,-20.0061 3256.7507,-16.1917 3246.6451,-13.0096 3246.8664,-20.0061"/>
</g>
-<!-- Node35->Node19 -->
+<!-- Node33->Node17 -->
<g id="edge46" class="edge">
-<title>Node35->Node19</title>
+<title>Node33->Node17</title>
<path fill="none" stroke="#191970" d="M976.7115,-179.1404C964.0874,-153.4215 943.9146,-101.1139 968,-67 979.9162,-50.1222 1030.3677,-34.8388 1068.4375,-25.3846"/>
<polygon fill="#191970" stroke="#191970" points="1069.4157,-28.749 1078.3105,-22.9928 1067.7675,-21.9458 1069.4157,-28.749"/>
</g>
-<!-- Node35->Node20 -->
+<!-- Node33->Node18 -->
<g id="edge45" class="edge">
-<title>Node35->Node20</title>
+<title>Node33->Node18</title>
<path fill="none" stroke="#191970" d="M948.0897,-179.4182C911.977,-164.7424 855.3613,-141.9474 806,-123 738.787,-97.2003 722.6724,-88.6173 654,-67 597.0081,-49.0596 529.3771,-32.2582 490.3027,-22.9796"/>
<polygon fill="#191970" stroke="#191970" points="491.0759,-19.566 480.5395,-20.6769 489.469,-26.3791 491.0759,-19.566"/>
</g>
-<!-- Node35->Node22 -->
+<!-- Node33->Node20 -->
<g id="edge47" class="edge">
-<title>Node35->Node22</title>
+<title>Node33->Node20</title>
<path fill="none" stroke="#191970" d="M1048.064,-192.2748C1331.3622,-182.279 2481.7637,-141.6882 2694.1868,-134.1931"/>
<polygon fill="#191970" stroke="#191970" points="2694.5365,-137.683 2704.4068,-133.8325 2694.2896,-130.6874 2694.5365,-137.683"/>
</g>
-<!-- Node35->Node24 -->
+<!-- Node33->Node22 -->
<g id="edge41" class="edge">
-<title>Node35->Node24</title>
+<title>Node33->Node22</title>
<path fill="none" stroke="#191970" d="M921.9105,-183.9278C847.1684,-171.4029 724.7086,-150.8817 660.4316,-140.1105"/>
<polygon fill="#191970" stroke="#191970" points="660.6969,-136.6062 650.2559,-138.4053 659.5399,-143.5099 660.6969,-136.6062"/>
</g>
-<!-- Node35->Node25 -->
+<!-- Node33->Node23 -->
<g id="edge42" class="edge">
-<title>Node35->Node25</title>
+<title>Node33->Node23</title>
<path fill="none" stroke="#191970" d="M1048.1978,-188.5752C1182.2942,-176.0037 1491.9962,-146.9691 1601.6467,-136.6894"/>
<polygon fill="#191970" stroke="#191970" points="1602.2917,-140.1443 1611.9213,-135.7261 1601.6383,-133.1749 1602.2917,-140.1443"/>
</g>
-<!-- Node35->Node27 -->
+<!-- Node33->Node25 -->
<g id="edge36" class="edge">
-<title>Node35->Node27</title>
+<title>Node33->Node25</title>
<path fill="none" stroke="#191970" d="M1016.3123,-179.4554C1037.0138,-169.509 1063.9523,-156.5659 1084.1353,-146.8686"/>
<polygon fill="#191970" stroke="#191970" points="1085.6667,-150.0159 1093.1645,-142.5303 1082.6352,-143.7064 1085.6667,-150.0159"/>
</g>
-<!-- Node35->Node38 -->
+<!-- Node33->Node36 -->
<g id="edge44" class="edge">
-<title>Node35->Node38</title>
+<title>Node33->Node36</title>
<path fill="none" stroke="#191970" d="M1048.1047,-184.0674C1118.7901,-172.3815 1232.3366,-153.6097 1300.2991,-142.3739"/>
<polygon fill="#191970" stroke="#191970" points="1301.0981,-145.7894 1310.3933,-140.7051 1299.9563,-138.8832 1301.0981,-145.7894"/>
</g>
-<!-- Node27->Node10 -->
+<!-- Node25->Node10 -->
<g id="edge37" class="edge">
-<title>Node27->Node10</title>
+<title>Node25->Node10</title>
<path fill="none" stroke="#191970" d="M1171.9156,-123.4369C1232.2001,-113.6516 1325.8019,-98.4582 1389.3894,-88.1368"/>
<polygon fill="#191970" stroke="#191970" points="1390.0174,-91.5808 1399.3274,-86.5237 1388.8958,-84.6712 1390.0174,-91.5808"/>
</g>
-<!-- Node27->Node19 -->
+<!-- Node25->Node17 -->
<g id="edge38" class="edge">
-<title>Node27->Node19</title>
+<title>Node25->Node17</title>
<path fill="none" stroke="#191970" d="M1113,-123.3845C1113,-104.1913 1113,-60.786 1113,-35.2663"/>
<polygon fill="#191970" stroke="#191970" points="1116.5001,-35.2483 1113,-25.2484 1109.5001,-35.2484 1116.5001,-35.2483"/>
</g>
-<!-- Node27->Node20 -->
+<!-- Node25->Node18 -->
<g id="edge39" class="edge">
-<title>Node27->Node20</title>
+<title>Node25->Node18</title>
<path fill="none" stroke="#191970" d="M1065.0702,-123.4479C998.504,-110.2691 874.2551,-85.9866 768,-67 667.0907,-48.9686 547.3151,-29.6634 490.7248,-20.6674"/>
<polygon fill="#191970" stroke="#191970" points="491.0175,-17.1701 480.5925,-19.0596 489.9204,-24.0836 491.0175,-17.1701"/>
</g>
-<!-- Node43->Node10 -->
+<!-- Node41->Node10 -->
<g id="edge79" class="edge">
-<title>Node43->Node10</title>
+<title>Node41->Node10</title>
<path fill="none" stroke="#191970" d="M2289.936,-318.8626C2242.8958,-310.2571 2175.2659,-296.0648 2118,-277 1956.3674,-223.1897 1931.0386,-168.9562 1767,-123 1686.6079,-100.4778 1591.1745,-88.409 1527.8542,-82.3655"/>
<polygon fill="#191970" stroke="#191970" points="1527.8667,-78.8517 1517.5857,-81.4114 1527.219,-85.8217 1527.8667,-78.8517"/>
</g>
-<!-- Node43->Node11 -->
+<!-- Node41->Node11 -->
<g id="edge61" class="edge">
-<title>Node43->Node11</title>
+<title>Node41->Node11</title>
<path fill="none" stroke="#191970" d="M2378.7591,-313.4038C2412.6153,-294.6672 2464.4442,-259.0838 2481,-210 2501.5277,-149.1405 2463.8043,-76.1869 2440.4614,-39.3259"/>
<polygon fill="#191970" stroke="#191970" points="2443.2725,-37.2289 2434.8807,-30.7614 2437.4077,-41.0505 2443.2725,-37.2289"/>
</g>
-<!-- Node43->Node17 -->
+<!-- Node41->Node15 -->
<g id="edge68" class="edge">
-<title>Node43->Node17</title>
+<title>Node41->Node15</title>
<path fill="none" stroke="#191970" d="M2340.4055,-313.3193C2330.1065,-294.3943 2309.8681,-262.5293 2283,-246 2214.2546,-203.7077 2156.1036,-272.4739 2105,-210 2063.1762,-158.8706 2095.9584,-73.0124 2114.0577,-34.8234"/>
<polygon fill="#191970" stroke="#191970" points="2117.4271,-35.9028 2118.6991,-25.3846 2111.1455,-32.8139 2117.4271,-35.9028"/>
</g>
-<!-- Node43->Node18 -->
+<!-- Node41->Node16 -->
<g id="edge83" class="edge">
-<title>Node43->Node18</title>
+<title>Node41->Node16</title>
<path fill="none" stroke="#191970" d="M2384.2234,-313.4595C2406.4856,-303.8221 2435.3641,-290.6054 2460,-277 2627.2156,-184.6537 2649.8814,-114.1167 2835,-67 2985.8809,-28.5975 3172.9782,-18.7564 3246.8506,-16.2959"/>
<polygon fill="#191970" stroke="#191970" points="3247.0238,-19.7924 3256.9105,-15.984 3246.8067,-12.7957 3247.0238,-19.7924"/>
</g>
-<!-- Node43->Node19 -->
+<!-- Node41->Node17 -->
<g id="edge85" class="edge">
-<title>Node43->Node19</title>
+<title>Node41->Node17</title>
<path fill="none" stroke="#191970" d="M2298.5569,-313.4599C2267.2379,-303.6684 2225.9979,-290.2869 2190,-277 1934.3739,-182.6475 1881.5953,-118.3797 1614,-67 1447.5265,-35.0362 1245.868,-21.9494 1157.7893,-17.4798"/>
<polygon fill="#191970" stroke="#191970" points="1157.7818,-13.9752 1147.621,-16.9767 1157.4359,-20.9667 1157.7818,-13.9752"/>
</g>
-<!-- Node43->Node20 -->
+<!-- Node41->Node18 -->
<g id="edge86" class="edge">
-<title>Node43->Node20</title>
+<title>Node41->Node18</title>
<path fill="none" stroke="#191970" d="M2289.5678,-323.6835C2020.5298,-301.5061 914.0602,-210.2853 913,-210 789.9455,-176.885 768.2646,-143.4146 654,-87 636.5332,-78.3763 632.9766,-74.5033 615,-67 572.8389,-49.4023 522.3717,-33.7681 490.2512,-24.4652"/>
<polygon fill="#191970" stroke="#191970" points="491.0781,-21.0614 480.5009,-21.6728 489.1508,-27.7908 491.0781,-21.0614"/>
</g>
-<!-- Node43->Node22 -->
+<!-- Node41->Node20 -->
<g id="edge87" class="edge">
-<title>Node43->Node22</title>
+<title>Node41->Node20</title>
<path fill="none" stroke="#191970" d="M2406.1803,-321.0332C2470.9178,-311.9905 2570.7895,-295.6346 2604,-277 2658.001,-246.6997 2699.4718,-183.4268 2717.7591,-151.8162"/>
<polygon fill="#191970" stroke="#191970" points="2721.0092,-153.1779 2722.8685,-142.7475 2714.9105,-149.7419 2721.0092,-153.1779"/>
</g>
-<!-- Node43->Node23 -->
+<!-- Node41->Node21 -->
<g id="edge62" class="edge">
-<title>Node43->Node23</title>
+<title>Node41->Node21</title>
<path fill="none" stroke="#191970" d="M2289.6144,-323.0411C2252.5215,-319.7329 2203.4761,-315.6529 2160,-313 1756.8772,-288.4017 1654.8752,-305.3668 1252,-277 1226.0177,-275.1706 1197.6186,-272.4844 1172.418,-269.8603"/>
<polygon fill="#191970" stroke="#191970" points="1172.5108,-266.3507 1162.1986,-268.7815 1171.7759,-273.312 1172.5108,-266.3507"/>
</g>
-<!-- Node43->Node25 -->
+<!-- Node41->Node23 -->
<g id="edge82" class="edge">
-<title>Node43->Node25</title>
+<title>Node41->Node23</title>
<path fill="none" stroke="#191970" d="M2331.8232,-313.2679C2308.7648,-291.871 2267.2243,-254.5629 2249,-246 2235.8878,-239.839 1812.4371,-163.6959 1680.1098,-139.9947"/>
<polygon fill="#191970" stroke="#191970" points="1680.4739,-136.5043 1670.0135,-138.1869 1679.2401,-143.3947 1680.4739,-136.5043"/>
</g>
-<!-- Node43->Node30 -->
+<!-- Node41->Node28 -->
<g id="edge67" class="edge">
-<title>Node43->Node30</title>
+<title>Node41->Node28</title>
<path fill="none" stroke="#191970" d="M2377.9566,-313.3925C2391.6188,-304.7636 2406.4135,-292.5426 2414,-277 2423.7629,-256.9985 2417.7952,-231.1992 2411.4922,-213.82"/>
<polygon fill="#191970" stroke="#191970" points="2414.6169,-212.2067 2407.6619,-204.2142 2408.1147,-214.7994 2414.6169,-212.2067"/>
</g>
-<!-- Node43->Node32 -->
+<!-- Node41->Node30 -->
<g id="edge78" class="edge">
-<title>Node43->Node32</title>
+<title>Node41->Node30</title>
<path fill="none" stroke="#191970" d="M2289.9029,-315.742C2284.8717,-314.7655 2279.8518,-313.8361 2275,-313 2155.2622,-292.3667 2123.8452,-296.9997 2004,-277 1996.7873,-275.7964 1989.2177,-274.4168 1981.7571,-272.988"/>
<polygon fill="#191970" stroke="#191970" points="1982.1808,-269.5046 1971.695,-271.0208 1980.8377,-276.3745 1982.1808,-269.5046"/>
</g>
-<!-- Node43->Node42 -->
+<!-- Node41->Node40 -->
<g id="edge80" class="edge">
-<title>Node43->Node42</title>
+<title>Node41->Node40</title>
<path fill="none" stroke="#191970" d="M2289.912,-323.3654C2163.8228,-312.0925 1874.7319,-285.5058 1855,-277 1821.2483,-262.4507 1790.6225,-231.6989 1773.4443,-212.1254"/>
<polygon fill="#191970" stroke="#191970" points="1775.8873,-209.5962 1766.7351,-204.2589 1770.5614,-214.1387 1775.8873,-209.5962"/>
</g>
-<!-- Node44 -->
+<!-- Node42 -->
<g id="node28" class="node">
-<title>Node44</title>
+<title>Node42</title>
<g id="a_node28"><a xlink:href="map_8h.html" target="_top" xlink:title="Runtime Map container types. ">
<polygon fill="#ffffff" stroke="#ff0000" points="284,-246.5 284,-276.5 410,-276.5 410,-246.5 284,-246.5"/>
<text text-anchor="start" x="292" y="-264.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
@@ -920,441 +920,441 @@
</a>
</g>
</g>
-<!-- Node43->Node44 -->
+<!-- Node41->Node42 -->
<g id="edge63" class="edge">
-<title>Node43->Node44</title>
+<title>Node41->Node42</title>
<path fill="none" stroke="#191970" d="M2289.6243,-322.8721C2252.5351,-319.4983 2203.4904,-315.4088 2160,-313 1812.773,-293.7685 715.0419,-269.3842 420.3939,-263.0559"/>
<polygon fill="#191970" stroke="#191970" points="420.3265,-259.5538 410.2537,-262.8384 420.1763,-266.5521 420.3265,-259.5538"/>
</g>
-<!-- Node45 -->
+<!-- Node43 -->
<g id="node29" class="node">
-<title>Node45</title>
+<title>Node43</title>
<g id="a_node29"><a xlink:href="runtime_2module_8h.html" target="_top" xlink:title="Runtime container of the functions generated by TVM, This is used to support dynamically link...">
<polygon fill="#ffffff" stroke="#ff0000" points="2114,-185 2114,-204 2240,-204 2240,-185 2114,-185"/>
<text text-anchor="middle" x="2177" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/module.h</text>
</a>
</g>
</g>
-<!-- Node43->Node45 -->
+<!-- Node41->Node43 -->
<g id="edge69" class="edge">
-<title>Node43->Node45</title>
+<title>Node41->Node43</title>
<path fill="none" stroke="#191970" d="M2340.4642,-313.3586C2335.5253,-295.0667 2329.1191,-264.4571 2311,-246 2291.9066,-226.5504 2264.9415,-214.4118 2240.1561,-206.8509"/>
<polygon fill="#191970" stroke="#191970" points="2241.0011,-203.4524 2230.4246,-204.0756 2239.0812,-210.184 2241.0011,-203.4524"/>
</g>
-<!-- Node43->Node46 -->
+<!-- Node41->Node44 -->
<g id="edge81" class="edge">
-<title>Node43->Node46</title>
+<title>Node41->Node44</title>
<path fill="none" stroke="#191970" d="M2406.2322,-321.9538C2530.0315,-308.0367 2815.8999,-275.9005 2911.8984,-265.1087"/>
<polygon fill="#191970" stroke="#191970" points="2912.3292,-268.5824 2921.8756,-263.9871 2911.5472,-261.6263 2912.3292,-268.5824"/>
</g>
-<!-- Node47 -->
+<!-- Node45 -->
<g id="node31" class="node">
-<title>Node47</title>
+<title>Node45</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="2198.5,-252 2198.5,-271 2239.5,-271 2239.5,-252 2198.5,-252"/>
<text text-anchor="middle" x="2219" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tuple</text>
</g>
-<!-- Node43->Node47 -->
+<!-- Node41->Node45 -->
<g id="edge84" class="edge">
-<title>Node43->Node47</title>
+<title>Node41->Node45</title>
<path fill="none" stroke="#191970" d="M2319.0499,-313.4639C2297.3078,-302.1715 2267.7284,-286.8086 2246.3773,-275.7192"/>
<polygon fill="#191970" stroke="#191970" points="2247.9788,-272.6071 2237.4911,-271.1039 2244.7523,-278.8192 2247.9788,-272.6071"/>
</g>
-<!-- Node44->Node20 -->
+<!-- Node42->Node18 -->
<g id="edge66" class="edge">
-<title>Node44->Node20</title>
+<title>Node42->Node18</title>
<path fill="none" stroke="#191970" d="M353.7765,-246.4819C373.0309,-203.8098 428.0468,-81.8828 449.3904,-34.5808"/>
<polygon fill="#191970" stroke="#191970" points="452.7025,-35.7501 453.6252,-25.1956 446.322,-32.8711 452.7025,-35.7501"/>
</g>
-<!-- Node44->Node24 -->
+<!-- Node42->Node22 -->
<g id="edge64" class="edge">
-<title>Node44->Node24</title>
+<title>Node42->Node22</title>
<path fill="none" stroke="#191970" d="M378.884,-246.3816C431.8559,-221.2639 537.2336,-171.297 588.5494,-146.9646"/>
<polygon fill="#191970" stroke="#191970" points="590.2329,-150.0399 597.769,-142.5929 587.2338,-143.7149 590.2329,-150.0399"/>
</g>
-<!-- Node44->Node38 -->
+<!-- Node42->Node36 -->
<g id="edge65" class="edge">
-<title>Node44->Node38</title>
+<title>Node42->Node36</title>
<path fill="none" stroke="#191970" d="M410.1228,-251.4254C514.5588,-234.9538 729.9621,-201.8115 913,-179 1051.7853,-161.7036 1215.5524,-145.9328 1300.1043,-138.1327"/>
<polygon fill="#191970" stroke="#191970" points="1300.6255,-141.5996 1310.2631,-137.1988 1299.9846,-134.629 1300.6255,-141.5996"/>
</g>
-<!-- Node45->Node10 -->
+<!-- Node43->Node10 -->
<g id="edge72" class="edge">
-<title>Node45->Node10</title>
+<title>Node43->Node10</title>
<path fill="none" stroke="#191970" d="M2156.9273,-184.9022C2122.7263,-169.0291 2050.5782,-137.6139 1986,-123 1900.6734,-103.6908 1650.789,-87.687 1527.8886,-80.7318"/>
<polygon fill="#191970" stroke="#191970" points="1528.0116,-77.2333 1517.831,-80.1668 1527.6189,-84.2223 1528.0116,-77.2333"/>
</g>
-<!-- Node45->Node11 -->
+<!-- Node43->Node11 -->
<g id="edge70" class="edge">
-<title>Node45->Node11</title>
+<title>Node43->Node11</title>
<path fill="none" stroke="#191970" d="M2190.2798,-184.8762C2228.7223,-157.017 2340.7777,-75.8109 2394.5458,-36.8454"/>
<polygon fill="#191970" stroke="#191970" points="2396.9175,-39.4491 2402.9609,-30.7469 2392.8098,-33.781 2396.9175,-39.4491"/>
</g>
-<!-- Node45->Node18 -->
+<!-- Node43->Node16 -->
<g id="edge74" class="edge">
-<title>Node45->Node18</title>
+<title>Node43->Node16</title>
<path fill="none" stroke="#191970" d="M2200.5714,-184.8974C2261.8879,-160.3939 2431.9104,-95.4773 2580,-67 2710.5057,-41.9041 3127.1963,-22.1407 3246.6879,-16.8804"/>
<polygon fill="#191970" stroke="#191970" points="3247.1545,-20.3635 3256.9922,-16.4305 3246.849,-13.3702 3247.1545,-20.3635"/>
</g>
-<!-- Node45->Node22 -->
+<!-- Node43->Node20 -->
<g id="edge76" class="edge">
-<title>Node45->Node22</title>
+<title>Node43->Node20</title>
<path fill="none" stroke="#191970" d="M2220.4362,-184.9519C2231.3547,-182.774 2243.0749,-180.6254 2254,-179 2418.0195,-154.5977 2615.9804,-140.1965 2694.0216,-135.104"/>
<polygon fill="#191970" stroke="#191970" points="2694.4673,-138.5826 2704.2217,-134.4471 2694.0174,-131.5971 2694.4673,-138.5826"/>
</g>
-<!-- Node45->Node25 -->
+<!-- Node43->Node23 -->
<g id="edge73" class="edge">
-<title>Node45->Node25</title>
+<title>Node43->Node23</title>
<path fill="none" stroke="#191970" d="M2113.9378,-187.2643C2002.0231,-174.4234 1771.8406,-148.0125 1680.0461,-137.4801"/>
<polygon fill="#191970" stroke="#191970" points="1680.3981,-133.9976 1670.0643,-136.3348 1679.6001,-140.952 1680.3981,-133.9976"/>
</g>
-<!-- Node45->Node27 -->
+<!-- Node43->Node25 -->
<g id="edge71" class="edge">
-<title>Node45->Node27</title>
+<title>Node43->Node25</title>
<path fill="none" stroke="#191970" d="M2113.6283,-190.8371C1928.1696,-180.1174 1387.0754,-148.8418 1187.8036,-137.3237"/>
<polygon fill="#191970" stroke="#191970" points="1187.9084,-133.824 1177.7231,-136.741 1187.5044,-140.8123 1187.9084,-133.824"/>
</g>
-<!-- Node45->Node38 -->
+<!-- Node43->Node36 -->
<g id="edge75" class="edge">
-<title>Node45->Node38</title>
+<title>Node43->Node36</title>
<path fill="none" stroke="#191970" d="M2113.6783,-189.7509C1958.6814,-178.1261 1562.1061,-148.383 1413.962,-137.2721"/>
<polygon fill="#191970" stroke="#191970" points="1413.9654,-133.7627 1403.7316,-136.5049 1413.4418,-140.7431 1413.9654,-133.7627"/>
</g>
-<!-- Node45->Node43 -->
+<!-- Node43->Node41 -->
<g id="edge77" class="edge">
-<title>Node45->Node43</title>
+<title>Node43->Node41</title>
<path fill="none" stroke="#191970" d="M2240.0837,-202.4984C2270.7029,-209.4273 2305.7084,-222.2738 2329,-246 2344.0756,-261.3569 2351.0427,-285.1267 2352.7127,-303.2146"/>
<polygon fill="#191970" stroke="#191970" points="2349.2209,-303.4993 2353.0996,-313.3586 2356.2158,-303.2324 2349.2209,-303.4993"/>
</g>
-<!-- Node48->Node8 -->
+<!-- Node46->Node8 -->
<g id="edge92" class="edge">
-<title>Node48->Node8</title>
+<title>Node46->Node8</title>
<path fill="none" stroke="#191970" d="M1548.26,-380.4549C1538.869,-372.0029 1526.4091,-358.7442 1521,-344 1516.2547,-331.0652 1516.4663,-326.0105 1521,-313 1535.776,-270.5967 1572.7533,-231.969 1595.6796,-210.9925"/>
<polygon fill="#191970" stroke="#191970" points="1598.2882,-213.3559 1603.4232,-204.0886 1593.6298,-208.1309 1598.2882,-213.3559"/>
</g>
-<!-- Node49 -->
+<!-- Node47 -->
<g id="node33" class="node">
-<title>Node49</title>
+<title>Node47</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1530,-319 1530,-338 1590,-338 1590,-319 1530,-319"/>
<text text-anchor="middle" x="1560" y="-326" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">iostream</text>
</g>
-<!-- Node48->Node49 -->
+<!-- Node46->Node47 -->
<g id="edge93" class="edge">
-<title>Node48->Node49</title>
+<title>Node46->Node47</title>
<path fill="none" stroke="#191970" d="M1560,-380.3906C1560,-371.8657 1560,-359.1392 1560,-348.4235"/>
<polygon fill="#191970" stroke="#191970" points="1563.5001,-348.2448 1560,-338.2449 1556.5001,-348.2449 1563.5001,-348.2448"/>
</g>
-<!-- Node50->Node4 -->
+<!-- Node48->Node4 -->
<g id="edge106" class="edge">
-<title>Node50->Node4</title>
+<title>Node48->Node4</title>
<path fill="none" stroke="#191970" d="M1974.4421,-548.3733C1998.7326,-539.1822 2035.5208,-525.2624 2062.4543,-515.0713"/>
<polygon fill="#191970" stroke="#191970" points="2063.7757,-518.3136 2071.8899,-511.5011 2061.2984,-511.7666 2063.7757,-518.3136"/>
</g>
-<!-- Node50->Node5 -->
+<!-- Node48->Node5 -->
<g id="edge107" class="edge">
-<title>Node50->Node5</title>
+<title>Node48->Node5</title>
<path fill="none" stroke="#191970" d="M1951.3925,-548.3232C1955.0703,-534.7844 1963.0748,-509.8479 1976,-492 1984.3521,-480.467 1996.1754,-469.9681 2006.6305,-461.9219"/>
<polygon fill="#191970" stroke="#191970" points="2009.007,-464.5185 2014.9847,-455.7711 2004.8567,-458.8815 2009.007,-464.5185"/>
</g>
-<!-- Node50->Node10 -->
+<!-- Node48->Node10 -->
<g id="edge110" class="edge">
-<title>Node50->Node10</title>
+<title>Node48->Node10</title>
<path fill="none" stroke="#191970" d="M1908.7334,-556.6793C1774.8174,-552.0801 1349.6634,-535.7437 1293,-512 1220.2534,-481.5169 1059.4211,-348.9038 1027,-277 1021.3367,-264.44 1019.0598,-257.2597 1027,-246 1054.4639,-207.0546 1084.1846,-228.741 1128,-210 1207.1295,-176.1543 1220.9321,-154.5619 1301,-123 1336.1691,-109.1367 1377.2064,-97.2548 1408.5037,-89.0412"/>
<polygon fill="#191970" stroke="#191970" points="1409.4988,-92.399 1418.3031,-86.5055 1407.7452,-85.6222 1409.4988,-92.399"/>
</g>
-<!-- Node50->Node18 -->
+<!-- Node48->Node16 -->
<g id="edge111" class="edge">
-<title>Node50->Node18</title>
+<title>Node48->Node16</title>
<path fill="none" stroke="#191970" d="M1989.0395,-555.1307C2201.157,-538.8625 3184,-449.7138 3184,-194.5 3184,-194.5 3184,-194.5 3184,-133 3184,-88.262 3226.161,-50.9849 3254.3091,-31.0828"/>
<polygon fill="#191970" stroke="#191970" points="3256.5884,-33.7652 3262.8845,-25.2441 3252.6488,-27.979 3256.5884,-33.7652"/>
</g>
-<!-- Node50->Node23 -->
+<!-- Node48->Node21 -->
<g id="edge108" class="edge">
-<title>Node50->Node23</title>
+<title>Node48->Node21</title>
<path fill="none" stroke="#191970" d="M1908.7904,-556.2744C1781.0098,-550.6119 1389.4747,-531.8211 1336,-512 1296.6586,-497.4176 1288.248,-487.0233 1260,-456 1210.0686,-401.1629 1224.8376,-366.998 1174,-313 1162.9157,-301.2266 1148.7025,-290.6337 1135.6356,-282.1599"/>
<polygon fill="#191970" stroke="#191970" points="1137.1581,-278.9837 1126.8249,-276.6435 1133.4434,-284.9168 1137.1581,-278.9837"/>
</g>
-<!-- Node50->Node30 -->
+<!-- Node48->Node28 -->
<g id="edge109" class="edge">
-<title>Node50->Node30</title>
+<title>Node48->Node28</title>
<path fill="none" stroke="#191970" d="M1960.0318,-548.4703C1977.4745,-533.9316 2013.0431,-506.3876 2048,-492 2097.3092,-471.7052 2485.1016,-441.1603 2519,-400 2571.2389,-336.5701 2470.4134,-246.3862 2424.3049,-210.308"/>
<polygon fill="#191970" stroke="#191970" points="2426.1904,-207.343 2416.1274,-204.0283 2421.9269,-212.8949 2426.1904,-207.343"/>
</g>
-<!-- Node51->Node4 -->
+<!-- Node49->Node4 -->
<g id="edge129" class="edge">
-<title>Node51->Node4</title>
+<title>Node49->Node4</title>
<path fill="none" stroke="#191970" d="M3286.261,-844.6004C3133.8852,-829.7906 2662.16,-773.9881 2304,-624 2232.3485,-593.9942 2155.407,-543.0873 2118.7849,-517.5641"/>
<polygon fill="#191970" stroke="#191970" points="2120.5647,-514.537 2110.3707,-511.6502 2116.5395,-520.264 2120.5647,-514.537"/>
</g>
-<!-- Node51->Node5 -->
+<!-- Node49->Node5 -->
<g id="edge130" class="edge">
-<title>Node51->Node5</title>
+<title>Node49->Node5</title>
<path fill="none" stroke="#191970" d="M3286.3498,-840.2976C3163.9008,-818.3411 2838.1685,-756.9215 2573,-680 2502.653,-659.5934 2484.8933,-653.8399 2418,-624 2304.6843,-573.4518 2286.1445,-540.6476 2172,-492 2140.3325,-478.5035 2103.2951,-466.6279 2074.9874,-458.325"/>
<polygon fill="#191970" stroke="#191970" points="2075.9406,-454.9573 2065.3618,-455.5404 2073.9952,-461.6816 2075.9406,-454.9573"/>
</g>
-<!-- Node51->Node18 -->
+<!-- Node49->Node16 -->
<g id="edge157" class="edge">
-<title>Node51->Node18</title>
+<title>Node49->Node16</title>
<path fill="none" stroke="#191970" d="M3367.212,-839.4178C3393.1209,-831.0375 3430.6636,-817.8989 3462,-803 3486.2996,-791.4467 3489.0947,-782.1821 3514,-772 3576.6613,-746.382 3608.762,-777.8142 3662,-736 3709.9717,-698.3221 3754,-477.4482 3754,-328.5 3754,-328.5 3754,-328.5 3754,-133 3754,-98.6999 3742.8423,-85.5639 3714,-67 3680.2495,-45.277 3406.3344,-24.3587 3311.3172,-17.6911"/>
<polygon fill="#191970" stroke="#191970" points="3311.3713,-14.1865 3301.1526,-16.9845 3310.8857,-21.1696 3311.3713,-14.1865"/>
</g>
-<!-- Node51->Node22 -->
+<!-- Node49->Node20 -->
<g id="edge158" class="edge">
-<title>Node51->Node22</title>
+<title>Node49->Node20</title>
<path fill="none" stroke="#191970" d="M3353.1366,-839.3204C3379.4216,-824.5312 3431.342,-795.5466 3476,-772 3564.965,-725.0918 3678,-770.5742 3678,-670 3678,-670 3678,-670 3678,-502 3678,-309.0737 3561.3505,-263.6767 3388,-179 3331.5277,-151.4149 2889.1828,-137.374 2762.0591,-133.8814"/>
<polygon fill="#191970" stroke="#191970" points="2761.8555,-130.3747 2751.7645,-133.6028 2761.6661,-137.3722 2761.8555,-130.3747"/>
</g>
-<!-- Node52 -->
+<!-- Node50 -->
<g id="node36" class="node">
-<title>Node52</title>
+<title>Node50</title>
<g id="a_node36"><a xlink:href="tir_2expr_8h.html" target="_top" xlink:title="TIR expressions. ">
<polygon fill="#ffffff" stroke="#000000" points="2224.5,-778 2224.5,-797 2307.5,-797 2307.5,-778 2224.5,-778"/>
<text text-anchor="middle" x="2266" y="-785" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/tir/expr.h</text>
</a>
</g>
</g>
-<!-- Node51->Node52 -->
+<!-- Node49->Node50 -->
<g id="edge131" class="edge">
-<title>Node51->Node52</title>
+<title>Node49->Node50</title>
<path fill="none" stroke="#191970" d="M3286.4931,-843.6986C3269.8287,-842.0335 3251.1216,-840.2941 3234,-839 2884.4631,-812.5809 2462.9094,-795.0601 2317.9155,-789.4468"/>
<polygon fill="#191970" stroke="#191970" points="2317.6825,-785.9353 2307.5551,-789.0478 2317.413,-792.9302 2317.6825,-785.9353"/>
</g>
-<!-- Node52->Node3 -->
+<!-- Node50->Node3 -->
<g id="edge132" class="edge">
-<title>Node52->Node3</title>
+<title>Node50->Node3</title>
<path fill="none" stroke="#191970" d="M2235.8878,-777.9688C2206.2799,-768.3263 2159.9859,-752.5213 2121,-736 2039.2144,-701.3413 1946.1629,-652.5045 1901.5716,-628.4892"/>
<polygon fill="#191970" stroke="#191970" points="1903.0236,-625.2955 1892.5622,-623.6191 1899.6949,-631.4534 1903.0236,-625.2955"/>
</g>
-<!-- Node52->Node5 -->
+<!-- Node50->Node5 -->
<g id="edge134" class="edge">
-<title>Node52->Node5</title>
+<title>Node50->Node5</title>
<path fill="none" stroke="#191970" d="M2261.0247,-777.6821C2253.9379,-763.8055 2240.4167,-737.7241 2228,-716 2184.5391,-639.9614 2190.3186,-606.3455 2125,-548 2096.8256,-522.8333 2072.3329,-540.8976 2048,-512 2037.1347,-499.0965 2032.7851,-480.1952 2031.0657,-465.9233"/>
<polygon fill="#191970" stroke="#191970" points="2034.5278,-465.3332 2030.1765,-455.6731 2027.554,-465.9382 2034.5278,-465.3332"/>
</g>
-<!-- Node52->Node8 -->
+<!-- Node50->Node8 -->
<g id="edge133" class="edge">
-<title>Node52->Node8</title>
+<title>Node50->Node8</title>
<path fill="none" stroke="#191970" d="M2224.4121,-786.4869C1981.2945,-780.5176 754.2703,-749.726 741,-736 676.927,-669.7267 569.1438,-522.1224 1027,-246 1070.9549,-219.4918 1406.2897,-202.9307 1549.9767,-196.9841"/>
<polygon fill="#191970" stroke="#191970" points="1550.5479,-200.4638 1560.3964,-196.5577 1550.2616,-193.4696 1550.5479,-200.4638"/>
</g>
-<!-- Node52->Node11 -->
+<!-- Node50->Node11 -->
<g id="edge135" class="edge">
-<title>Node52->Node11</title>
+<title>Node50->Node11</title>
<path fill="none" stroke="#191970" d="M2307.9506,-786.5969C2450.4201,-783.2463 2914.8977,-769.9241 3060,-736 3208.608,-701.2563 3236.0459,-662.5098 3377,-604 3415.6488,-587.957 3435.0618,-598.2273 3464,-568 3485.5846,-545.4539 3488,-533.2125 3488,-502 3488,-502 3488,-502 3488,-261.5 3488,-58.1181 2740.7051,-22.8519 2498.6715,-16.7635"/>
<polygon fill="#191970" stroke="#191970" points="2498.6403,-13.2618 2488.5585,-16.5188 2498.4709,-20.2598 2498.6403,-13.2618"/>
</g>
-<!-- Node52->Node18 -->
+<!-- Node50->Node16 -->
<g id="edge154" class="edge">
-<title>Node52->Node18</title>
+<title>Node50->Node16</title>
<path fill="none" stroke="#191970" d="M2307.6585,-786.6576C2455.3663,-783.4059 2953.5154,-770.1093 3109,-736 3175.5981,-721.3901 3482.1313,-612.4099 3540,-568 3644.7932,-487.5791 3716,-460.5952 3716,-328.5 3716,-328.5 3716,-328.5 3716,-133 3716,-101.7875 3716.406,-86.4569 3692,-67 3662.3125,-43.3327 3402.9876,-23.8137 3311.1049,-17.5844"/>
<polygon fill="#191970" stroke="#191970" points="3311.2147,-14.084 3301.003,-16.9076 3310.7466,-21.0684 3311.2147,-14.084"/>
</g>
-<!-- Node52->Node20 -->
+<!-- Node50->Node18 -->
<g id="edge156" class="edge">
-<title>Node52->Node20</title>
+<title>Node50->Node18</title>
<path fill="none" stroke="#191970" d="M2224.0694,-786.5958C1949.8475,-780.6517 422.8084,-747.0745 402,-736 196.4673,-626.6129 180,-494.3288 180,-261.5 180,-261.5 180,-261.5 180,-133 180,-94.2415 203.8303,-87.0498 237,-67 268.5291,-47.9419 371.9501,-29.2424 425.3868,-20.55"/>
<polygon fill="#191970" stroke="#191970" points="426.0524,-23.988 435.3715,-18.9478 424.9433,-17.0764 426.0524,-23.988"/>
</g>
-<!-- Node52->Node23 -->
+<!-- Node50->Node21 -->
<g id="edge136" class="edge">
-<title>Node52->Node23</title>
+<title>Node50->Node21</title>
<path fill="none" stroke="#191970" d="M2224.3964,-781.0094C2053.3501,-753.6192 1406.0904,-642.8435 1252,-512 1176.2054,-447.6401 1127.0676,-335.8648 1107.9832,-286.3424"/>
<polygon fill="#191970" stroke="#191970" points="1111.1725,-284.8803 1104.3682,-276.7592 1104.623,-287.351 1111.1725,-284.8803"/>
</g>
-<!-- Node52->Node24 -->
+<!-- Node50->Node22 -->
<g id="edge151" class="edge">
-<title>Node52->Node24</title>
+<title>Node50->Node22</title>
<path fill="none" stroke="#191970" d="M2224.1294,-786.6022C1963.1614,-780.965 569.3923,-750.2328 528,-736 481.0927,-719.8709 438,-719.6029 438,-670 438,-670 438,-670 438,-502 438,-352.7531 558.4178,-200.7928 602.27,-150.4021"/>
<polygon fill="#191970" stroke="#191970" points="605.0785,-152.5093 609.0738,-142.6966 599.8313,-147.876 605.0785,-152.5093"/>
</g>
-<!-- Node52->Node30 -->
+<!-- Node50->Node28 -->
<g id="edge139" class="edge">
-<title>Node52->Node30</title>
+<title>Node50->Node28</title>
<path fill="none" stroke="#191970" d="M2307.5778,-786.7557C2458.3835,-783.779 2971.1027,-771.1438 3035,-736 3062.0281,-721.1344 3063.6204,-708.2534 3076,-680 3154.8825,-499.9699 3145.0586,-368.0624 2991,-246 2951.9878,-215.0902 2633.5181,-201.4344 2482.2379,-196.6468"/>
<polygon fill="#191970" stroke="#191970" points="2482.1842,-193.1435 2472.0802,-196.3307 2481.9664,-200.1401 2482.1842,-193.1435"/>
</g>
-<!-- Node52->Node35 -->
+<!-- Node50->Node33 -->
<g id="edge138" class="edge">
-<title>Node52->Node35</title>
+<title>Node50->Node33</title>
<path fill="none" stroke="#191970" d="M2224.2339,-786.7984C1992.7045,-782.817 871.7338,-762.3082 720,-736 625.2735,-719.5759 514,-766.1398 514,-670 514,-670 514,-670 514,-614 514,-493.6876 666,-510.3124 666,-390 666,-390 666,-390 666,-328.5 666,-282.0938 690.563,-270.4592 730,-246 759.8294,-227.4995 847.442,-212.4793 911.6731,-203.5333"/>
<polygon fill="#191970" stroke="#191970" points="912.3363,-206.9752 921.7686,-202.1501 911.386,-200.04 912.3363,-206.9752"/>
</g>
-<!-- Node52->Node38 -->
+<!-- Node50->Node36 -->
<g id="edge155" class="edge">
-<title>Node52->Node38</title>
+<title>Node50->Node36</title>
<path fill="none" stroke="#191970" d="M2307.557,-786.0163C2455.5409,-780.5348 2949.2497,-760.4703 2972,-736 2978.0524,-729.49 2975.0406,-724.3527 2972,-716 2924.2669,-584.8766 2872.2303,-562.8271 2752,-492 2653.0026,-433.6811 2628.7969,-413.8559 2519,-380 2416.7687,-348.4769 2385.8387,-365.3024 2281,-344 2226.5971,-332.9457 2214.5761,-323.1652 2160,-313 2092.9056,-300.5032 1911.7697,-314.8818 1855,-277 1813.4542,-249.277 1841.7391,-206.431 1800,-179 1768.3951,-158.2292 1526.3653,-142.339 [...]
<polygon fill="#191970" stroke="#191970" points="1414.0392,-132.4987 1403.8597,-135.4361 1413.6491,-139.4879 1414.0392,-132.4987"/>
</g>
-<!-- Node52->Node44 -->
+<!-- Node50->Node42 -->
<g id="edge137" class="edge">
-<title>Node52->Node44</title>
+<title>Node50->Node42</title>
<path fill="none" stroke="#191970" d="M2224.0862,-786.6487C1951.6583,-781.077 442.8805,-749.6217 424,-736 398.6876,-717.7379 400,-701.2125 400,-670 400,-670 400,-670 400,-614 400,-581.2634 363.7922,-361.8272 351.2309,-286.6845"/>
<polygon fill="#191970" stroke="#191970" points="354.6599,-285.9689 349.5563,-276.6843 347.756,-287.125 354.6599,-285.9689"/>
</g>
-<!-- Node52->Node46 -->
+<!-- Node50->Node44 -->
<g id="edge153" class="edge">
-<title>Node52->Node46</title>
+<title>Node50->Node44</title>
<path fill="none" stroke="#191970" d="M2307.7209,-787.0313C2455.9914,-785.0384 2951.7426,-775.4573 3010,-736 3162.239,-632.8896 3002.3198,-354.0047 2955.8509,-279.828"/>
<polygon fill="#191970" stroke="#191970" points="2958.6307,-277.6762 2950.3165,-271.1093 2952.7208,-281.4277 2958.6307,-277.6762"/>
</g>
-<!-- Node52->Node49 -->
+<!-- Node50->Node47 -->
<g id="edge152" class="edge">
-<title>Node52->Node49</title>
+<title>Node50->Node47</title>
<path fill="none" stroke="#191970" d="M2277.0474,-777.6948C2299.0895,-756.8794 2345.1624,-706.3765 2330,-660 2301.2429,-572.0421 2278.314,-550.9735 2207,-492 2162.2643,-455.0055 2144.1892,-454.0007 2089,-436 1880.4151,-367.9674 1815.8506,-401.0562 1604,-344 1600.8719,-343.1575 1597.647,-342.1999 1594.4321,-341.1837"/>
<polygon fill="#191970" stroke="#191970" points="1595.5173,-337.8562 1584.9237,-338.0158 1593.3047,-344.4973 1595.5173,-337.8562"/>
</g>
-<!-- Node53 -->
+<!-- Node51 -->
<g id="node37" class="node">
-<title>Node53</title>
+<title>Node51</title>
<g id="a_node37"><a xlink:href="buffer_8h.html" target="_top" xlink:title="Symbolic n-dimensional array, to represent a memory buffer. ">
<polygon fill="#ffffff" stroke="#000000" points="2129.5,-716.5 2129.5,-735.5 2218.5,-735.5 2218.5,-716.5 2129.5,-716.5"/>
<text text-anchor="middle" x="2174" y="-723.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/tir/buffer.h</text>
</a>
</g>
</g>
-<!-- Node52->Node53 -->
+<!-- Node50->Node51 -->
<g id="edge140" class="edge">
-<title>Node52->Node53</title>
+<title>Node50->Node51</title>
<path fill="none" stroke="#191970" d="M2251.625,-777.8906C2236.9069,-768.0519 2213.8168,-752.6167 2196.6411,-741.1351"/>
<polygon fill="#191970" stroke="#191970" points="2198.5327,-738.1896 2188.2741,-735.5419 2194.6425,-744.0091 2198.5327,-738.1896"/>
</g>
-<!-- Node54 -->
+<!-- Node52 -->
<g id="node38" class="node">
-<title>Node54</title>
+<title>Node52</title>
<g id="a_node38"><a xlink:href="var_8h.html" target="_top" xlink:title="Variables in the TIR. ">
<polygon fill="#ffffff" stroke="#000000" points="2243.5,-660.5 2243.5,-679.5 2320.5,-679.5 2320.5,-660.5 2243.5,-660.5"/>
<text text-anchor="middle" x="2282" y="-667.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/tir/var.h</text>
</a>
</g>
</g>
-<!-- Node52->Node54 -->
+<!-- Node50->Node52 -->
<g id="edge150" class="edge">
-<title>Node52->Node54</title>
+<title>Node50->Node52</title>
<path fill="none" stroke="#191970" d="M2267.3093,-777.8845C2269.9229,-758.6913 2275.8334,-715.286 2279.3084,-689.7663"/>
<polygon fill="#191970" stroke="#191970" points="2282.7912,-690.1292 2280.6726,-679.7484 2275.8552,-689.1847 2282.7912,-690.1292"/>
</g>
-<!-- Node53->Node3 -->
+<!-- Node51->Node3 -->
<g id="edge141" class="edge">
-<title>Node53->Node3</title>
+<title>Node51->Node3</title>
<path fill="none" stroke="#191970" d="M2168.1615,-716.0996C2158.8915,-701.3285 2139.339,-673.7858 2115,-660 2083.1573,-641.964 1985.2372,-627.3463 1924.7606,-619.7307"/>
<polygon fill="#191970" stroke="#191970" points="1925.015,-616.2356 1914.6607,-618.48 1924.1547,-623.1825 1925.015,-616.2356"/>
</g>
-<!-- Node53->Node18 -->
+<!-- Node51->Node16 -->
<g id="edge149" class="edge">
-<title>Node53->Node18</title>
+<title>Node51->Node16</title>
<path fill="none" stroke="#191970" d="M2218.6978,-724.6236C2315.8646,-721.1521 2553.4046,-710.0665 2750,-680 2763.8552,-677.881 3229.8251,-574.9446 3242,-568 3365.6955,-497.444 3322.103,-398.4763 3436,-313 3475.1217,-283.6403 3500.4707,-305.8084 3540,-277 3577.0673,-249.9858 3602,-240.3667 3602,-194.5 3602,-194.5 3602,-194.5 3602,-133 3602,-92.7538 3575.2082,-86.497 3540,-67 3500.9309,-45.365 3371.9455,-27.0279 3311.1891,-19.3525"/>
<polygon fill="#191970" stroke="#191970" points="3311.3732,-15.8485 3301.0175,-18.0868 3310.5088,-22.7949 3311.3732,-15.8485"/>
</g>
-<!-- Node53->Node23 -->
+<!-- Node51->Node21 -->
<g id="edge142" class="edge">
-<title>Node53->Node23</title>
+<title>Node51->Node21</title>
<path fill="none" stroke="#191970" d="M2170.6942,-716.2706C2165.7457,-702.6693 2155.4272,-677.6591 2141,-660 2089.9244,-597.4826 2071.3631,-581.6896 1998,-548 1807.7375,-460.6281 1738.2267,-510.2037 1536,-456 1461.1985,-435.9506 1442.3562,-430.0919 1371,-400 1369.5111,-399.3721 1212.2254,-319.2153 1137.5414,-281.147"/>
<polygon fill="#191970" stroke="#191970" points="1138.9295,-277.9261 1128.4307,-276.503 1135.7505,-284.1627 1138.9295,-277.9261"/>
</g>
-<!-- Node53->Node35 -->
+<!-- Node51->Node33 -->
<g id="edge143" class="edge">
-<title>Node53->Node35</title>
+<title>Node51->Node33</title>
<path fill="none" stroke="#191970" d="M2129.1188,-725.5832C2001.3756,-723.9696 1625.0205,-716.2476 1315,-680 1176.3739,-663.7918 1137.7376,-670.099 1006,-624 850.4305,-569.5615 768.5662,-473.9655 804,-313 811.0538,-280.9565 810.6398,-267.9801 835,-246 856.3926,-226.6975 885.3608,-214.5703 912.1364,-206.9775"/>
<polygon fill="#191970" stroke="#191970" points="913.122,-210.3371 921.8838,-204.3805 911.3197,-203.5731 913.122,-210.3371"/>
</g>
-<!-- Node53->Node54 -->
+<!-- Node51->Node52 -->
<g id="edge144" class="edge">
-<title>Node53->Node54</title>
+<title>Node51->Node52</title>
<path fill="none" stroke="#191970" d="M2192.8122,-716.2455C2209.8473,-707.4125 2235.1575,-694.2887 2254.4681,-684.2758"/>
<polygon fill="#191970" stroke="#191970" points="2256.137,-687.353 2263.4034,-679.6427 2252.9147,-681.1388 2256.137,-687.353"/>
</g>
-<!-- Node54->Node3 -->
+<!-- Node52->Node3 -->
<g id="edge145" class="edge">
-<title>Node54->Node3</title>
+<title>Node52->Node3</title>
<path fill="none" stroke="#191970" d="M2243.3435,-664.6812C2169.1315,-654.4702 2006.7932,-632.1337 1924.6762,-620.835"/>
<polygon fill="#191970" stroke="#191970" points="1924.9769,-617.3435 1914.5932,-619.4477 1924.0227,-624.2782 1924.9769,-617.3435"/>
</g>
-<!-- Node54->Node5 -->
+<!-- Node52->Node5 -->
<g id="edge146" class="edge">
-<title>Node54->Node5</title>
+<title>Node52->Node5</title>
<path fill="none" stroke="#191970" d="M2277.2155,-660.4325C2262.1126,-631.1371 2212.5443,-541.6474 2147,-492 2126.8431,-476.7319 2100.9678,-465.862 2078.6782,-458.5364"/>
<polygon fill="#191970" stroke="#191970" points="2079.6378,-455.1693 2069.0486,-455.5143 2077.5417,-461.8481 2079.6378,-455.1693"/>
</g>
-<!-- Node54->Node18 -->
+<!-- Node52->Node16 -->
<g id="edge148" class="edge">
-<title>Node54->Node18</title>
+<title>Node52->Node16</title>
<path fill="none" stroke="#191970" d="M2320.5881,-662.8409C2391.549,-649.8457 2547.6223,-622.0524 2680,-604 2828.8107,-583.7066 2876.4953,-623.6316 3016,-568 3073.1378,-545.2146 3194.7022,-445.5894 3236,-400 3279.142,-352.3749 3285.8981,-335.7203 3312,-277 3327.5111,-242.1055 3336,-232.6867 3336,-194.5 3336,-194.5 3336,-194.5 3336,-133 3336,-94.341 3310.7138,-55.0543 3293.825,-33.1259"/>
<polygon fill="#191970" stroke="#191970" points="3296.3871,-30.7255 3287.4105,-25.0977 3290.9183,-35.095 3296.3871,-30.7255"/>
</g>
-<!-- Node54->Node30 -->
+<!-- Node52->Node28 -->
<g id="edge147" class="edge">
-<title>Node54->Node30</title>
+<title>Node52->Node28</title>
<path fill="none" stroke="#191970" d="M2302.6756,-660.427C2397.688,-616.3775 2789.9157,-433.8829 2840,-400 2907.2297,-354.518 2942.2037,-351.2486 2975,-277 2980.5669,-264.3969 2984.2953,-256.1698 2975,-246 2942.409,-210.3428 2631.3607,-199.1585 2482.1279,-195.8285"/>
<polygon fill="#191970" stroke="#191970" points="2482.1762,-192.3288 2472.1027,-195.6112 2482.0245,-199.3272 2482.1762,-192.3288"/>
</g>
-<!-- Node55->Node1 -->
+<!-- Node53->Node1 -->
<g id="edge160" class="edge">
-<title>Node55->Node1</title>
+<title>Node53->Node1</title>
<path fill="none" stroke="#191970" d="M3242.6244,-958.0776C3333.6911,-952.106 3522.5678,-937.1699 3542,-915 3564.2646,-889.5987 3555.9004,-869.785 3542,-839 3504.3591,-755.637 3464.0274,-744.1506 3377,-716 3314.0619,-695.6415 2863.1303,-677.4379 2710.4149,-671.7963"/>
<polygon fill="#191970" stroke="#191970" points="2710.4905,-668.2968 2700.3688,-671.4275 2710.2336,-675.2921 2710.4905,-668.2968"/>
</g>
-<!-- Node55->Node3 -->
+<!-- Node53->Node3 -->
<g id="edge161" class="edge">
-<title>Node55->Node3</title>
+<title>Node53->Node3</title>
<path fill="none" stroke="#191970" d="M3145.2318,-959.5878C2902.3646,-952.2416 1835,-916.1372 1835,-849 1835,-849 1835,-849 1835,-726 1835,-691.4432 1852.097,-654.1996 1863.9358,-632.5529"/>
<polygon fill="#191970" stroke="#191970" points="1867.0287,-634.1941 1868.9249,-623.7703 1860.9422,-630.7365 1867.0287,-634.1941"/>
</g>
-<!-- Node55->Node18 -->
+<!-- Node53->Node16 -->
<g id="edge214" class="edge">
-<title>Node55->Node18</title>
+<title>Node53->Node16</title>
<path fill="none" stroke="#191970" d="M3242.7436,-958.0522C3397.5646,-948.0461 3868,-912.4238 3868,-849 3868,-849 3868,-849 3868,-133 3868,-98.6999 3857.0658,-85.212 3828,-67 3784.1338,-39.5144 3422.476,-21.71 3311.3533,-16.8486"/>
<polygon fill="#191970" stroke="#191970" points="3311.2483,-13.3409 3301.1066,-16.4061 3310.9463,-20.3344 3311.2483,-13.3409"/>
</g>
-<!-- Node55->Node20 -->
+<!-- Node53->Node18 -->
<g id="edge215" class="edge">
-<title>Node55->Node20</title>
+<title>Node53->Node18</title>
<path fill="none" stroke="#191970" d="M3145.4251,-960.6123C2744.7084,-957.2602 0,-931.2962 0,-849 0,-849 0,-849 0,-133 0,-93.9495 23.5249,-85.3415 58,-67 121.5856,-33.1711 341.3069,-20.4541 425.1436,-16.7642"/>
<polygon fill="#191970" stroke="#191970" points="425.5179,-20.2516 435.3602,-16.33 425.2206,-13.2579 425.5179,-20.2516"/>
</g>
-<!-- Node55->Node42 -->
+<!-- Node53->Node40 -->
<g id="edge212" class="edge">
-<title>Node55->Node42</title>
+<title>Node53->Node40</title>
<path fill="none" stroke="#191970" d="M3242.5381,-951.6107C3316.2033,-935.2445 3442.8636,-897.5451 3395,-839 3169.9638,-563.7438 2952.9252,-777.4396 2611,-680 2593.7907,-675.0958 2476.1903,-631.6214 2460,-624 2354.0604,-574.13 2333.6257,-550.1587 2232,-492 2188.1251,-466.8912 2180.1592,-454.2063 2133,-436 2080.2928,-415.6518 1940.4177,-390.9991 1885,-380 1838.5585,-370.7825 1705.5205,-380.1978 1675,-344 1638.4011,-300.5931 1702.9496,-238.8911 1738.2545,-210.242"/>
<polygon fill="#191970" stroke="#191970" points="1740.4493,-212.9685 1746.1145,-204.0154 1736.1026,-207.4815 1740.4493,-212.9685"/>
</g>
-<!-- Node55->Node51 -->
+<!-- Node53->Node49 -->
<g id="edge216" class="edge">
-<title>Node55->Node51</title>
+<title>Node53->Node49</title>
<path fill="none" stroke="#191970" d="M3189.701,-951.41C3184.1992,-937.5256 3176.7513,-911.6674 3189,-895 3199.7475,-880.3754 3240.4255,-868.3271 3276.2725,-860.2345"/>
<polygon fill="#191970" stroke="#191970" points="3277.2047,-863.6135 3286.2262,-858.058 3275.7094,-856.775 3277.2047,-863.6135"/>
</g>
-<!-- Node56 -->
+<!-- Node54 -->
<g id="node40" class="node">
-<title>Node56</title>
+<title>Node54</title>
<g id="a_node40"><a xlink:href="ir_2module_8h.html" target="_top" xlink:title="IRModule that holds the functions and type definitions. ">
<polygon fill="#ffffff" stroke="#000000" points="1378.5,-778 1378.5,-797 1473.5,-797 1473.5,-778 1378.5,-778"/>
<text text-anchor="middle" x="1426" y="-785" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/module.h</text>
</a>
</g>
</g>
-<!-- Node55->Node56 -->
+<!-- Node53->Node54 -->
<g id="edge162" class="edge">
-<title>Node55->Node56</title>
+<title>Node53->Node54</title>
<path fill="none" stroke="#191970" d="M3145.4373,-960.1749C2916.8206,-956.1825 1953.3782,-938.1134 1821,-915 1681.5705,-890.6554 1523.5686,-828.6113 1457.573,-801.049"/>
<polygon fill="#191970" stroke="#191970" points="1458.7266,-797.7372 1448.1517,-797.0866 1456.0128,-804.1898 1458.7266,-797.7372"/>
</g>
-<!-- Node64 -->
+<!-- Node62 -->
<g id="node45" class="node">
-<title>Node64</title>
+<title>Node62</title>
<g id="a_node45"><a xlink:href="ir_2op_8h.html" target="_top" xlink:title="Primitive operators(builtin intrinsics) and registry for them. ">
<polygon fill="#ffffff" stroke="#ff0000" points="3020,-895.5 3020,-914.5 3090,-914.5 3090,-895.5 3020,-895.5"/>
<text text-anchor="middle" x="3055" y="-902.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/op.h</text>
</a>
</g>
</g>
-<!-- Node55->Node64 -->
+<!-- Node53->Node62 -->
<g id="edge195" class="edge">
-<title>Node55->Node64</title>
+<title>Node53->Node62</title>
<path fill="none" stroke="#191970" d="M3170.105,-951.3733C3147.4918,-942.2629 3113.3458,-928.5062 3088.1127,-918.3404"/>
<polygon fill="#191970" stroke="#191970" points="3089.1666,-914.9917 3078.5831,-914.5011 3086.5507,-921.4845 3089.1666,-914.9917"/>
</g>
-<!-- Node71 -->
+<!-- Node69 -->
<g id="node48" class="node">
-<title>Node71</title>
+<title>Node69</title>
<g id="a_node48"><a xlink:href="virtual__device_8h.html" target="_top" xlink:title="A compile time representation for where data is to be stored at runtime, and how to compile code to c...">
<polygon fill="#ffffff" stroke="#ff0000" points="676.5,-772.5 676.5,-802.5 779.5,-802.5 779.5,-772.5 676.5,-772.5"/>
<text text-anchor="start" x="684.5" y="-790.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/target/virtual</text>
@@ -1362,402 +1362,402 @@
</a>
</g>
</g>
-<!-- Node55->Node71 -->
+<!-- Node53->Node69 -->
<g id="edge208" class="edge">
-<title>Node55->Node71</title>
+<title>Node53->Node69</title>
<path fill="none" stroke="#191970" d="M3145.145,-960.091C2923.8822,-955.878 2014.236,-937.498 1730,-915 1372.274,-886.6851 946.2115,-822.1706 789.6306,-797.4203"/>
<polygon fill="#191970" stroke="#191970" points="790.1347,-793.9566 779.7102,-795.8483 789.0392,-800.8703 790.1347,-793.9566"/>
</g>
-<!-- Node78 -->
+<!-- Node76 -->
<g id="node49" class="node">
-<title>Node78</title>
+<title>Node76</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="3197.5,-895.5 3197.5,-914.5 3240.5,-914.5 3240.5,-895.5 3197.5,-895.5"/>
<text text-anchor="middle" x="3219" y="-902.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stack</text>
</g>
-<!-- Node55->Node78 -->
+<!-- Node53->Node76 -->
<g id="edge213" class="edge">
-<title>Node55->Node78</title>
+<title>Node53->Node76</title>
<path fill="none" stroke="#191970" d="M3198.3547,-951.2455C3201.7244,-943.6973 3206.493,-933.0158 3210.5983,-923.8197"/>
<polygon fill="#191970" stroke="#191970" points="3213.8147,-925.2009 3214.6952,-914.6427 3207.4227,-922.3473 3213.8147,-925.2009"/>
</g>
-<!-- Node79 -->
+<!-- Node77 -->
<g id="node50" class="node">
-<title>Node79</title>
+<title>Node77</title>
<g id="a_node50"><a xlink:href="relay_2type_8h.html" target="_top" xlink:title="Relay typed AST nodes. ">
<polygon fill="#ffffff" stroke="#ff0000" points="3258.5,-895.5 3258.5,-914.5 3311.5,-914.5 3311.5,-895.5 3258.5,-895.5"/>
<text text-anchor="middle" x="3285" y="-902.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">./type.h</text>
</a>
</g>
</g>
-<!-- Node55->Node79 -->
+<!-- Node53->Node77 -->
<g id="edge217" class="edge">
-<title>Node55->Node79</title>
+<title>Node53->Node77</title>
<path fill="none" stroke="#191970" d="M3209.851,-951.2455C3223.8132,-942.6534 3244.3728,-930.0014 3260.4596,-920.1018"/>
<polygon fill="#191970" stroke="#191970" points="3262.6484,-922.8645 3269.3307,-914.6427 3258.9797,-916.9029 3262.6484,-922.8645"/>
</g>
-<!-- Node56->Node3 -->
+<!-- Node54->Node3 -->
<g id="edge171" class="edge">
-<title>Node56->Node3</title>
+<title>Node54->Node3</title>
<path fill="none" stroke="#191970" d="M1473.6406,-780.4913C1519.9569,-772.7787 1591.7856,-758.4148 1651,-736 1728.6408,-706.6102 1812.6931,-654.8397 1852.1298,-629.2286"/>
<polygon fill="#191970" stroke="#191970" points="1854.2794,-632.0046 1860.7281,-623.5983 1850.4447,-626.1484 1854.2794,-632.0046"/>
</g>
-<!-- Node56->Node18 -->
+<!-- Node54->Node16 -->
<g id="edge190" class="edge">
-<title>Node56->Node18</title>
+<title>Node54->Node16</title>
<path fill="none" stroke="#191970" d="M1473.5314,-786.4138C1714.1535,-780.6242 2799.3426,-750.9576 3128,-680 3222.7047,-659.5531 3526,-598.8869 3526,-502 3526,-502 3526,-502 3526,-390 3526,-314.1144 3613.3487,-341.0788 3654,-277 3674.4563,-244.7547 3678,-232.6867 3678,-194.5 3678,-194.5 3678,-194.5 3678,-133 3678,-101.7875 3678.2847,-86.6081 3654,-67 3627.4613,-45.572 3397.3141,-25.0269 3311.2736,-18.0269"/>
<polygon fill="#191970" stroke="#191970" points="3311.3005,-14.5178 3301.0519,-17.204 3310.7388,-21.4952 3311.3005,-14.5178"/>
</g>
-<!-- Node56->Node20 -->
+<!-- Node54->Node18 -->
<g id="edge193" class="edge">
-<title>Node56->Node20</title>
+<title>Node54->Node18</title>
<path fill="none" stroke="#191970" d="M1378.4721,-786.5239C1168.4362,-782.0619 331.7355,-762.6722 283,-736 90.0091,-630.3793 142,-481.5028 142,-261.5 142,-261.5 142,-261.5 142,-133 142,-101.7875 142.0932,-87.0671 166,-67 204.9228,-34.3287 357.4929,-21.4195 425.2598,-17.2229"/>
<polygon fill="#191970" stroke="#191970" points="425.6024,-20.7089 435.3775,-16.6225 425.1877,-13.7212 425.6024,-20.7089"/>
</g>
-<!-- Node56->Node22 -->
+<!-- Node54->Node20 -->
<g id="edge194" class="edge">
-<title>Node56->Node22</title>
+<title>Node54->Node20</title>
<path fill="none" stroke="#191970" d="M1473.5077,-786.3235C1654.0031,-781.6828 2297.2221,-763.5422 2500,-736 2611.9692,-720.7918 2646.1739,-728.9882 2748,-680 2957.2758,-579.318 3096.9689,-470.8841 3039,-246 3030.509,-213.0601 3029.7587,-198.6619 3002,-179 2963.8753,-151.9958 2826.5567,-139.5203 2761.9949,-135.0483"/>
<polygon fill="#191970" stroke="#191970" points="2762.1297,-131.5496 2751.9185,-134.3748 2761.6628,-138.534 2762.1297,-131.5496"/>
</g>
-<!-- Node56->Node23 -->
+<!-- Node54->Node21 -->
<g id="edge187" class="edge">
-<title>Node56->Node23</title>
+<title>Node54->Node21</title>
<path fill="none" stroke="#191970" d="M1378.2701,-778.4602C1289.8426,-758.7023 1103.9159,-702.228 1042,-568 996.8614,-470.1437 1056.4173,-339.0988 1085.1177,-285.6859"/>
<polygon fill="#191970" stroke="#191970" points="1088.3389,-287.0913 1090.0854,-276.6414 1082.2034,-283.7214 1088.3389,-287.0913"/>
</g>
-<!-- Node56->Node35 -->
+<!-- Node54->Node33 -->
<g id="edge189" class="edge">
-<title>Node56->Node35</title>
+<title>Node54->Node33</title>
<path fill="none" stroke="#191970" d="M1378.1922,-786.339C1181.6658,-781.4077 444.3677,-761.2867 402,-736 372.5469,-718.4213 362,-704.3001 362,-670 362,-670 362,-670 362,-614 362,-405.7627 472.1646,-350.9858 652,-246 674.0347,-233.1364 820.6889,-213.8878 911.8768,-202.9309"/>
<polygon fill="#191970" stroke="#191970" points="912.3746,-206.3964 921.8886,-201.7344 911.5439,-199.4459 912.3746,-206.3964"/>
</g>
-<!-- Node56->Node38 -->
+<!-- Node54->Node36 -->
<g id="edge191" class="edge">
-<title>Node56->Node38</title>
+<title>Node54->Node36</title>
<path fill="none" stroke="#191970" d="M1424.5628,-777.8572C1414.9285,-713.2157 1360.1149,-345.4154 1360,-344 1354.3067,-273.8444 1355.4692,-190.1486 1356.4038,-152.7501"/>
<polygon fill="#191970" stroke="#191970" points="1359.904,-152.7882 1356.6787,-142.6962 1352.9066,-152.5968 1359.904,-152.7882"/>
</g>
-<!-- Node56->Node44 -->
+<!-- Node54->Node42 -->
<g id="edge188" class="edge">
-<title>Node56->Node44</title>
+<title>Node54->Node42</title>
<path fill="none" stroke="#191970" d="M1378.2127,-786.8099C1182.2642,-783.7644 446.041,-770.1324 348,-736 300.4753,-719.4546 256,-720.3224 256,-670 256,-670 256,-670 256,-390 256,-346.1504 291.7276,-306.8014 318.4099,-283.5289"/>
<polygon fill="#191970" stroke="#191970" points="320.9605,-285.9565 326.3507,-276.8353 316.4489,-280.6043 320.9605,-285.9565"/>
</g>
-<!-- Node56->Node50 -->
+<!-- Node54->Node48 -->
<g id="edge180" class="edge">
-<title>Node56->Node50</title>
+<title>Node54->Node48</title>
<path fill="none" stroke="#191970" d="M1445.7532,-777.8209C1503.5922,-749.6141 1677.8462,-665.592 1826,-604 1854.4941,-592.1542 1887.2412,-579.9756 1911.7787,-571.1275"/>
<polygon fill="#191970" stroke="#191970" points="1913.1894,-574.34 1921.4208,-567.6696 1910.8263,-567.7509 1913.1894,-574.34"/>
</g>
-<!-- Node57 -->
+<!-- Node55 -->
<g id="node41" class="node">
-<title>Node57</title>
+<title>Node55</title>
<g id="a_node41"><a xlink:href="ir_2adt_8h.html" target="_top" xlink:title="Algebraic data type definitions. ">
<polygon fill="#ffffff" stroke="#ff0000" points="1324,-660.5 1324,-679.5 1398,-679.5 1398,-660.5 1324,-660.5"/>
<text text-anchor="middle" x="1361" y="-667.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/adt.h</text>
</a>
</g>
</g>
-<!-- Node56->Node57 -->
+<!-- Node54->Node55 -->
<g id="edge163" class="edge">
-<title>Node56->Node57</title>
+<title>Node54->Node55</title>
<path fill="none" stroke="#191970" d="M1420.6808,-777.8845C1409.9205,-758.4331 1385.4034,-714.1138 1371.3702,-688.7462"/>
<polygon fill="#191970" stroke="#191970" points="1374.296,-686.8045 1366.3927,-679.7484 1368.1708,-690.193 1374.296,-686.8045"/>
</g>
-<!-- Node59 -->
+<!-- Node57 -->
<g id="node42" class="node">
-<title>Node59</title>
+<title>Node57</title>
<g id="a_node42"><a xlink:href="ir_2function_8h.html" target="_top" xlink:title="Function nodes. ">
<polygon fill="#ffffff" stroke="#000000" points="750,-716.5 750,-735.5 848,-735.5 848,-716.5 750,-716.5"/>
<text text-anchor="middle" x="799" y="-723.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/function.h</text>
</a>
</g>
</g>
-<!-- Node56->Node59 -->
+<!-- Node54->Node57 -->
<g id="edge172" class="edge">
-<title>Node56->Node59</title>
+<title>Node54->Node57</title>
<path fill="none" stroke="#191970" d="M1378.4816,-783.4816C1283.6296,-775.3578 1064.1004,-756.0424 880,-736 872.9306,-735.2304 865.5199,-734.3782 858.1734,-733.5054"/>
<polygon fill="#191970" stroke="#191970" points="858.5802,-730.0291 848.2334,-732.3082 857.7431,-736.9789 858.5802,-730.0291"/>
</g>
-<!-- Node60 -->
+<!-- Node58 -->
<g id="node43" class="node">
-<title>Node60</title>
+<title>Node58</title>
<g id="a_node43"><a xlink:href="source__map_8h.html" target="_top" xlink:title="A map from source names to source code. ">
<polygon fill="#ffffff" stroke="#ff0000" points="2680.5,-548.5 2680.5,-567.5 2821.5,-567.5 2821.5,-548.5 2680.5,-548.5"/>
<text text-anchor="middle" x="2751" y="-555.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/parser/source_map.h</text>
</a>
</g>
</g>
-<!-- Node56->Node60 -->
+<!-- Node54->Node58 -->
<g id="edge181" class="edge">
-<title>Node56->Node60</title>
+<title>Node54->Node58</title>
<path fill="none" stroke="#191970" d="M1473.825,-786.8685C1677.8271,-783.6619 2475.442,-765.7725 2709,-680 2751.1187,-664.5322 2771.8282,-663.5599 2793,-624 2802.4087,-606.4198 2787.2126,-587.2971 2772.5384,-574.183"/>
<polygon fill="#191970" stroke="#191970" points="2774.5212,-571.2792 2764.5993,-567.5635 2770.0385,-576.6556 2774.5212,-571.2792"/>
</g>
-<!-- Node63 -->
+<!-- Node61 -->
<g id="node44" class="node">
-<title>Node63</title>
+<title>Node61</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="104,-716.5 104,-735.5 190,-735.5 190,-716.5 104,-716.5"/>
<text text-anchor="middle" x="147" y="-723.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">unordered_set</text>
</g>
-<!-- Node56->Node63 -->
+<!-- Node54->Node61 -->
<g id="edge192" class="edge">
-<title>Node56->Node63</title>
+<title>Node54->Node61</title>
<path fill="none" stroke="#191970" d="M1378.4713,-786.3343C1182.6829,-781.4028 437.9322,-761.3386 204,-736 202.7358,-735.8631 201.4573,-735.7152 200.169,-735.5579"/>
<polygon fill="#191970" stroke="#191970" points="200.5573,-732.0785 190.1749,-734.1894 199.6076,-739.0138 200.5573,-732.0785"/>
</g>
-<!-- Node57->Node3 -->
+<!-- Node55->Node3 -->
<g id="edge164" class="edge">
-<title>Node57->Node3</title>
+<title>Node55->Node3</title>
<path fill="none" stroke="#191970" d="M1398.3796,-665.9275C1489.393,-656.0117 1722.5428,-630.6101 1825.2011,-619.4256"/>
<polygon fill="#191970" stroke="#191970" points="1825.7431,-622.8873 1835.3052,-618.3247 1824.9849,-615.9285 1825.7431,-622.8873"/>
</g>
-<!-- Node57->Node5 -->
+<!-- Node55->Node5 -->
<g id="edge166" class="edge">
-<title>Node57->Node5</title>
+<title>Node55->Node5</title>
<path fill="none" stroke="#191970" d="M1384.3428,-660.4686C1417.1091,-647.2092 1478.7078,-622.7188 1532,-604 1699.3161,-545.2305 1901.529,-484.0731 1987.75,-458.4523"/>
<polygon fill="#191970" stroke="#191970" points="1988.8451,-461.7783 1997.4366,-455.5785 1986.8541,-455.0674 1988.8451,-461.7783"/>
</g>
-<!-- Node57->Node10 -->
+<!-- Node55->Node10 -->
<g id="edge169" class="edge">
-<title>Node57->Node10</title>
+<title>Node55->Node10</title>
<path fill="none" stroke="#191970" d="M1323.769,-667.6377C1258.6385,-662.9611 1120.0904,-650.5877 1006,-624 762.7785,-567.3194 512.7632,-561.4602 538,-313 546.7113,-227.2362 511.6398,-179.2081 577,-123 607.5861,-96.6968 1184.6811,-82.5294 1387.9101,-78.34"/>
<polygon fill="#191970" stroke="#191970" points="1388.2405,-81.834 1398.1669,-78.1305 1388.0974,-74.8355 1388.2405,-81.834"/>
</g>
-<!-- Node57->Node18 -->
+<!-- Node55->Node16 -->
<g id="edge170" class="edge">
-<title>Node57->Node18</title>
+<title>Node55->Node16</title>
<path fill="none" stroke="#191970" d="M1398.3594,-667.9671C1623.2932,-655.6638 2790.0645,-590.9265 2862,-568 2983.555,-529.2593 3298,-322.0792 3298,-194.5 3298,-194.5 3298,-194.5 3298,-133 3298,-98.1936 3289.7747,-58.3226 3284.1479,-35.1197"/>
<polygon fill="#191970" stroke="#191970" points="3287.477,-34.0065 3281.6419,-25.1633 3280.6887,-35.7152 3287.477,-34.0065"/>
</g>
-<!-- Node57->Node23 -->
+<!-- Node55->Node21 -->
<g id="edge167" class="edge">
-<title>Node57->Node23</title>
+<title>Node55->Node21</title>
<path fill="none" stroke="#191970" d="M1323.9392,-664.8369C1240.6715,-652.198 1046,-616.4042 1046,-558 1046,-558 1046,-558 1046,-502 1046,-459.7004 1077.6627,-339.1125 1092.0425,-286.5281"/>
<polygon fill="#191970" stroke="#191970" points="1095.4749,-287.2462 1094.7524,-276.676 1088.7255,-285.3897 1095.4749,-287.2462"/>
</g>
-<!-- Node57->Node35 -->
+<!-- Node55->Node33 -->
<g id="edge168" class="edge">
-<title>Node57->Node35</title>
+<title>Node55->Node33</title>
<path fill="none" stroke="#191970" d="M1323.6774,-668.3023C1200.9983,-660.5393 818,-618.0422 818,-390 818,-390 818,-390 818,-328.5 818,-270.6501 879.7156,-233.6149 928.1118,-213.4662"/>
<polygon fill="#191970" stroke="#191970" points="929.6152,-216.6343 937.5912,-209.6605 927.0071,-210.1382 929.6152,-216.6343"/>
</g>
-<!-- Node57->Node50 -->
+<!-- Node55->Node48 -->
<g id="edge165" class="edge">
-<title>Node57->Node50</title>
+<title>Node55->Node48</title>
<path fill="none" stroke="#191970" d="M1394.2432,-660.4121C1443.7419,-646.4131 1539.9124,-620.2791 1623,-604 1719.8358,-585.0272 1834.3223,-570.7241 1898.6145,-563.4209"/>
<polygon fill="#191970" stroke="#191970" points="1899.3277,-566.8628 1908.8736,-562.2666 1898.545,-559.9067 1899.3277,-566.8628"/>
</g>
-<!-- Node59->Node1 -->
+<!-- Node57->Node1 -->
<g id="edge173" class="edge">
-<title>Node59->Node1</title>
+<title>Node57->Node1</title>
<path fill="none" stroke="#191970" d="M848.2388,-724.5252C1038.8907,-718.8138 1746.9066,-697.5945 2330,-680 2429.0439,-677.0114 2544.8083,-673.4989 2609.5439,-671.533"/>
<polygon fill="#191970" stroke="#191970" points="2609.9802,-675.0215 2619.8693,-671.2194 2609.7677,-668.0247 2609.9802,-675.0215"/>
</g>
-<!-- Node59->Node3 -->
+<!-- Node57->Node3 -->
<g id="edge174" class="edge">
-<title>Node59->Node3</title>
+<title>Node57->Node3</title>
<path fill="none" stroke="#191970" d="M848.3548,-723.3497C952.0204,-717.5445 1199.9351,-702.4101 1407,-680 1560.0584,-663.4349 1740.1474,-635.6993 1825.4307,-622.0649"/>
<polygon fill="#191970" stroke="#191970" points="1826.0333,-625.5131 1835.3529,-620.4738 1824.9249,-618.6014 1826.0333,-625.5131"/>
</g>
-<!-- Node59->Node18 -->
+<!-- Node57->Node16 -->
<g id="edge178" class="edge">
-<title>Node59->Node18</title>
+<title>Node57->Node16</title>
<path fill="none" stroke="#191970" d="M848.3656,-725.4067C1138.8594,-721.8404 2616.0458,-702.6452 2709,-680 2728.6985,-675.2011 2730.9861,-667.038 2750,-660 2809.4443,-637.9966 2826.8117,-640.546 2888,-624 2977.1072,-599.9044 3002.8387,-603.6125 3088,-568 3252.5901,-499.1722 3267.7743,-434.9361 3398,-313 3454.59,-260.0123 3526,-272.025 3526,-194.5 3526,-194.5 3526,-194.5 3526,-133 3526,-86.009 3378.3744,-41.4482 3311.3801,-23.6453"/>
<polygon fill="#191970" stroke="#191970" points="3311.9156,-20.1677 3301.3552,-21.0216 3310.1432,-26.9396 3311.9156,-20.1677"/>
</g>
-<!-- Node59->Node19 -->
+<!-- Node57->Node17 -->
<g id="edge179" class="edge">
-<title>Node59->Node19</title>
+<title>Node57->Node17</title>
<path fill="none" stroke="#191970" d="M785.9993,-716.4612C743.9466,-684.8167 610.8743,-578.3407 551,-456 496.768,-345.1883 504.6552,-299.0705 533,-179 539.3932,-151.9178 540.3057,-141.6029 561,-123 636.1511,-55.4438 950.6866,-26.8522 1068.3358,-18.3823"/>
<polygon fill="#191970" stroke="#191970" points="1068.7091,-21.8648 1078.4381,-17.6699 1068.2166,-14.8821 1068.7091,-21.8648"/>
</g>
-<!-- Node59->Node23 -->
+<!-- Node57->Node21 -->
<g id="edge175" class="edge">
-<title>Node59->Node23</title>
+<title>Node57->Node21</title>
<path fill="none" stroke="#191970" d="M802.7013,-716.4225C821.4447,-668.7629 909.8119,-453.8534 1032,-313 1041.9022,-301.5852 1054.6822,-291.0189 1066.4213,-282.4661"/>
<polygon fill="#191970" stroke="#191970" points="1068.5715,-285.2326 1074.7292,-276.6109 1064.539,-279.5109 1068.5715,-285.2326"/>
</g>
-<!-- Node59->Node35 -->
+<!-- Node57->Node33 -->
<g id="edge177" class="edge">
-<title>Node59->Node35</title>
+<title>Node57->Node33</title>
<path fill="none" stroke="#191970" d="M793.0573,-716.2405C772.2595,-681.1257 704,-557.3923 704,-446 704,-446 704,-446 704,-390 704,-320.32 710.2316,-287.7755 766,-246 788.9174,-228.8329 857.4242,-214.4365 911.7799,-205.2685"/>
<polygon fill="#191970" stroke="#191970" points="912.4564,-208.7043 921.7505,-203.618 911.3131,-201.7982 912.4564,-208.7043"/>
</g>
-<!-- Node59->Node44 -->
+<!-- Node57->Node42 -->
<g id="edge176" class="edge">
-<title>Node59->Node44</title>
+<title>Node57->Node42</title>
<path fill="none" stroke="#191970" d="M749.6257,-717.8364C702.5515,-709.3307 635.7145,-695.1493 614,-680 465.1146,-576.129 380.4739,-359.8565 355.154,-286.477"/>
<polygon fill="#191970" stroke="#191970" points="358.3715,-285.0661 351.8481,-276.7177 351.7416,-287.312 358.3715,-285.0661"/>
</g>
-<!-- Node60->Node4 -->
+<!-- Node58->Node4 -->
<g id="edge182" class="edge">
-<title>Node60->Node4</title>
+<title>Node58->Node4</title>
<path fill="none" stroke="#191970" d="M2680.3892,-551.9538C2547.4936,-540.5744 2263.1905,-516.2304 2147.5193,-506.3258"/>
<polygon fill="#191970" stroke="#191970" points="2147.7779,-502.8352 2137.5157,-505.4692 2147.1806,-509.8097 2147.7779,-502.8352"/>
</g>
-<!-- Node60->Node18 -->
+<!-- Node58->Node16 -->
<g id="edge184" class="edge">
-<title>Node60->Node18</title>
+<title>Node58->Node16</title>
<path fill="none" stroke="#191970" d="M2763.3813,-548.3732C2779.924,-535.5139 2810.1543,-512.025 2836,-492 2911.4763,-433.5219 3146,-289.9796 3146,-194.5 3146,-194.5 3146,-194.5 3146,-133 3146,-101.7875 3148.9748,-90.0686 3170,-67 3190.478,-44.5318 3222.8079,-30.9807 3246.9482,-23.4609"/>
<polygon fill="#191970" stroke="#191970" points="3248.1187,-26.7653 3256.7342,-20.599 3246.1538,-20.0467 3248.1187,-26.7653"/>
</g>
-<!-- Node60->Node20 -->
+<!-- Node58->Node18 -->
<g id="edge185" class="edge">
-<title>Node60->Node20</title>
+<title>Node58->Node18</title>
<path fill="none" stroke="#191970" d="M2680.3434,-550.147C2567.592,-537.6452 2340.0039,-512.5324 2147,-492 1511.6348,-424.4079 1350.822,-424.7918 717,-344 503.3354,-316.7647 256,-409.8934 256,-194.5 256,-194.5 256,-194.5 256,-133 256,-55.6578 368.0149,-28.3468 425.217,-19.4423"/>
<polygon fill="#191970" stroke="#191970" points="425.8293,-22.8902 435.2207,-17.986 424.8208,-15.9632 425.8293,-22.8902"/>
</g>
-<!-- Node60->Node22 -->
+<!-- Node58->Node20 -->
<g id="edge186" class="edge">
-<title>Node60->Node22</title>
+<title>Node58->Node20</title>
<path fill="none" stroke="#191970" d="M2751.9817,-548.4401C2754.8611,-525.4729 2765.3973,-466.8152 2800,-436 2845.4122,-395.5586 2884.2022,-437.5919 2932,-400 2966.712,-372.6998 2955.7024,-349.8761 2980,-313 2991.1525,-296.0741 3001.4217,-296.1726 3008,-277 3012.4714,-263.968 3015.7067,-257.4208 3008,-246 2951.9867,-162.9919 2823.656,-140.945 2762.1261,-135.101"/>
<polygon fill="#191970" stroke="#191970" points="2762.0584,-131.5824 2751.7956,-134.2139 2761.4594,-138.5567 2762.0584,-131.5824"/>
</g>
-<!-- Node60->Node43 -->
+<!-- Node58->Node41 -->
<g id="edge183" class="edge">
-<title>Node60->Node43</title>
+<title>Node58->Node41</title>
<path fill="none" stroke="#191970" d="M2749.5525,-548.3226C2744.4553,-517.9163 2724.3796,-423.7318 2666,-380 2627.5816,-351.2209 2497.705,-337.8813 2416.2967,-332.2153"/>
<polygon fill="#191970" stroke="#191970" points="2416.341,-328.7104 2406.1282,-331.5298 2415.8701,-335.6946 2416.341,-328.7104"/>
</g>
-<!-- Node64->Node1 -->
+<!-- Node62->Node1 -->
<g id="edge197" class="edge">
-<title>Node64->Node1</title>
+<title>Node62->Node1</title>
<path fill="none" stroke="#191970" d="M3079.6898,-895.2957C3105.1654,-883.732 3139.316,-862.9397 3125,-839 3065.7061,-739.8466 3004.8119,-754.8217 2896,-716 2833.9684,-693.8685 2758.7551,-681.5441 2710.1709,-675.3448"/>
<polygon fill="#191970" stroke="#191970" points="2710.5763,-671.8683 2700.2221,-674.1136 2709.7165,-678.8153 2710.5763,-671.8683"/>
</g>
-<!-- Node64->Node3 -->
+<!-- Node62->Node3 -->
<g id="edge198" class="edge">
-<title>Node64->Node3</title>
+<title>Node62->Node3</title>
<path fill="none" stroke="#191970" d="M3019.8058,-904.7751C2830.1865,-903.4177 1937.0975,-895.1192 1892,-859 1866.336,-838.4453 1873,-820.3806 1873,-787.5 1873,-787.5 1873,-787.5 1873,-726 1873,-693.7064 1873.8415,-656.1836 1874.4328,-633.8029"/>
<polygon fill="#191970" stroke="#191970" points="1877.9374,-633.6797 1874.7134,-623.5873 1870.9401,-633.4874 1877.9374,-633.6797"/>
</g>
-<!-- Node64->Node18 -->
+<!-- Node62->Node16 -->
<g id="edge205" class="edge">
-<title>Node64->Node18</title>
+<title>Node62->Node16</title>
<path fill="none" stroke="#191970" d="M3090.1274,-901.4589C3164.9798,-893.7161 3338.0081,-874.715 3395,-859 3529.1774,-822.0017 3564.5562,-809.0994 3683,-736 3721.6445,-712.15 3761,-715.4117 3761,-670 3761,-670 3761,-670 3761,-614 3761,-486.3653 3792,-456.1347 3792,-328.5 3792,-328.5 3792,-328.5 3792,-133 3792,-95.6561 3772.6395,-85.1447 3740,-67 3666.1151,-25.9264 3403.3829,-17.5848 3311.2277,-15.9127"/>
<polygon fill="#191970" stroke="#191970" points="3311.1629,-12.4112 3301.1059,-15.7438 3311.0461,-19.4102 3311.1629,-12.4112"/>
</g>
-<!-- Node64->Node20 -->
+<!-- Node62->Node18 -->
<g id="edge206" class="edge">
-<title>Node64->Node20</title>
+<title>Node62->Node18</title>
<path fill="none" stroke="#191970" d="M3019.8111,-904.586C2682.7593,-900.51 38,-866.0249 38,-787.5 38,-787.5 38,-787.5 38,-133 38,-94.5305 60.3114,-85.5737 94,-67 150.7175,-35.7297 346.2844,-21.6724 424.8963,-17.1894"/>
<polygon fill="#191970" stroke="#191970" points="425.3948,-20.6672 435.1859,-16.6196 425.0076,-13.6779 425.3948,-20.6672"/>
</g>
-<!-- Node64->Node22 -->
+<!-- Node62->Node20 -->
<g id="edge207" class="edge">
-<title>Node64->Node22</title>
+<title>Node62->Node20</title>
<path fill="none" stroke="#191970" d="M3079.384,-895.3554C3130.1203,-874.1092 3247.8184,-818.5081 3314,-736 3384.5496,-648.0463 3412,-614.7523 3412,-502 3412,-502 3412,-502 3412,-390 3412,-329.9589 3352.7396,-343.4623 3301,-313 3184.8701,-244.6273 3157.5552,-219.4297 3029,-179 2934.96,-149.4251 2818.2061,-138.5285 2761.7303,-134.7914"/>
<polygon fill="#191970" stroke="#191970" points="2761.7837,-131.2879 2751.5842,-134.1549 2761.3455,-138.2742 2761.7837,-131.2879"/>
</g>
-<!-- Node64->Node50 -->
+<!-- Node62->Node48 -->
<g id="edge199" class="edge">
-<title>Node64->Node50</title>
+<title>Node62->Node48</title>
<path fill="none" stroke="#191970" d="M3019.9615,-903.5088C2866.3806,-896.6473 2252.2749,-865.7296 2071,-803 2010.4067,-782.0319 1949,-790.1187 1949,-726 1949,-726 1949,-726 1949,-670 1949,-637.7116 1949,-600.1872 1949,-577.8048"/>
<polygon fill="#191970" stroke="#191970" points="1952.5001,-577.5883 1949,-567.5884 1945.5001,-577.5884 1952.5001,-577.5883"/>
</g>
-<!-- Node65 -->
+<!-- Node63 -->
<g id="node46" class="node">
-<title>Node65</title>
+<title>Node63</title>
<polygon fill="#ffffff" stroke="#bfbfbf" points="1901.5,-839.5 1901.5,-858.5 1990.5,-858.5 1990.5,-839.5 1901.5,-839.5"/>
<text text-anchor="middle" x="1946" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/registry.h</text>
</g>
-<!-- Node64->Node65 -->
+<!-- Node62->Node63 -->
<g id="edge196" class="edge">
-<title>Node64->Node65</title>
+<title>Node62->Node63</title>
<path fill="none" stroke="#191970" d="M3019.9416,-904.0502C2861.1658,-899.6585 2208.4458,-880.6132 2005,-859 2003.6925,-858.8611 2002.37,-858.7115 2001.0374,-858.5526"/>
<polygon fill="#191970" stroke="#191970" points="2001.0718,-855.0263 1990.6972,-857.1751 2000.1474,-861.965 2001.0718,-855.0263"/>
</g>
-<!-- Node66 -->
+<!-- Node64 -->
<g id="node47" class="node">
-<title>Node66</title>
+<title>Node64</title>
<g id="a_node47"><a xlink:href="type__relation_8h.html" target="_top" xlink:title="Type relation and function for type inference(checking). ">
<polygon fill="#ffffff" stroke="#ff0000" points="2994,-839.5 2994,-858.5 3116,-858.5 3116,-839.5 2994,-839.5"/>
<text text-anchor="middle" x="3055" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type_relation.h</text>
</a>
</g>
</g>
-<!-- Node64->Node66 -->
+<!-- Node62->Node64 -->
<g id="edge200" class="edge">
-<title>Node64->Node66</title>
+<title>Node62->Node64</title>
<path fill="none" stroke="#191970" d="M3055,-895.2455C3055,-887.9382 3055,-877.6944 3055,-868.7046"/>
<polygon fill="#191970" stroke="#191970" points="3058.5001,-868.6426 3055,-858.6427 3051.5001,-868.6427 3058.5001,-868.6426"/>
</g>
-<!-- Node66->Node1 -->
+<!-- Node64->Node1 -->
<g id="edge201" class="edge">
-<title>Node66->Node1</title>
+<title>Node64->Node1</title>
<path fill="none" stroke="#191970" d="M3044.3044,-839.2443C3016.6814,-814.594 2939.8024,-749.3996 2864,-716 2814.1237,-694.0238 2752.5919,-682.0034 2710.226,-675.8195"/>
<polygon fill="#191970" stroke="#191970" points="2710.544,-672.3299 2700.1547,-674.4066 2709.5714,-679.262 2710.544,-672.3299"/>
</g>
-<!-- Node66->Node17 -->
+<!-- Node64->Node15 -->
<g id="edge204" class="edge">
-<title>Node66->Node17</title>
+<title>Node64->Node15</title>
<path fill="none" stroke="#191970" d="M3116.1556,-839.8016C3258.8934,-814.5146 3602,-730.0161 3602,-502 3602,-502 3602,-502 3602,-390 3602,-269.572 3537.1924,-237.63 3432,-179 3323.3206,-118.4265 2455.3873,-42.779 2196.5854,-21.3799"/>
<polygon fill="#191970" stroke="#191970" points="2196.7962,-17.8855 2186.5423,-20.5514 2196.2206,-24.8618 2196.7962,-17.8855"/>
</g>
-<!-- Node66->Node50 -->
+<!-- Node64->Node48 -->
<g id="edge203" class="edge">
-<title>Node66->Node50</title>
+<title>Node64->Node48</title>
<path fill="none" stroke="#191970" d="M2993.9071,-848.3122C2799.7745,-845.8039 2205.0293,-835.5213 2125,-803 2024.6508,-762.2214 1971.5696,-627.4185 1954.8794,-577.1339"/>
<polygon fill="#191970" stroke="#191970" points="1958.1765,-575.9534 1951.789,-567.5006 1951.5111,-578.0918 1958.1765,-575.9534"/>
</g>
-<!-- Node66->Node56 -->
+<!-- Node64->Node54 -->
<g id="edge202" class="edge">
-<title>Node66->Node56</title>
+<title>Node64->Node54</title>
<path fill="none" stroke="#191970" d="M2993.6233,-846.6828C2734.8937,-836.915 1734.9769,-799.1649 1483.6001,-789.6746"/>
<polygon fill="#191970" stroke="#191970" points="1483.7072,-786.1762 1473.5822,-789.2964 1483.443,-793.1712 1483.7072,-786.1762"/>
</g>
-<!-- Node71->Node18 -->
+<!-- Node69->Node16 -->
<g id="edge209" class="edge">
-<title>Node71->Node18</title>
+<title>Node69->Node16</title>
<path fill="none" stroke="#191970" d="M779.5681,-786.1583C1008.8331,-780.1274 1937.1067,-754.9944 2228,-736 2486.8676,-719.0968 2552.7366,-720.3394 2809,-680 2955.0363,-657.0119 3450,-649.8346 3450,-502 3450,-502 3450,-502 3450,-390 3450,-268.8367 3640,-315.6633 3640,-194.5 3640,-194.5 3640,-194.5 3640,-133 3640,-90.5631 3608.8569,-86.1766 3571,-67 3525.8691,-44.1388 3377.9012,-26.023 3311.619,-18.8428"/>
<polygon fill="#191970" stroke="#191970" points="3311.6554,-15.3269 3301.3405,-17.7463 3310.9128,-22.2874 3311.6554,-15.3269"/>
</g>
-<!-- Node71->Node20 -->
+<!-- Node69->Node18 -->
<g id="edge211" class="edge">
-<title>Node71->Node20</title>
+<title>Node69->Node18</title>
<path fill="none" stroke="#191970" d="M676.3338,-786.6192C533.2773,-783.7229 139.5839,-772.4859 95,-736 71.3773,-716.6681 76,-700.5246 76,-670 76,-670 76,-670 76,-133 76,-95.6561 95.7634,-85.8512 128,-67 178.2132,-37.6364 352.4127,-22.6663 425.4149,-17.5644"/>
<polygon fill="#191970" stroke="#191970" points="425.6856,-21.0541 435.4246,-16.8827 425.2099,-14.0703 425.6856,-21.0541"/>
</g>
-<!-- Node71->Node63 -->
+<!-- Node69->Node61 -->
<g id="edge210" class="edge">
-<title>Node71->Node63</title>
+<title>Node69->Node61</title>
<path fill="none" stroke="#191970" d="M676.1574,-783.7134C581.9107,-776.5388 376.307,-759.5476 204,-736 202.7401,-735.8278 201.4655,-735.6482 200.1807,-735.4624"/>
<polygon fill="#191970" stroke="#191970" points="200.621,-731.9891 190.2062,-733.934 199.5607,-738.9083 200.621,-731.9891"/>
</g>
-<!-- Node79->Node1 -->
+<!-- Node77->Node1 -->
<g id="edge218" class="edge">
-<title>Node79->Node1</title>
+<title>Node77->Node1</title>
<path fill="none" stroke="#191970" d="M3311.5718,-902.1583C3337.7854,-897.7834 3376.5316,-886.7414 3395,-859 3423.7919,-815.7518 3479.3118,-812.8694 3352,-716 3326.534,-696.6233 2866.0447,-677.7027 2710.6946,-671.8462"/>
<polygon fill="#191970" stroke="#191970" points="2710.6025,-668.3403 2700.4784,-671.4633 2710.3403,-675.3354 2710.6025,-668.3403"/>
</g>
-<!-- Node79->Node18 -->
+<!-- Node77->Node16 -->
<g id="edge222" class="edge">
-<title>Node79->Node18</title>
+<title>Node77->Node16</title>
<path fill="none" stroke="#191970" d="M3311.6473,-899.6056C3404.707,-880.22 3712.9455,-811.2334 3775,-736 3858.4744,-634.7977 3830,-577.1864 3830,-446 3830,-446 3830,-446 3830,-133 3830,-101.7875 3830.663,-86.13 3806,-67 3766.8528,-36.6353 3419.9437,-20.8911 3311.3497,-16.6757"/>
<polygon fill="#191970" stroke="#191970" points="3311.4404,-13.1767 3301.3142,-16.2928 3311.1735,-20.1716 3311.4404,-13.1767"/>
</g>
-<!-- Node79->Node50 -->
+<!-- Node77->Node48 -->
<g id="edge219" class="edge">
-<title>Node79->Node50</title>
+<title>Node77->Node48</title>
<path fill="none" stroke="#191970" d="M3267.4365,-895.3278C3238.6792,-880.0582 3179.3046,-850.8821 3125,-839 2927.5137,-795.7892 2407.4515,-867.918 2216,-803 2167.0699,-786.4086 2160.236,-769.615 2121,-736 2058.39,-682.3595 1993.0199,-609.1449 1964.1919,-575.8226"/>
<polygon fill="#191970" stroke="#191970" points="1966.5413,-573.1867 1957.367,-567.8875 1961.2343,-577.7513 1966.5413,-573.1867"/>
</g>
-<!-- Node79->Node51 -->
+<!-- Node77->Node49 -->
<g id="edge223" class="edge">
-<title>Node79->Node51</title>
+<title>Node77->Node49</title>
<path fill="none" stroke="#191970" d="M3293.8835,-895.2455C3301.1966,-887.2155 3311.7391,-875.6394 3320.4467,-866.0781"/>
<polygon fill="#191970" stroke="#191970" points="3323.0727,-868.3928 3327.2183,-858.6427 3317.8972,-863.6795 3323.0727,-868.3928"/>
</g>
-<!-- Node79->Node52 -->
+<!-- Node77->Node50 -->
<g id="edge221" class="edge">
-<title>Node79->Node52</title>
+<title>Node77->Node50</title>
<path fill="none" stroke="#191970" d="M3275.7395,-895.1286C3260.7025,-879.861 3229.2889,-851.0272 3196,-839 3113.3873,-809.1523 2501.0485,-792.8715 2317.7152,-788.6295"/>
<polygon fill="#191970" stroke="#191970" points="2317.6079,-785.1262 2307.5303,-788.3961 2317.4474,-792.1244 2317.6079,-785.1262"/>
</g>
-<!-- Node79->Node66 -->
+<!-- Node77->Node64 -->
<g id="edge220" class="edge">
-<title>Node79->Node66</title>
+<title>Node77->Node64</title>
<path fill="none" stroke="#191970" d="M3258.2822,-897.2276C3255.4904,-896.457 3252.6907,-895.7013 3250,-895 3202.1615,-882.5307 3147.3363,-869.7283 3108.0275,-860.8061"/>
<polygon fill="#191970" stroke="#191970" points="3108.6556,-857.3598 3098.1298,-858.5671 3107.1111,-864.1873 3108.6556,-857.3598"/>
</g>
diff --git a/docs/reference/api/doxygen/algorithms_8h.html b/docs/reference/api/doxygen/algorithms_8h.html
index c05a5a568..21dbb4c3a 100644
--- a/docs/reference/api/doxygen/algorithms_8h.html
+++ b/docs/reference/api/doxygen/algorithms_8h.html
@@ -76,7 +76,7 @@ $(function() {
</div><div class="textblock"><div class="dynheader">
Include dependency graph for algorithms.h:</div>
<div class="dyncontent">
-<div class="center"><iframe scrolling="no" frameborder="0" src="algorithms_8h__incl.svg" width="4284" height="1395"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
+<div class="center"><iframe scrolling="no" frameborder="0" src="algorithms_8h__incl.svg" width="4392" height="1395"><p><b>This browser is not able to show SVG: try Firefox, Chrome, Safari, or Opera instead.</b></p></iframe>
</div>
</div>
</div>
diff --git a/docs/reference/api/doxygen/algorithms_8h__incl.svg b/docs/reference/api/doxygen/algorithms_8h__incl.svg
index 035f8504a..9ab219824 100644
--- a/docs/reference/api/doxygen/algorithms_8h__incl.svg
+++ b/docs/reference/api/doxygen/algorithms_8h__incl.svg
@@ -4,1445 +4,1472 @@
<!-- Generated by graphviz version 2.40.1 (20161225.0304)
-->
<!-- Title: include/tvm/tir/usmp/algorithms.h Pages: 1 -->
-<svg width="3213pt" height="1046pt"
- viewBox="0.00 0.00 3212.50 1046.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+<svg width="3294pt" height="1046pt"
+ viewBox="0.00 0.00 3294.00 1046.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 1042)">
<title>include/tvm/tir/usmp/algorithms.h</title>
-<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-1042 3208.5,-1042 3208.5,4 -4,4"/>
+<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-1042 3290,-1042 3290,4 -4,4"/>
<!-- Node0 -->
<g id="node1" class="node">
<title>Node0</title>
-<polygon fill="#bfbfbf" stroke="#000000" points="810.5,-1007.5 810.5,-1037.5 928.5,-1037.5 928.5,-1007.5 810.5,-1007.5"/>
-<text text-anchor="start" x="818.5" y="-1025.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">include/tvm/tir/usmp</text>
-<text text-anchor="middle" x="869.5" y="-1014.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/algorithms.h</text>
+<polygon fill="#bfbfbf" stroke="#000000" points="835,-1007.5 835,-1037.5 953,-1037.5 953,-1007.5 835,-1007.5"/>
+<text text-anchor="start" x="843" y="-1025.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">include/tvm/tir/usmp</text>
+<text text-anchor="middle" x="894" y="-1014.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/algorithms.h</text>
</g>
<!-- Node1 -->
<g id="node2" class="node">
<title>Node1</title>
<g id="a_node2"><a xlink:href="tir_2usmp_2utils_8h.html" target="_top" xlink:title="Utilities for Unified Static Memory Planner. ">
-<polygon fill="#ffffff" stroke="#000000" points="813,-951.5 813,-970.5 926,-970.5 926,-951.5 813,-951.5"/>
-<text text-anchor="middle" x="869.5" y="-958.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/tir/usmp/utils.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="837.5,-951.5 837.5,-970.5 950.5,-970.5 950.5,-951.5 837.5,-951.5"/>
+<text text-anchor="middle" x="894" y="-958.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/tir/usmp/utils.h</text>
</a>
</g>
</g>
<!-- Node0->Node1 -->
<g id="edge1" class="edge">
<title>Node0->Node1</title>
-<path fill="none" stroke="#191970" d="M869.5,-1007.2977C869.5,-999.3834 869.5,-989.6043 869.5,-981.0759"/>
-<polygon fill="#191970" stroke="#191970" points="873.0001,-980.8469 869.5,-970.8469 866.0001,-980.847 873.0001,-980.8469"/>
+<path fill="none" stroke="#191970" d="M894,-1007.2977C894,-999.3834 894,-989.6043 894,-981.0759"/>
+<polygon fill="#191970" stroke="#191970" points="897.5001,-980.8469 894,-970.8469 890.5001,-980.847 897.5001,-980.8469"/>
</g>
<!-- Node2 -->
<g id="node3" class="node">
<title>Node2</title>
<g id="a_node3"><a xlink:href="ir_2expr_8h.html" target="_top" xlink:title="Base expr nodes in TVM. ">
-<polygon fill="#ffffff" stroke="#000000" points="1514,-727.5 1514,-746.5 1593,-746.5 1593,-727.5 1514,-727.5"/>
-<text text-anchor="middle" x="1553.5" y="-734.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/expr.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2546.5,-727.5 2546.5,-746.5 2625.5,-746.5 2625.5,-727.5 2546.5,-727.5"/>
+<text text-anchor="middle" x="2586" y="-734.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/expr.h</text>
</a>
</g>
</g>
<!-- Node1->Node2 -->
<g id="edge2" class="edge">
<title>Node1->Node2</title>
-<path fill="none" stroke="#191970" d="M900.8645,-951.499C979.7341,-927.5143 1192.5924,-862.2058 1368.5,-803 1420.6414,-785.4506 1480.5373,-763.8083 1517.7056,-750.2016"/>
-<polygon fill="#191970" stroke="#191970" points="1519.3777,-753.3164 1527.561,-746.587 1516.9673,-746.7445 1519.3777,-753.3164"/>
+<path fill="none" stroke="#191970" d="M950.6966,-960.0806C1209.6303,-955.7687 2270.0969,-936.8389 2337,-915 2437.9722,-882.04 2533.0673,-792.0345 2569.9065,-754.1585"/>
+<polygon fill="#191970" stroke="#191970" points="2572.8486,-756.1472 2577.2427,-746.5066 2567.7956,-751.3028 2572.8486,-756.1472"/>
</g>
-<!-- Node50 -->
+<!-- Node48 -->
<g id="node41" class="node">
-<title>Node50</title>
+<title>Node48</title>
<g id="a_node41"><a xlink:href="memory__pools_8h.html" target="_top" xlink:title="The object definition for relay.build argument type of memory pools. ">
-<polygon fill="#ffffff" stroke="#000000" points="538,-895.5 538,-914.5 669,-914.5 669,-895.5 538,-895.5"/>
-<text text-anchor="middle" x="603.5" y="-902.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/memory_pools.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="390.5,-895.5 390.5,-914.5 521.5,-914.5 521.5,-895.5 390.5,-895.5"/>
+<text text-anchor="middle" x="456" y="-902.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/memory_pools.h</text>
</a>
</g>
</g>
-<!-- Node1->Node50 -->
-<g id="edge120" class="edge">
-<title>Node1->Node50</title>
-<path fill="none" stroke="#191970" d="M824.0752,-951.4369C778.2645,-941.7925 707.5002,-926.8948 658.5325,-916.5858"/>
-<polygon fill="#191970" stroke="#191970" points="659.244,-913.1589 648.7374,-914.5237 657.8019,-920.0088 659.244,-913.1589"/>
+<!-- Node1->Node48 -->
+<g id="edge124" class="edge">
+<title>Node1->Node48</title>
+<path fill="none" stroke="#191970" d="M837.1882,-953.7364C759.1482,-943.7587 618.596,-925.7885 531.6995,-914.6785"/>
+<polygon fill="#191970" stroke="#191970" points="532.0005,-911.1885 521.6373,-913.392 531.1127,-918.132 532.0005,-911.1885"/>
</g>
-<!-- Node52 -->
+<!-- Node50 -->
<g id="node43" class="node">
-<title>Node52</title>
+<title>Node50</title>
<g id="a_node43"><a xlink:href="target_8h.html" target="_top" xlink:title="Compilation target object. ">
-<polygon fill="#ffffff" stroke="#000000" points="548.5,-839.5 548.5,-858.5 658.5,-858.5 658.5,-839.5 548.5,-839.5"/>
-<text text-anchor="middle" x="603.5" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/target/target.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="401,-839.5 401,-858.5 511,-858.5 511,-839.5 401,-839.5"/>
+<text text-anchor="middle" x="456" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/target/target.h</text>
</a>
</g>
</g>
-<!-- Node1->Node52 -->
-<g id="edge156" class="edge">
-<title>Node1->Node52</title>
-<path fill="none" stroke="#191970" d="M846.821,-951.4509C799.5107,-931.5308 690.1738,-885.4942 635.6393,-862.5323"/>
-<polygon fill="#191970" stroke="#191970" points="636.6962,-859.1798 626.1216,-858.5249 633.9797,-865.6312 636.6962,-859.1798"/>
+<!-- Node1->Node50 -->
+<g id="edge160" class="edge">
+<title>Node1->Node50</title>
+<path fill="none" stroke="#191970" d="M856.6564,-951.4509C777.0575,-931.0969 590.8262,-883.4761 503.1831,-861.0651"/>
+<polygon fill="#191970" stroke="#191970" points="503.8045,-857.6114 493.2491,-858.5249 502.0703,-864.3932 503.8045,-857.6114"/>
</g>
-<!-- Node70 -->
+<!-- Node68 -->
<g id="node48" class="node">
-<title>Node70</title>
+<title>Node68</title>
<g id="a_node48"><a xlink:href="device__api_8h.html" target="_top" xlink:title="Abstract device memory management API. ">
-<polygon fill="#ffffff" stroke="#000000" points="1948,-492.5 1948,-522.5 2061,-522.5 2061,-492.5 1948,-492.5"/>
-<text text-anchor="start" x="1956" y="-510.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/device</text>
-<text text-anchor="middle" x="2004.5" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1611.5,-492.5 1611.5,-522.5 1724.5,-522.5 1724.5,-492.5 1611.5,-492.5"/>
+<text text-anchor="start" x="1619.5" y="-510.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/device</text>
+<text text-anchor="middle" x="1668" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
</a>
</g>
</g>
-<!-- Node1->Node70 -->
-<g id="edge151" class="edge">
-<title>Node1->Node70</title>
-<path fill="none" stroke="#191970" d="M926.1495,-959.5977C1267.8497,-950.9587 3039.5,-903.5374 3039.5,-849 3039.5,-849 3039.5,-849 3039.5,-793 3039.5,-591.4095 2300.6167,-526.6588 2071.4509,-511.4242"/>
-<polygon fill="#191970" stroke="#191970" points="2071.4575,-507.9173 2061.2502,-510.7566 2071.0002,-514.9023 2071.4575,-507.9173"/>
+<!-- Node1->Node68 -->
+<g id="edge155" class="edge">
+<title>Node1->Node68</title>
+<path fill="none" stroke="#191970" d="M950.9003,-959.5732C1216.6771,-952.5422 2323,-919.0671 2323,-849 2323,-849 2323,-849 2323,-793 2323,-727.9193 1882.8457,-577.685 1724.1707,-525.6402"/>
+<polygon fill="#191970" stroke="#191970" points="1725.1735,-522.2858 1714.5809,-522.502 1722.9964,-528.9386 1725.1735,-522.2858"/>
</g>
-<!-- Node71 -->
+<!-- Node69 -->
<g id="node49" class="node">
-<title>Node71</title>
+<title>Node69</title>
<g id="a_node49"><a xlink:href="stmt_8h.html" target="_top" xlink:title="TIR statements. ">
-<polygon fill="#ffffff" stroke="#000000" points="827.5,-839.5 827.5,-858.5 911.5,-858.5 911.5,-839.5 827.5,-839.5"/>
-<text text-anchor="middle" x="869.5" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/tir/stmt.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="852,-839.5 852,-858.5 936,-858.5 936,-839.5 852,-839.5"/>
+<text text-anchor="middle" x="894" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/tir/stmt.h</text>
</a>
</g>
</g>
-<!-- Node1->Node71 -->
-<g id="edge157" class="edge">
-<title>Node1->Node71</title>
-<path fill="none" stroke="#191970" d="M869.5,-951.4509C869.5,-933.184 869.5,-892.9553 869.5,-868.6976"/>
-<polygon fill="#191970" stroke="#191970" points="873.0001,-868.5249 869.5,-858.5249 866.0001,-868.5249 873.0001,-868.5249"/>
+<!-- Node1->Node69 -->
+<g id="edge161" class="edge">
+<title>Node1->Node69</title>
+<path fill="none" stroke="#191970" d="M894,-951.4509C894,-933.184 894,-892.9553 894,-868.6976"/>
+<polygon fill="#191970" stroke="#191970" points="897.5001,-868.5249 894,-858.5249 890.5001,-868.5249 897.5001,-868.5249"/>
</g>
<!-- Node3 -->
<g id="node4" class="node">
<title>Node3</title>
<g id="a_node4"><a xlink:href="ir_2span_8h.html" target="_top" xlink:title="Span information for debugging purposes. ">
-<polygon fill="#ffffff" stroke="#000000" points="1513,-615.5 1513,-634.5 1594,-634.5 1594,-615.5 1513,-615.5"/>
-<text text-anchor="middle" x="1553.5" y="-622.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/span.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1310.5,-615.5 1310.5,-634.5 1391.5,-634.5 1391.5,-615.5 1310.5,-615.5"/>
+<text text-anchor="middle" x="1351" y="-622.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/span.h</text>
</a>
</g>
</g>
<!-- Node2->Node3 -->
<g id="edge3" class="edge">
<title>Node2->Node3</title>
-<path fill="none" stroke="#191970" d="M1553.5,-727.4509C1553.5,-709.184 1553.5,-668.9553 1553.5,-644.6976"/>
-<polygon fill="#191970" stroke="#191970" points="1557.0001,-644.5249 1553.5,-634.5249 1550.0001,-644.5249 1557.0001,-644.5249"/>
+<path fill="none" stroke="#191970" d="M2546.3081,-733.4004C2363.8951,-716.8577 1606.2242,-648.1458 1401.9376,-629.6194"/>
+<polygon fill="#191970" stroke="#191970" points="1402.0649,-626.1167 1391.7897,-628.6991 1401.4326,-633.0881 1402.0649,-626.1167"/>
</g>
<!-- Node4 -->
<g id="node5" class="node">
<title>Node4</title>
<g id="a_node5"><a xlink:href="node_8h.html" target="_top" xlink:title="Definitions and helper macros for IR/AST nodes. ">
-<polygon fill="#ffffff" stroke="#000000" points="1397,-559.5 1397,-578.5 1496,-578.5 1496,-559.5 1397,-559.5"/>
-<text text-anchor="middle" x="1446.5" y="-566.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/node.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1370.5,-559.5 1370.5,-578.5 1469.5,-578.5 1469.5,-559.5 1370.5,-559.5"/>
+<text text-anchor="middle" x="1420" y="-566.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/node.h</text>
</a>
</g>
</g>
<!-- Node2->Node4 -->
-<g id="edge113" class="edge">
+<g id="edge117" class="edge">
<title>Node2->Node4</title>
-<path fill="none" stroke="#191970" d="M1563.571,-727.3004C1584.9901,-705.4935 1630.946,-651.6413 1603.5,-615 1591.5814,-599.0883 1545.75,-586.8234 1506.6595,-578.975"/>
-<polygon fill="#191970" stroke="#191970" points="1506.8542,-575.4475 1496.3704,-576.9773 1505.5199,-582.3191 1506.8542,-575.4475"/>
+<path fill="none" stroke="#191970" d="M2566.8401,-727.4249C2517.5836,-703.3439 2381.9152,-640.3774 2261,-615 2111.5411,-583.6319 1646.0951,-572.8281 1479.8051,-569.9135"/>
+<polygon fill="#191970" stroke="#191970" points="1479.818,-566.4133 1469.7592,-569.7407 1479.6975,-573.4123 1479.818,-566.4133"/>
</g>
-<!-- Node23 -->
+<!-- Node21 -->
<g id="node9" class="node">
-<title>Node23</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2281.5,-179.5 2281.5,-198.5 2345.5,-198.5 2345.5,-179.5 2281.5,-179.5"/>
-<text text-anchor="middle" x="2313.5" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
+<title>Node21</title>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2414,-179.5 2414,-198.5 2478,-198.5 2478,-179.5 2414,-179.5"/>
+<text text-anchor="middle" x="2446" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
</g>
-<!-- Node2->Node23 -->
-<g id="edge116" class="edge">
-<title>Node2->Node23</title>
-<path fill="none" stroke="#191970" d="M1593.3017,-736.1855C1724.3972,-733.2321 2141.7264,-721.6019 2272.5,-691 2425.4679,-655.2045 2478.4837,-639.8421 2583.5,-523 2658.4395,-439.6215 2750.3948,-380.2087 2682.5,-291 2642.6669,-238.6623 2443.8781,-206.2854 2355.7188,-194.2873"/>
-<polygon fill="#191970" stroke="#191970" points="2356.16,-190.8152 2345.7841,-192.958 2355.2316,-197.7534 2356.16,-190.8152"/>
+<!-- Node2->Node21 -->
+<g id="edge120" class="edge">
+<title>Node2->Node21</title>
+<path fill="none" stroke="#191970" d="M2625.8137,-728.8817C2675.4935,-716.1572 2754,-686.3444 2754,-625 2754,-625 2754,-625 2754,-373.5 2754,-335.8735 2759.8691,-319.2361 2735,-291 2718.1089,-271.8221 2563.7983,-223.9805 2487.844,-201.3"/>
+<polygon fill="#191970" stroke="#191970" points="2488.6898,-197.9001 2478.1068,-198.4021 2486.693,-204.6092 2488.6898,-197.9001"/>
</g>
<!-- Node9 -->
<g id="node17" class="node">
<title>Node9</title>
<g id="a_node17"><a xlink:href="object_8h.html" target="_top" xlink:title="A managed object in the TVM runtime. ">
-<polygon fill="#ffffff" stroke="#000000" points="1920,-123.5 1920,-142.5 2039,-142.5 2039,-123.5 1920,-123.5"/>
-<text text-anchor="middle" x="1979.5" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/object.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1848.5,-123.5 1848.5,-142.5 1967.5,-142.5 1967.5,-123.5 1848.5,-123.5"/>
+<text text-anchor="middle" x="1908" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/object.h</text>
</a>
</g>
</g>
<!-- Node2->Node9 -->
-<g id="edge115" class="edge">
+<g id="edge119" class="edge">
<title>Node2->Node9</title>
-<path fill="none" stroke="#191970" d="M1593.2715,-736.1248C1730.4633,-732.8725 2183.2796,-720.1378 2325.5,-691 2397.6662,-676.2147 2417.0743,-670.7197 2481.5,-635 2613.4702,-561.8314 2675.3805,-534.4655 2715.5,-389 2759.4523,-229.6376 2569.2966,-217.375 2408.5,-179 2284.2255,-149.3412 2135.1103,-138.7838 2049.4241,-135.0404"/>
-<polygon fill="#191970" stroke="#191970" points="2049.4324,-131.5378 2039.2949,-134.6169 2049.1399,-138.5317 2049.4324,-131.5378"/>
+<path fill="none" stroke="#191970" d="M2625.65,-729.1183C2665.5813,-720.5124 2723.6856,-705.9519 2741,-691 2895.9395,-557.2013 2950.1862,-380.3227 2806,-235 2747.8863,-176.4282 2179.0212,-145.2463 1977.9572,-135.9868"/>
+<polygon fill="#191970" stroke="#191970" points="1977.9569,-132.4833 1967.8077,-135.5237 1977.6377,-139.476 1977.9569,-132.4833"/>
</g>
-<!-- Node17 -->
-<g id="node24" class="node">
-<title>Node17</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1149.5,-62 1149.5,-81 1193.5,-81 1193.5,-62 1149.5,-62"/>
-<text text-anchor="middle" x="1171.5" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
+<!-- Node15 -->
+<g id="node22" class="node">
+<title>Node15</title>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="320,-62 320,-81 364,-81 364,-62 320,-62"/>
+<text text-anchor="middle" x="342" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
</g>
-<!-- Node2->Node17 -->
-<g id="edge118" class="edge">
-<title>Node2->Node17</title>
-<path fill="none" stroke="#191970" d="M1513.9792,-728.5275C1440.7165,-712.0632 1287.9506,-673.9344 1251.5,-635 1115.8926,-490.1521 1157.2714,-397.4115 1155.5,-199 1155.1982,-165.1973 1153.4353,-156.4225 1158.5,-123 1160.1262,-112.2686 1163.0051,-100.5184 1165.6411,-90.9549"/>
-<polygon fill="#191970" stroke="#191970" points="1169.0297,-91.8354 1168.4342,-81.2573 1162.3032,-89.898 1169.0297,-91.8354"/>
+<!-- Node2->Node15 -->
+<g id="edge122" class="edge">
+<title>Node2->Node15</title>
+<path fill="none" stroke="#191970" d="M2546.2481,-736.7564C2378.8048,-735.5094 1714.4513,-728.3428 1170,-691 931.9245,-674.6709 863.1273,-708.2103 636,-635 571.2017,-614.1135 567.4663,-580.8897 503,-559 390.8686,-520.9256 337.7153,-588.4094 239,-523 203.4496,-499.4441 190,-483.1464 190,-440.5 190,-440.5 190,-440.5 190,-189 190,-128.8142 265.5179,-95.0962 310.0946,-80.4359"/>
+<polygon fill="#191970" stroke="#191970" points="311.4082,-83.6918 319.8979,-77.3534 309.3085,-77.0142 311.4082,-83.6918"/>
</g>
-<!-- Node18 -->
-<g id="node25" class="node">
-<title>Node18</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2700,-62 2700,-81 2769,-81 2769,-62 2700,-62"/>
-<text text-anchor="middle" x="2734.5" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
+<!-- Node16 -->
+<g id="node23" class="node">
+<title>Node16</title>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2929.5,-62 2929.5,-81 2998.5,-81 2998.5,-62 2929.5,-62"/>
+<text text-anchor="middle" x="2964" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
</g>
-<!-- Node2->Node18 -->
-<g id="edge119" class="edge">
-<title>Node2->Node18</title>
-<path fill="none" stroke="#191970" d="M1593.198,-736.4994C1778.1331,-734.0189 2556.3375,-721.906 2797.5,-691 2932.0535,-673.7564 3095.5,-760.6539 3095.5,-625 3095.5,-625 3095.5,-625 3095.5,-245 3095.5,-175.4899 2875.3287,-108.7321 2778.8745,-82.8215"/>
-<polygon fill="#191970" stroke="#191970" points="2779.6779,-79.4135 2769.1142,-80.2258 2777.8787,-86.1784 2779.6779,-79.4135"/>
+<!-- Node2->Node16 -->
+<g id="edge123" class="edge">
+<title>Node2->Node16</title>
+<path fill="none" stroke="#191970" d="M2625.9202,-728.3029C2769.0454,-695.9884 3248,-577.6081 3248,-440.5 3248,-440.5 3248,-440.5 3248,-189 3248,-137.7657 3088.1891,-97.2785 3008.3293,-80.245"/>
+<polygon fill="#191970" stroke="#191970" points="3009.0459,-76.8193 2998.54,-78.1892 3007.6072,-83.6698 3009.0459,-76.8193"/>
</g>
-<!-- Node34 -->
-<g id="node33" class="node">
-<title>Node34</title>
-<g id="a_node33"><a xlink:href="string_8h.html" target="_top" xlink:title="Runtime String container types. ">
-<polygon fill="#ffffff" stroke="#000000" points="2087.5,-291.5 2087.5,-321.5 2213.5,-321.5 2213.5,-291.5 2087.5,-291.5"/>
-<text text-anchor="start" x="2095.5" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="2150.5" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/string.h</text>
+<!-- Node32 -->
+<g id="node31" class="node">
+<title>Node32</title>
+<g id="a_node31"><a xlink:href="string_8h.html" target="_top" xlink:title="Runtime String container types. ">
+<polygon fill="#ffffff" stroke="#000000" points="2372,-291.5 2372,-321.5 2498,-321.5 2498,-291.5 2372,-291.5"/>
+<text text-anchor="start" x="2380" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
+<text text-anchor="middle" x="2435" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/string.h</text>
</a>
</g>
</g>
-<!-- Node2->Node34 -->
-<g id="edge114" class="edge">
-<title>Node2->Node34</title>
-<path fill="none" stroke="#191970" d="M1593.1686,-735.9441C1759.6722,-730.9967 2393.5,-706.7917 2393.5,-625 2393.5,-625 2393.5,-625 2393.5,-569 2393.5,-566.9668 2230.7008,-392.3986 2171.6073,-329.1002"/>
-<polygon fill="#191970" stroke="#191970" points="2173.9979,-326.532 2164.615,-321.6112 2168.8813,-331.3092 2173.9979,-326.532"/>
+<!-- Node2->Node32 -->
+<g id="edge118" class="edge">
+<title>Node2->Node32</title>
+<path fill="none" stroke="#191970" d="M2613.5697,-727.4257C2651.73,-712.2096 2716,-678.6227 2716,-625 2716,-625 2716,-625 2716,-440.5 2716,-402.8735 2724.0317,-384.1733 2697,-358 2683.4222,-344.8534 2580.8219,-327.5573 2508.1291,-316.7334"/>
+<polygon fill="#191970" stroke="#191970" points="2508.4867,-313.2483 2498.0827,-315.2501 2507.4642,-320.1732 2508.4867,-313.2483"/>
</g>
-<!-- Node45 -->
+<!-- Node43 -->
<g id="node38" class="node">
-<title>Node45</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2877.5,-364 2877.5,-383 2921.5,-383 2921.5,-364 2877.5,-364"/>
-<text text-anchor="middle" x="2899.5" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
+<title>Node43</title>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2644,-364 2644,-383 2688,-383 2688,-364 2644,-364"/>
+<text text-anchor="middle" x="2666" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
</g>
-<!-- Node2->Node45 -->
-<g id="edge117" class="edge">
-<title>Node2->Node45</title>
-<path fill="none" stroke="#191970" d="M1593.0221,-736.7422C1787.4416,-735.3051 2637.4787,-726.9957 2749.5,-691 2800.1987,-674.7091 2849.5,-678.2518 2849.5,-625 2849.5,-625 2849.5,-625 2849.5,-507.5 2849.5,-463.8776 2873.0469,-417.0212 2887.8226,-391.9129"/>
-<polygon fill="#191970" stroke="#191970" points="2890.9628,-393.4851 2893.1675,-383.1221 2884.9816,-389.8484 2890.9628,-393.4851"/>
+<!-- Node2->Node43 -->
+<g id="edge121" class="edge">
+<title>Node2->Node43</title>
+<path fill="none" stroke="#191970" d="M2602.4478,-727.232C2628.9102,-710.1274 2678,-672.0556 2678,-625 2678,-625 2678,-625 2678,-507.5 2678,-466.8078 2672.4424,-419.4407 2668.8915,-393.3298"/>
+<polygon fill="#191970" stroke="#191970" points="2672.3291,-392.6406 2667.4714,-383.225 2665.3972,-393.6148 2672.3291,-392.6406"/>
</g>
-<!-- Node49 -->
+<!-- Node47 -->
<g id="node40" class="node">
-<title>Node49</title>
+<title>Node47</title>
<g id="a_node40"><a xlink:href="ir_2type_8h.html" target="_top" xlink:title="IR/AST nodes for the unified type system in TVM. ">
-<polygon fill="#ffffff" stroke="#000000" points="1407.5,-671.5 1407.5,-690.5 1487.5,-690.5 1487.5,-671.5 1407.5,-671.5"/>
-<text text-anchor="middle" x="1447.5" y="-678.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1179,-671.5 1179,-690.5 1259,-690.5 1259,-671.5 1179,-671.5"/>
+<text text-anchor="middle" x="1219" y="-678.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
</a>
</g>
</g>
-<!-- Node2->Node49 -->
-<g id="edge106" class="edge">
-<title>Node2->Node49</title>
-<path fill="none" stroke="#191970" d="M1535.0362,-727.2455C1518.3926,-718.4527 1493.7007,-705.4079 1474.7808,-695.4125"/>
-<polygon fill="#191970" stroke="#191970" points="1476.2291,-692.2193 1465.7522,-690.6427 1472.9592,-698.4086 1476.2291,-692.2193"/>
+<!-- Node2->Node47 -->
+<g id="edge110" class="edge">
+<title>Node2->Node47</title>
+<path fill="none" stroke="#191970" d="M2546.1964,-735.3694C2350.1446,-727.338 1487.8468,-692.0135 1269.4233,-683.0656"/>
+<polygon fill="#191970" stroke="#191970" points="1269.3506,-679.5598 1259.2157,-682.6475 1269.064,-686.5539 1269.3506,-679.5598"/>
</g>
<!-- Node3->Node4 -->
<g id="edge4" class="edge">
<title>Node3->Node4</title>
-<path fill="none" stroke="#191970" d="M1534.862,-615.2455C1518.0614,-606.4527 1493.1366,-593.4079 1474.0382,-583.4125"/>
-<polygon fill="#191970" stroke="#191970" points="1475.4073,-580.1787 1464.9244,-578.6427 1472.1614,-586.3807 1475.4073,-580.1787"/>
+<path fill="none" stroke="#191970" d="M1363.0189,-615.2455C1373.2098,-606.9746 1388.0362,-594.9416 1400.0107,-585.2232"/>
+<polygon fill="#191970" stroke="#191970" points="1402.5598,-587.662 1408.1188,-578.6427 1398.1486,-582.2268 1402.5598,-587.662"/>
</g>
<!-- Node3->Node9 -->
-<g id="edge104" class="edge">
+<g id="edge108" class="edge">
<title>Node3->Node9</title>
-<path fill="none" stroke="#191970" d="M1594.255,-624.2957C1755.1285,-620.5307 2342.138,-597.1677 2442.5,-456 2515.6562,-353.0995 2490.6888,-238.8114 2379.5,-179 2351.1352,-163.7419 2156.2868,-146.6112 2049.3731,-138.2127"/>
-<polygon fill="#191970" stroke="#191970" points="2049.5516,-134.7161 2039.3097,-137.4276 2049.0071,-141.6949 2049.5516,-134.7161"/>
+<path fill="none" stroke="#191970" d="M1391.5374,-618.9786C1446.7247,-610.6569 1548.5345,-594.8636 1635,-579 1756.8264,-556.6488 1787.1357,-550.0749 1908,-523 1919.5346,-520.4161 2727.695,-331.2931 2735,-322 2836.136,-193.3389 3031.5471,-319.054 2487,-179 2392.5342,-154.704 2110.1852,-140.8591 1977.869,-135.5446"/>
+<polygon fill="#191970" stroke="#191970" points="1977.9701,-132.046 1967.8391,-135.1466 1977.6924,-139.0404 1977.9701,-132.046"/>
</g>
-<!-- Node3->Node17 -->
-<g id="edge105" class="edge">
-<title>Node3->Node17</title>
-<path fill="none" stroke="#191970" d="M1512.9672,-615.8464C1478.9696,-607.6775 1429.3943,-594.6121 1387.5,-579 1333.1037,-558.7289 1314.0444,-560.2247 1269.5,-523 1241.3917,-499.5106 1241.9319,-486.4033 1221.5,-456 1129.8174,-319.5735 1070.0633,-300.9299 1024.5,-143 1010.0922,-93.0601 1091.5389,-77.9806 1139.0613,-73.4425"/>
-<polygon fill="#191970" stroke="#191970" points="1139.5366,-76.9148 1149.2054,-72.5832 1138.9457,-69.9398 1139.5366,-76.9148"/>
+<!-- Node3->Node15 -->
+<g id="edge109" class="edge">
+<title>Node3->Node15</title>
+<path fill="none" stroke="#191970" d="M1310.238,-621.3934C1122.4457,-604.7214 348.8318,-535.4046 327,-523 287.352,-500.4725 266,-486.1011 266,-440.5 266,-440.5 266,-440.5 266,-189 266,-147.4563 299.9099,-108.986 322.4281,-88.0193"/>
+<polygon fill="#191970" stroke="#191970" points="325.0192,-90.3968 330.125,-81.1133 320.3444,-85.1865 325.0192,-90.3968"/>
</g>
<!-- Node5 -->
<g id="node6" class="node">
<title>Node5</title>
<g id="a_node6"><a xlink:href="reflection_8h.html" target="_top" xlink:title="Reflection and serialization of compiler IR/AST nodes. ">
-<polygon fill="#ffffff" stroke="#000000" points="1748,-498 1748,-517 1869,-517 1869,-498 1748,-498"/>
-<text text-anchor="middle" x="1808.5" y="-505" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/reflection.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1742.5,-498 1742.5,-517 1863.5,-517 1863.5,-498 1742.5,-498"/>
+<text text-anchor="middle" x="1803" y="-505" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/reflection.h</text>
</a>
</g>
</g>
<!-- Node4->Node5 -->
<g id="edge5" class="edge">
<title>Node4->Node5</title>
-<path fill="none" stroke="#191970" d="M1496.0755,-560.5776C1559.9337,-549.7288 1671.1448,-530.8352 1742.2156,-518.761"/>
-<polygon fill="#191970" stroke="#191970" points="1743.1273,-522.1564 1752.3998,-517.0308 1741.9548,-515.2553 1743.1273,-522.1564"/>
+<path fill="none" stroke="#191970" d="M1469.5131,-562.8432C1531.37,-554.8944 1640.3597,-540.0412 1733,-523 1739.2662,-521.8473 1745.8252,-520.5359 1752.3094,-519.1753"/>
+<polygon fill="#191970" stroke="#191970" points="1753.3417,-522.5335 1762.3843,-517.0125 1751.8724,-515.6895 1753.3417,-522.5335"/>
</g>
<!-- Node6 -->
<g id="node7" class="node">
<title>Node6</title>
<g id="a_node7"><a xlink:href="structural__equal_8h.html" target="_top" xlink:title="Structural equality comparison. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="1352,-358.5 1352,-388.5 1465,-388.5 1465,-358.5 1352,-358.5"/>
-<text text-anchor="start" x="1360" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
-<text text-anchor="middle" x="1408.5" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_equal.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="1244.5,-358.5 1244.5,-388.5 1357.5,-388.5 1357.5,-358.5 1244.5,-358.5"/>
+<text text-anchor="start" x="1252.5" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
+<text text-anchor="middle" x="1301" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_equal.h</text>
</a>
</g>
</g>
<!-- Node4->Node6 -->
-<g id="edge95" class="edge">
+<g id="edge99" class="edge">
<title>Node4->Node6</title>
-<path fill="none" stroke="#191970" d="M1407.6145,-559.4838C1375.7324,-550.8019 1334.3185,-537.1953 1324.5,-523 1294.6322,-479.818 1345.7832,-425.1806 1380.4987,-395.3654"/>
-<polygon fill="#191970" stroke="#191970" points="1383.0539,-397.7913 1388.4809,-388.692 1378.564,-392.4209 1383.0539,-397.7913"/>
+<path fill="none" stroke="#191970" d="M1370.2641,-561.9255C1326.6742,-554.5865 1268.1211,-541.5664 1254,-523 1224.9436,-484.7968 1258.2532,-428.2516 1281.7722,-396.786"/>
+<polygon fill="#191970" stroke="#191970" points="1284.7428,-398.6658 1288.0862,-388.6123 1279.2031,-394.3865 1284.7428,-398.6658"/>
</g>
-<!-- Node19 -->
+<!-- Node17 -->
<g id="node11" class="node">
-<title>Node19</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="450,-62 450,-81 495,-81 495,-62 450,-62"/>
-<text text-anchor="middle" x="472.5" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
+<title>Node17</title>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1204.5,-62 1204.5,-81 1249.5,-81 1249.5,-62 1204.5,-62"/>
+<text text-anchor="middle" x="1227" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
</g>
-<!-- Node4->Node19 -->
-<g id="edge102" class="edge">
-<title>Node4->Node19</title>
-<path fill="none" stroke="#191970" d="M1396.7479,-568.628C1184.8215,-566.8319 366.2717,-557.6663 324.5,-523 295.5457,-498.9708 305.5,-478.1265 305.5,-440.5 305.5,-440.5 305.5,-440.5 305.5,-373.5 305.5,-250.8247 412.3339,-131.1909 455.1043,-88.2116"/>
-<polygon fill="#191970" stroke="#191970" points="457.6527,-90.6139 462.3048,-81.095 452.7321,-85.6352 457.6527,-90.6139"/>
+<!-- Node4->Node17 -->
+<g id="edge106" class="edge">
+<title>Node4->Node17</title>
+<path fill="none" stroke="#191970" d="M1370.2523,-564.3659C1222.0362,-549.4143 788.6932,-497.019 708,-389 681.9331,-354.1058 691.3921,-331.2649 708,-291 728.072,-242.3366 827.9921,-146.6912 875,-123 930.7044,-94.9259 1117.6607,-79.0914 1194.0557,-73.6603"/>
+<polygon fill="#191970" stroke="#191970" points="1194.3436,-77.1488 1204.0764,-72.9628 1193.8575,-70.1657 1194.3436,-77.1488"/>
</g>
-<!-- Node21 -->
+<!-- Node19 -->
<g id="node12" class="node">
-<title>Node21</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="724,-235.5 724,-254.5 771,-254.5 771,-235.5 724,-235.5"/>
-<text text-anchor="middle" x="747.5" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
+<title>Node19</title>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="869.5,-235.5 869.5,-254.5 916.5,-254.5 916.5,-235.5 869.5,-235.5"/>
+<text text-anchor="middle" x="893" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
</g>
-<!-- Node4->Node21 -->
-<g id="edge103" class="edge">
-<title>Node4->Node21</title>
-<path fill="none" stroke="#191970" d="M1396.9532,-561.734C1353.1942,-554.5124 1288.0498,-541.7841 1233.5,-523 1040.861,-456.665 985.7704,-442.5212 821.5,-322 797.8224,-304.6283 775.1165,-279.2768 761.2437,-262.4698"/>
-<polygon fill="#191970" stroke="#191970" points="763.8459,-260.1215 754.836,-254.5472 758.4033,-264.5235 763.8459,-260.1215"/>
+<!-- Node4->Node19 -->
+<g id="edge107" class="edge">
+<title>Node4->Node19</title>
+<path fill="none" stroke="#191970" d="M1370.3901,-560.3546C1333.3176,-552.9552 1281.8638,-540.673 1239,-523 1087.5371,-460.5511 1031.3306,-450.7362 930,-322 916.1542,-304.4095 905.7731,-280.7944 899.5844,-264.3738"/>
+<polygon fill="#191970" stroke="#191970" points="902.8044,-262.9847 896.1286,-254.7577 896.2169,-265.3522 902.8044,-262.9847"/>
</g>
-<!-- Node26 -->
+<!-- Node24 -->
<g id="node16" class="node">
-<title>Node26</title>
+<title>Node24</title>
<g id="a_node16"><a xlink:href="runtime_2memory_8h.html" target="_top" xlink:title="Runtime memory management. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="2134,-179.5 2134,-198.5 2263,-198.5 2263,-179.5 2134,-179.5"/>
-<text text-anchor="middle" x="2198.5" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="1904.5,-179.5 1904.5,-198.5 2033.5,-198.5 2033.5,-179.5 1904.5,-179.5"/>
+<text text-anchor="middle" x="1969" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
</a>
</g>
</g>
-<!-- Node4->Node26 -->
-<g id="edge98" class="edge">
-<title>Node4->Node26</title>
-<path fill="none" stroke="#191970" d="M1496.2169,-567.4459C1634.9343,-562.8152 2019.4037,-547.8661 2070.5,-523 2175.704,-471.8023 2204.7156,-433.3937 2240.5,-322 2244.3497,-310.0161 2242.3004,-244.6344 2238.5,-235 2234.1311,-223.9245 2225.8773,-213.6998 2217.9968,-205.7171"/>
-<polygon fill="#191970" stroke="#191970" points="2220.0695,-202.86 2210.3984,-198.5332 2215.2604,-207.9466 2220.0695,-202.86"/>
+<!-- Node4->Node24 -->
+<g id="edge102" class="edge">
+<title>Node4->Node24</title>
+<path fill="none" stroke="#191970" d="M1469.7577,-567.7096C1553.6529,-564.6979 1728.1755,-554.8265 1872,-523 1960.6548,-503.3819 1984.9889,-498.9275 2065,-456 2108.3163,-432.76 2121.753,-426.9476 2153,-389 2182.214,-353.5213 2213.4761,-327.8411 2186,-291 2152.4565,-246.0235 2115.6255,-279.1891 2065,-255 2036.4847,-241.3752 2007.0144,-219.7945 1988.3782,-205.0459"/>
+<polygon fill="#191970" stroke="#191970" points="1990.3702,-202.1562 1980.386,-198.6114 1985.9804,-207.6088 1990.3702,-202.1562"/>
</g>
<!-- Node4->Node9 -->
-<g id="edge99" class="edge">
+<g id="edge103" class="edge">
<title>Node4->Node9</title>
-<path fill="none" stroke="#191970" d="M1496.0213,-567.8125C1695.7626,-562.3788 2431.5,-535.8897 2431.5,-440.5 2431.5,-440.5 2431.5,-440.5 2431.5,-373.5 2431.5,-280.5279 2428.064,-235.852 2354.5,-179 2330.8857,-160.7503 2151.3119,-145.1996 2049.4682,-137.7342"/>
-<polygon fill="#191970" stroke="#191970" points="2049.4867,-134.2264 2039.2597,-136.9938 2048.9804,-141.2081 2049.4867,-134.2264"/>
+<path fill="none" stroke="#191970" d="M1418.4589,-559.4214C1411.6354,-515.7 1385.826,-335.3124 1417,-291 1466.5518,-220.5644 1718.8177,-166.7782 1841.2112,-144.3843"/>
+<polygon fill="#191970" stroke="#191970" points="1842.0029,-147.798 1851.2183,-142.5705 1840.7543,-140.9102 1842.0029,-147.798"/>
</g>
<!-- Node10 -->
<g id="node18" class="node">
<title>Node10</title>
<g id="a_node18"><a xlink:href="c__runtime__api_8h.html" target="_top" xlink:title="tvm/runtime/c_runtime\l_api.h">
-<polygon fill="#ffffff" stroke="#000000" points="2955,-56.5 2955,-86.5 3084,-86.5 3084,-56.5 2955,-56.5"/>
-<text text-anchor="start" x="2963" y="-74.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/c_runtime</text>
-<text text-anchor="middle" x="3019.5" y="-63.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2727.5,-56.5 2727.5,-86.5 2856.5,-86.5 2856.5,-56.5 2727.5,-56.5"/>
+<text text-anchor="start" x="2735.5" y="-74.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/c_runtime</text>
+<text text-anchor="middle" x="2792" y="-63.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
</a>
</g>
</g>
<!-- Node4->Node10 -->
-<g id="edge97" class="edge">
+<g id="edge101" class="edge">
<title>Node4->Node10</title>
-<path fill="none" stroke="#191970" d="M1496.1235,-568.2014C1660.8348,-565.2827 2188.4338,-553.711 2355.5,-523 2467.9317,-502.3322 2917.8283,-324.4824 2923.5,-322 2983.7713,-295.6207 3006.6752,-299.098 3055.5,-255 3100.3774,-214.4672 3117.0435,-176.3119 3088.5,-123 3081.8655,-110.6085 3070.7957,-100.3719 3059.4438,-92.3789"/>
-<polygon fill="#191970" stroke="#191970" points="3061.0605,-89.254 3050.7768,-86.7054 3057.2266,-95.1108 3061.0605,-89.254"/>
+<path fill="none" stroke="#191970" d="M1469.5498,-567.2787C1563.0702,-563.4515 1771.7654,-552.153 1945,-523 2241.893,-473.0369 2305.3302,-409.272 2602,-358 2766.1152,-329.6368 2823.0674,-390.2258 2975,-322 3020.9028,-301.3872 3058,-295.3185 3058,-245 3058,-245 3058,-245 3058,-189 3058,-146.9397 2945.3129,-109.854 2866.6286,-89.0855"/>
+<polygon fill="#191970" stroke="#191970" points="2867.2898,-85.6411 2856.7308,-86.5112 2865.5277,-92.4157 2867.2898,-85.6411"/>
</g>
-<!-- Node4->Node17 -->
-<g id="edge100" class="edge">
-<title>Node4->Node17</title>
-<path fill="none" stroke="#191970" d="M1397.0134,-559.4609C1370.1748,-552.5403 1337.4827,-541.131 1312.5,-523 1224.5963,-459.2045 1204.327,-426.1472 1173.5,-322 1149.0247,-239.3118 1161.2636,-134.4371 1168.0266,-91.3495"/>
-<polygon fill="#191970" stroke="#191970" points="1171.5037,-91.775 1169.6784,-81.3385 1164.5971,-90.6353 1171.5037,-91.775"/>
+<!-- Node4->Node15 -->
+<g id="edge104" class="edge">
+<title>Node4->Node15</title>
+<path fill="none" stroke="#191970" d="M1370.2615,-568.5456C1179.0582,-566.5531 497.5051,-557.1031 407,-523 352.1152,-502.3189 304,-499.1519 304,-440.5 304,-440.5 304,-440.5 304,-189 304,-152.7454 320.5805,-113.1181 331.837,-90.3929"/>
+<polygon fill="#191970" stroke="#191970" points="335.0965,-91.7076 336.5555,-81.2137 328.8708,-88.5073 335.0965,-91.7076"/>
</g>
-<!-- Node4->Node18 -->
-<g id="edge101" class="edge">
-<title>Node4->Node18</title>
-<path fill="none" stroke="#191970" d="M1496.0624,-568.3429C1650.9997,-565.9593 2124.7823,-556.0853 2274.5,-523 2482.8515,-476.9576 2553.7447,-466.769 2710.5,-322 2771.7377,-265.4448 2763.5286,-226.3049 2766.5,-143 2766.8169,-134.1168 2768.9393,-131.5476 2766.5,-123 2763.0413,-110.8801 2755.8178,-98.8316 2749.1078,-89.439"/>
-<polygon fill="#191970" stroke="#191970" points="2751.7974,-87.193 2742.9714,-81.3318 2746.2159,-91.4177 2751.7974,-87.193"/>
+<!-- Node4->Node16 -->
+<g id="edge105" class="edge">
+<title>Node4->Node16</title>
+<path fill="none" stroke="#191970" d="M1469.6466,-568.427C1574.4639,-566.5281 1825.7469,-558.3365 2033,-523 2258.9452,-484.4765 2310.4858,-451.5205 2531,-389 2577.403,-375.8438 2587.5798,-366.8118 2635,-358 2736.2447,-339.1863 3013.5705,-382.3652 3097,-322 3168.6875,-270.1307 3155.5382,-183.5339 3091,-123 3068.1492,-101.567 3035.205,-88.5574 3008.5966,-80.9762"/>
+<polygon fill="#191970" stroke="#191970" points="3009.3087,-77.5435 2998.7431,-78.3303 3007.4932,-84.304 3009.3087,-77.5435"/>
</g>
-<!-- Node30 -->
-<g id="node29" class="node">
-<title>Node30</title>
-<g id="a_node29"><a xlink:href="structural__hash_8h.html" target="_top" xlink:title="tvm/node/structural\l_hash.h">
-<polygon fill="#ffffff" stroke="#ff0000" points="1437,-425.5 1437,-455.5 1550,-455.5 1550,-425.5 1437,-425.5"/>
... 474946 lines suppressed ...