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 2022/09/20 23:44:41 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@534378b935aa08b77e7529ec183133a24f121ae4)
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 55c6cf2ee3 deploying docs (apache/tvm@534378b935aa08b77e7529ec183133a24f121ae4)
55c6cf2ee3 is described below
commit 55c6cf2ee304a23687c30a9df7fe56f6ced1ae55
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Tue Sep 20 23:44:28 2022 +0000
deploying docs (apache/tvm@534378b935aa08b77e7529ec183133a24f121ae4)
---
.../how_to/compile_models/from_darknet.rst.txt | 2 +-
.../how_to/compile_models/from_keras.rst.txt | 2 +-
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.rst.txt | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 2 +-
.../compile_models/sg_execution_times.rst.txt | 22 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 20 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 2 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 8 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 14 +-
.../tune_conv2d_layer_cuda.rst.txt | 1285 +--
.../tune_network_cuda.rst.txt | 2 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 86 +-
.../tune_with_autotvm/sg_execution_times.rst.txt | 6 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 26 +-
.../work_with_microtvm/micro_autotune.rst.txt | 16 +-
.../how_to/work_with_microtvm/micro_train.rst.txt | 16 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 10 +-
.../work_with_relay/sg_execution_times.rst.txt | 8 +-
.../how_to/work_with_schedules/intrin_math.rst.txt | 2 +-
.../work_with_schedules/sg_execution_times.rst.txt | 14 +-
.../how_to/work_with_schedules/tensorize.rst.txt | 2 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 4 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 2 +-
docs/_sources/tutorial/autotvm_matmul_x86.rst.txt | 20 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 58 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 24 +-
.../tutorial/tensor_expr_get_started.rst.txt | 46 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_darknet.html | 2 +-
docs/how_to/compile_models/from_keras.html | 2 +-
docs/how_to/compile_models/from_mxnet.html | 2 +-
docs/how_to/compile_models/from_oneflow.html | 17 +-
docs/how_to/compile_models/from_pytorch.html | 7 +-
docs/how_to/compile_models/from_tensorflow.html | 2 +-
docs/how_to/compile_models/sg_execution_times.html | 26 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 18 +-
docs/how_to/deploy_models/deploy_prequantized.html | 11 +-
.../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 | 37 +-
docs/how_to/deploy_models/sg_execution_times.html | 20 +-
.../extend_tvm/bring_your_own_datatypes.html | 2 +-
docs/how_to/extend_tvm/sg_execution_times.html | 8 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 1285 +--
.../tune_with_autoscheduler/tune_network_cuda.html | 2 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 86 +-
.../tune_with_autotvm/sg_execution_times.html | 6 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 26 +-
docs/how_to/work_with_microtvm/micro_autotune.html | 16 +-
docs/how_to/work_with_microtvm/micro_train.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 10 +-
.../how_to/work_with_relay/sg_execution_times.html | 8 +-
docs/how_to/work_with_schedules/intrin_math.html | 2 +-
.../work_with_schedules/sg_execution_times.html | 14 +-
docs/how_to/work_with_schedules/tensorize.html | 2 +-
docs/install/nnpack.html | 12 +-
docs/reference/api/doxygen/affine__type_8h.html | 2 +-
.../api/doxygen/affine__type_8h__incl.svg | 1084 +-
.../api/doxygen/affine__type_8h_source.html | 2 +-
docs/reference/api/doxygen/algorithm_8h__incl.svg | 1482 +--
docs/reference/api/doxygen/algorithms_8h.html | 2 +-
docs/reference/api/doxygen/algorithms_8h__incl.svg | 1412 +--
docs/reference/api/doxygen/analyzer_8h.html | 2 +-
docs/reference/api/doxygen/analyzer_8h__incl.svg | 1188 ++-
docs/reference/api/doxygen/annotated.html | 118 +-
docs/reference/api/doxygen/annotation_8h.html | 2 +-
docs/reference/api/doxygen/annotation_8h__incl.svg | 1136 +-
docs/reference/api/doxygen/arg__info_8h.html | 2 +-
docs/reference/api/doxygen/arg__info_8h__incl.svg | 1530 +--
.../reference/api/doxygen/arg__info_8h_source.html | 2 +-
docs/reference/api/doxygen/array_8h.html | 18 +-
docs/reference/api/doxygen/array_8h__dep__incl.svg | 776 +-
docs/reference/api/doxygen/array_8h__incl.svg | 377 +-
docs/reference/api/doxygen/array_8h_source.html | 111 +-
docs/reference/api/doxygen/array__utils_8h.html | 2 +-
.../api/doxygen/array__utils_8h__incl.svg | 1426 ++-
docs/reference/api/doxygen/auto__schedule_8h.html | 2 +-
.../api/doxygen/auto__schedule_8h__incl.svg | 1274 +--
.../api/doxygen/auto__schedule_8h_source.html | 2 +-
.../doxygen/auto__scheduler_2cost__model_8h.html | 2 +-
.../auto__scheduler_2cost__model_8h__incl.svg | 1214 +--
.../auto__scheduler_2cost__model_8h_source.html | 2 +-
.../api/doxygen/auto__scheduler_2feature_8h.html | 2 +-
.../doxygen/auto__scheduler_2feature_8h__incl.svg | 1252 +--
docs/reference/api/doxygen/autodiff_8h.html | 2 +-
docs/reference/api/doxygen/autodiff_8h__incl.svg | 1276 +--
docs/reference/api/doxygen/bias__add_8h.html | 2 +-
docs/reference/api/doxygen/bias__add_8h__incl.svg | 1356 +--
docs/reference/api/doxygen/bitserial_8h.html | 2 +-
docs/reference/api/doxygen/bitserial_8h__incl.svg | 1296 +--
.../reference/api/doxygen/bitserial_8h_source.html | 2 +-
docs/reference/api/doxygen/block__scope_8h.html | 2 +-
.../api/doxygen/block__scope_8h__incl.svg | 1230 +--
.../api/doxygen/block__scope_8h_source.html | 2 +-
docs/reference/api/doxygen/bound_8h.html | 2 +-
docs/reference/api/doxygen/bound_8h__incl.svg | 1206 +--
docs/reference/api/doxygen/broadcast_8h.html | 2 +-
docs/reference/api/doxygen/broadcast_8h__incl.svg | 1420 +--
.../reference/api/doxygen/broadcast_8h_source.html | 6 +-
docs/reference/api/doxygen/buffer_8h.html | 2 +-
docs/reference/api/doxygen/buffer_8h__incl.svg | 1118 +-
docs/reference/api/doxygen/buffer_8h_source.html | 4 +-
docs/reference/api/doxygen/builder_8h.html | 2 +-
docs/reference/api/doxygen/builder_8h__incl.svg | 1636 +--
docs/reference/api/doxygen/builder_8h_source.html | 2 +-
docs/reference/api/doxygen/builtin_8h.html | 2 +-
docs/reference/api/doxygen/builtin_8h__incl.svg | 1372 +--
docs/reference/api/doxygen/call_8h.html | 2 +-
docs/reference/api/doxygen/call_8h__incl.svg | 1136 +-
docs/reference/api/doxygen/classes.html | 495 +-
...sstvm_1_1CompilationConfigNode__coll__graph.svg | 4 +-
.../classtvm_1_1ConstructorNode__coll__graph.svg | 4 +-
...sstvm_1_1DiagnosticContextNode__coll__graph.svg | 4 +-
.../classtvm_1_1FuncTypeNode__coll__graph.svg | 12 +-
.../doxygen/classtvm_1_1OpNode__coll__graph.svg | 4 +-
.../classtvm_1_1TargetKindNode__coll__graph.svg | 4 +-
.../classtvm_1_1TargetNode__coll__graph.svg | 4 +-
.../classtvm_1_1TensorTypeNode__coll__graph.svg | 4 +-
...lasstvm_1_1TupleAffineTypeNode__coll__graph.svg | 4 +-
.../classtvm_1_1TupleTypeNode__coll__graph.svg | 4 +-
.../classtvm_1_1TypeCallNode__coll__graph.svg | 4 +-
.../classtvm_1_1TypeDataNode__coll__graph.svg | 8 +-
.../classtvm_1_1TypeRelationNode__coll__graph.svg | 4 +-
...1_1arith_1_1IntConstraintsNode__coll__graph.svg | 8 +-
...1_1arith_1_1IntGroupBoundsNode__coll__graph.svg | 4 +-
..._1_1arith_1_1IterMapResultNode__coll__graph.svg | 8 +-
...vm_1_1arith_1_1IterSumExprNode__coll__graph.svg | 4 +-
...cheduler_1_1AccessAnalyzerNode__coll__graph.svg | 4 +-
...__scheduler_1_1BuildResultNode__coll__graph.svg | 4 +-
...scheduler_1_1CacheReadStepNode__coll__graph.svg | 4 +-
...o__scheduler_1_1ComputeDAGNode__coll__graph.svg | 8 +-
...er_1_1FollowFusedSplitStepNode__coll__graph.svg | 4 +-
...uto__scheduler_1_1FuseStepNode__coll__graph.svg | 4 +-
...scheduler_1_1MeasureResultNode__coll__graph.svg | 4 +-
...__scheduler_1_1ReorderStepNode__coll__graph.svg | 4 +-
...o__scheduler_1_1SearchTaskNode__coll__graph.svg | 4 +-
...to__scheduler_1_1SplitStepNode__coll__graph.svg | 4 +-
..._1auto__scheduler_1_1StageNode__coll__graph.svg | 4 +-
..._1auto__scheduler_1_1StateNode__coll__graph.svg | 8 +-
...vm_1_1detail_1_1AttrDocVisitor__coll__graph.svg | 4 +-
..._schedule_1_1ExtractedTaskNode__coll__graph.svg | 4 +-
...hedule_1_1MeasureCandidateNode__coll__graph.svg | 4 +-
...chedule_1_1PyTaskSchedulerNode__coll__graph.svg | 8 +-
...a__schedule_1_1RunnerInputNode__coll__graph.svg | 4 +-
..._schedule_1_1TaskSchedulerNode__coll__graph.svg | 8 +-
...a__schedule_1_1TuneContextNode__coll__graph.svg | 8 +-
.../classtvm_1_1relay_1_1CallNode__coll__graph.svg | 8 +-
...vm_1_1relay_1_1CallPatternNode__coll__graph.svg | 4 +-
...sstvm_1_1relay_1_1FunctionNode__coll__graph.svg | 8 +-
..._1relay_1_1FunctionPatternNode__coll__graph.svg | 4 +-
...classtvm_1_1relay_1_1MatchNode__coll__graph.svg | 4 +-
...1relay_1_1OpSpecializationNode__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1OpStrategyNode__coll__graph.svg | 4 +-
...elay_1_1PatternConstructorNode__coll__graph.svg | 4 +-
...m_1_1relay_1_1PatternTupleNode__coll__graph.svg | 4 +-
...m_1_1relay_1_1ShapePatternNode__coll__graph.svg | 4 +-
...classtvm_1_1relay_1_1TupleNode__coll__graph.svg | 4 +-
...m_1_1relay_1_1TuplePatternNode__coll__graph.svg | 4 +-
.../classtvm_1_1runtime_1_1Array-members.html | 47 +-
.../api/doxygen/classtvm_1_1runtime_1_1Array.html | 65 +-
.../classtvm_1_1runtime_1_1Array__coll__graph.svg | 2 +-
...lasstvm_1_1runtime_1_1Array__inherit__graph.svg | 2 +-
...1metadata_1_1MetadataArrayNode__coll__graph.svg | 4 +-
...ime_1_1profiling_1_1ReportNode__coll__graph.svg | 4 +-
..._1ir__builder_1_1IRBuilderNode__coll__graph.svg | 4 +-
...der_1_1ir_1_1IRModuleFrameNode__coll__graph.svg | 8 +-
...1tir_1_1AllocateConstFrameNode__coll__graph.svg | 8 +-
...er_1_1tir_1_1AllocateFrameNode__coll__graph.svg | 8 +-
...lder_1_1tir_1_1AssertFrameNode__coll__graph.svg | 4 +-
...uilder_1_1tir_1_1AttrFrameNode__coll__graph.svg | 4 +-
...ilder_1_1tir_1_1BlockFrameNode__coll__graph.svg | 20 +-
...r_1_1tir_1_1BlockInitFrameNode__coll__graph.svg | 4 +-
..._1_1tir_1_1DeclBufferFrameNode__coll__graph.svg | 4 +-
...uilder_1_1tir_1_1ElseFrameNode__coll__graph.svg | 4 +-
...builder_1_1tir_1_1ForFrameNode__coll__graph.svg | 12 +-
..._builder_1_1tir_1_1IfFrameNode__coll__graph.svg | 4 +-
..._1tir_1_1LaunchThreadFrameNode__coll__graph.svg | 4 +-
...builder_1_1tir_1_1LetFrameNode__coll__graph.svg | 4 +-
...er_1_1tir_1_1PrimFuncFrameNode__coll__graph.svg | 12 +-
...der_1_1tir_1_1RealizeFrameNode__coll__graph.svg | 4 +-
...builder_1_1tir_1_1TIRFrameNode__coll__graph.svg | 4 +-
...uilder_1_1tir_1_1ThenFrameNode__coll__graph.svg | 4 +-
...ilder_1_1tir_1_1WhileFrameNode__coll__graph.svg | 4 +-
...pt_1_1printer_1_1AssertDocNode__coll__graph.svg | 4 +-
...pt_1_1printer_1_1AssignDocNode__coll__graph.svg | 4 +-
..._1printer_1_1AttrAccessDocNode__coll__graph.svg | 4 +-
...ript_1_1printer_1_1CallDocNode__coll__graph.svg | 12 +-
...ipt_1_1printer_1_1ClassDocNode__coll__graph.svg | 12 +-
...ript_1_1printer_1_1DictDocNode__coll__graph.svg | 8 +-
..._1script_1_1printer_1_1DocNode__coll__graph.svg | 4 +-
...ript_1_1printer_1_1ExprDocNode__coll__graph.svg | 4 +-
..._1_1printer_1_1ExprStmtDocNode__coll__graph.svg | 4 +-
...cript_1_1printer_1_1ForDocNode__coll__graph.svg | 8 +-
..._1_1printer_1_1FunctionDocNode__coll__graph.svg | 16 +-
..._1_1printer_1_1IRDocsifierNode__coll__graph.svg | 8 +-
...script_1_1printer_1_1IdDocNode__coll__graph.svg | 4 +-
...script_1_1printer_1_1IfDocNode__coll__graph.svg | 8 +-
...ipt_1_1printer_1_1IndexDocNode__coll__graph.svg | 8 +-
...pt_1_1printer_1_1LambdaDocNode__coll__graph.svg | 8 +-
...ript_1_1printer_1_1ListDocNode__coll__graph.svg | 8 +-
...t_1_1printer_1_1LiteralDocNode__coll__graph.svg | 4 +-
..._1printer_1_1MetadataFrameNode__coll__graph.svg | 4 +-
...1_1printer_1_1OperationDocNode__coll__graph.svg | 8 +-
...pt_1_1printer_1_1ReturnDocNode__coll__graph.svg | 4 +-
...ipt_1_1printer_1_1ScopeDocNode__coll__graph.svg | 8 +-
...ipt_1_1printer_1_1SliceDocNode__coll__graph.svg | 4 +-
...1_1printer_1_1StmtBlockDocNode__coll__graph.svg | 8 +-
...ript_1_1printer_1_1StmtDocNode__coll__graph.svg | 4 +-
...ipt_1_1printer_1_1TupleDocNode__coll__graph.svg | 8 +-
..._1_1printer_1_1VarDefFrameNode__coll__graph.svg | 4 +-
...ipt_1_1printer_1_1WhileDocNode__coll__graph.svg | 8 +-
...tvm_1_1te_1_1BaseComputeOpNode__coll__graph.svg | 4 +-
...lasstvm_1_1te_1_1ComputeOpNode__coll__graph.svg | 8 +-
...classtvm_1_1te_1_1ExternOpNode__coll__graph.svg | 8 +-
...classtvm_1_1te_1_1HybridOpNode__coll__graph.svg | 8 +-
...sstvm_1_1te_1_1IterVarAttrNode__coll__graph.svg | 8 +-
...tvm_1_1te_1_1PlaceholderOpNode__coll__graph.svg | 4 +-
.../classtvm_1_1te_1_1ScanOpNode__coll__graph.svg | 8 +-
...classtvm_1_1te_1_1ScheduleNode__coll__graph.svg | 8 +-
...te_1_1SpecializedConditionNode__coll__graph.svg | 4 +-
.../classtvm_1_1te_1_1StageNode__coll__graph.svg | 16 +-
...m_1_1te_1_1TensorComputeOpNode__coll__graph.svg | 16 +-
..._1_1te_1_1TensorIntrinCallNode__coll__graph.svg | 16 +-
...stvm_1_1te_1_1TensorIntrinNode__coll__graph.svg | 12 +-
.../classtvm_1_1te_1_1TensorNode__coll__graph.svg | 4 +-
...lasstvm_1_1te_1_1TransformNode__coll__graph.svg | 4 +-
...vm_1_1tir_1_1AllocateConstNode__coll__graph.svg | 4 +-
...lasstvm_1_1tir_1_1AllocateNode__coll__graph.svg | 4 +-
..._1_1tir_1_1BijectiveLayoutNode__coll__graph.svg | 4 +-
.../classtvm_1_1tir_1_1BlockNode__coll__graph.svg | 16 +-
...tvm_1_1tir_1_1BlockRealizeNode__coll__graph.svg | 4 +-
...sstvm_1_1tir_1_1BufferLoadNode__coll__graph.svg | 4 +-
.../classtvm_1_1tir_1_1BufferNode__coll__graph.svg | 8 +-
...vm_1_1tir_1_1BufferRealizeNode__coll__graph.svg | 4 +-
...tvm_1_1tir_1_1BufferRegionNode__coll__graph.svg | 4 +-
...stvm_1_1tir_1_1BufferStoreNode__coll__graph.svg | 4 +-
.../classtvm_1_1tir_1_1CallNode__coll__graph.svg | 4 +-
...stvm_1_1tir_1_1CommReducerNode__coll__graph.svg | 8 +-
...lasstvm_1_1tir_1_1IndexMapNode__coll__graph.svg | 8 +-
...stvm_1_1tir_1_1InstructionNode__coll__graph.svg | 4 +-
.../classtvm_1_1tir_1_1LayoutNode__coll__graph.svg | 4 +-
...lasstvm_1_1tir_1_1PrefetchNode__coll__graph.svg | 4 +-
...lasstvm_1_1tir_1_1PrimFuncNode__coll__graph.svg | 4 +-
...tvm_1_1tir_1_1ProducerLoadNode__coll__graph.svg | 4 +-
..._1_1tir_1_1ProducerRealizeNode__coll__graph.svg | 4 +-
...vm_1_1tir_1_1ProducerStoreNode__coll__graph.svg | 4 +-
.../classtvm_1_1tir_1_1ReduceNode__coll__graph.svg | 8 +-
...classtvm_1_1tir_1_1SeqStmtNode__coll__graph.svg | 4 +-
...classtvm_1_1tir_1_1ShuffleNode__coll__graph.svg | 4 +-
.../classtvm_1_1tir_1_1TraceNode__coll__graph.svg | 4 +-
..._1transform_1_1PassContextNode__coll__graph.svg | 8 +-
...m_1_1transform_1_1PassInfoNode__coll__graph.svg | 4 +-
...1_1transform_1_1SequentialNode__coll__graph.svg | 4 +-
docs/reference/api/doxygen/codegen_8h.html | 2 +-
docs/reference/api/doxygen/codegen_8h__incl.svg | 1488 +--
.../api/doxygen/compilation__config_8h.html | 2 +-
.../api/doxygen/compilation__config_8h__incl.svg | 1466 +--
.../api/doxygen/compilation__config_8h_source.html | 2 +-
docs/reference/api/doxygen/compute__dag_8h.html | 2 +-
.../api/doxygen/compute__dag_8h__incl.svg | 1288 +--
.../api/doxygen/compute__dag_8h_source.html | 2 +-
docs/reference/api/doxygen/constant__utils_8h.html | 2 +-
.../api/doxygen/constant__utils_8h__incl.svg | 1440 +--
docs/reference/api/doxygen/conv2d_8h.html | 2 +-
docs/reference/api/doxygen/conv2d_8h__incl.svg | 686 +-
docs/reference/api/doxygen/cublas_8h.html | 2 +-
docs/reference/api/doxygen/cublas_8h__incl.svg | 1453 +--
docs/reference/api/doxygen/cublas_8h_source.html | 2 +-
docs/reference/api/doxygen/cuda_2dense_8h.html | 2 +-
.../reference/api/doxygen/cuda_2dense_8h__incl.svg | 1658 +--
.../api/doxygen/cuda_2dense_8h_source.html | 4 +-
docs/reference/api/doxygen/cuda_2injective_8h.html | 2 +-
.../api/doxygen/cuda_2injective_8h__incl.svg | 1728 +--
.../api/doxygen/cuda_2injective_8h_source.html | 4 +-
docs/reference/api/doxygen/cuda_2pooling_8h.html | 2 +-
.../api/doxygen/cuda_2pooling_8h__incl.svg | 1686 +--
.../api/doxygen/cuda_2pooling_8h_source.html | 4 +-
docs/reference/api/doxygen/cuda_2reduction_8h.html | 2 +-
.../api/doxygen/cuda_2reduction_8h__incl.svg | 1728 +--
.../api/doxygen/cuda_2reduction_8h_source.html | 6 +-
docs/reference/api/doxygen/cuda_2softmax_8h.html | 2 +-
.../api/doxygen/cuda_2softmax_8h__incl.svg | 1728 +--
.../api/doxygen/cuda_2softmax_8h_source.html | 4 +-
docs/reference/api/doxygen/data__layout_8h.html | 2 +-
.../api/doxygen/data__layout_8h__incl.svg | 1310 +--
.../api/doxygen/data__layout_8h_source.html | 4 +-
docs/reference/api/doxygen/database_8h.html | 2 +-
docs/reference/api/doxygen/database_8h__incl.svg | 1630 +--
docs/reference/api/doxygen/database_8h_source.html | 2 +-
.../api/doxygen/dataflow__matcher_8h__incl.svg | 1544 +--
.../api/doxygen/dataflow__matcher_8h_source.html | 2 +-
.../api/doxygen/dataflow__pattern_8h.html | 2 +-
.../api/doxygen/dataflow__pattern_8h__incl.svg | 1566 +--
.../api/doxygen/dataflow__pattern_8h_source.html | 2 +-
.../api/doxygen/dataflow__pattern__functor_8h.html | 2 +-
.../dataflow__pattern__functor_8h__incl.svg | 1548 +--
.../api/doxygen/detail_2broadcast_8h.html | 2 +-
.../api/doxygen/detail_2broadcast_8h__incl.svg | 1418 +--
.../api/doxygen/detail_2broadcast_8h_source.html | 10 +-
docs/reference/api/doxygen/detail_2extern_8h.html | 2 +-
.../api/doxygen/detail_2extern_8h__incl.svg | 1366 +--
docs/reference/api/doxygen/device__api_8h.html | 2 +-
.../reference/api/doxygen/device__api_8h__incl.svg | 662 +-
.../api/doxygen/device__copy_8h__incl.svg | 1298 +--
docs/reference/api/doxygen/diagnostic_8h.html | 2 +-
docs/reference/api/doxygen/diagnostic_8h__incl.svg | 1488 +--
.../api/doxygen/diagnostic_8h_source.html | 2 +-
docs/reference/api/doxygen/dilate_8h.html | 2 +-
docs/reference/api/doxygen/dilate_8h__incl.svg | 1342 +--
docs/reference/api/doxygen/dilate_8h_source.html | 6 +-
docs/reference/api/doxygen/doc_8h.html | 2 +-
docs/reference/api/doxygen/doc_8h__incl.svg | 1122 +-
docs/reference/api/doxygen/doc_8h_source.html | 4 +-
docs/reference/api/doxygen/doc__printer_8h.html | 2 +-
.../api/doxygen/doc__printer_8h__incl.svg | 1122 +-
docs/reference/api/doxygen/driver__api_8h.html | 2 +-
.../reference/api/doxygen/driver__api_8h__incl.svg | 1520 +--
docs/reference/api/doxygen/einsum_8h.html | 2 +-
docs/reference/api/doxygen/einsum_8h__incl.svg | 1320 +--
docs/reference/api/doxygen/einsum_8h_source.html | 12 +-
docs/reference/api/doxygen/elemwise_8h.html | 2 +-
docs/reference/api/doxygen/elemwise_8h__incl.svg | 1308 +--
docs/reference/api/doxygen/elemwise_8h_source.html | 4 +-
docs/reference/api/doxygen/env__func_8h.html | 2 +-
docs/reference/api/doxygen/env__func_8h__incl.svg | 886 +-
docs/reference/api/doxygen/error_8h.html | 2 +-
docs/reference/api/doxygen/error_8h__incl.svg | 1500 +--
docs/reference/api/doxygen/executable_8h.html | 2 +-
docs/reference/api/doxygen/executable_8h__incl.svg | 744 +-
docs/reference/api/doxygen/executor_8h.html | 2 +-
docs/reference/api/doxygen/executor_8h__incl.svg | 1582 +--
docs/reference/api/doxygen/executor_8h_source.html | 2 +-
docs/reference/api/doxygen/extracted__task_8h.html | 2 +-
.../api/doxygen/extracted__task_8h__incl.svg | 1350 +--
.../api/doxygen/extracted__task_8h_source.html | 2 +-
.../api/doxygen/feature__extractor_8h.html | 2 +-
.../api/doxygen/feature__extractor_8h__incl.svg | 1384 +--
.../api/doxygen/feature__extractor_8h_source.html | 2 +-
docs/reference/api/doxygen/flatten_8h.html | 2 +-
docs/reference/api/doxygen/flatten_8h__incl.svg | 1424 +--
docs/reference/api/doxygen/flatten_8h_source.html | 2 +-
docs/reference/api/doxygen/functions_func_l.html | 4 +-
docs/reference/api/doxygen/functions_func_m.html | 7 +-
docs/reference/api/doxygen/functions_func_s.html | 2 +-
docs/reference/api/doxygen/functions_func_u.html | 2 +-
docs/reference/api/doxygen/functions_m.html | 11 +-
docs/reference/api/doxygen/functions_s.html | 4 +-
docs/reference/api/doxygen/functions_t.html | 8 +-
docs/reference/api/doxygen/functions_u.html | 2 +-
docs/reference/api/doxygen/functions_v.html | 8 +-
docs/reference/api/doxygen/fuse_8h.html | 2 +-
docs/reference/api/doxygen/fuse_8h__incl.svg | 1426 ++-
.../reference/api/doxygen/generic_2default_8h.html | 2 +-
.../api/doxygen/generic_2default_8h__incl.svg | 1728 +--
.../api/doxygen/generic_2default_8h_source.html | 4 +-
docs/reference/api/doxygen/generic_2extern_8h.html | 2 +-
.../api/doxygen/generic_2extern_8h__incl.svg | 1706 +--
.../api/doxygen/generic_2extern_8h_source.html | 4 +-
.../api/doxygen/generic_2injective_8h.html | 2 +-
.../api/doxygen/generic_2injective_8h__incl.svg | 1728 +--
.../api/doxygen/generic_2injective_8h_source.html | 4 +-
docs/reference/api/doxygen/generic__func_8h.html | 2 +-
.../api/doxygen/generic__func_8h__incl.svg | 1716 +--
.../api/doxygen/global__var__supply_8h.html | 2 +-
.../api/doxygen/global__var__supply_8h__incl.svg | 1296 +--
.../api/doxygen/global__var__supply_8h_source.html | 2 +-
docs/reference/api/doxygen/greedy_8h.html | 2 +-
docs/reference/api/doxygen/greedy_8h__incl.svg | 1348 +--
docs/reference/api/doxygen/greedy_8h_source.html | 2 +-
docs/reference/api/doxygen/hierarchy.html | 2853 ++---
docs/reference/api/doxygen/image_8h.html | 2 +-
docs/reference/api/doxygen/image_8h__incl.svg | 1296 +--
docs/reference/api/doxygen/image_8h_source.html | 2 +-
docs/reference/api/doxygen/index__map_8h.html | 2 +-
docs/reference/api/doxygen/index__map_8h__incl.svg | 1116 +-
.../api/doxygen/index__map_8h_source.html | 2 +-
docs/reference/api/doxygen/inherit_graph_101.svg | 47 +-
docs/reference/api/doxygen/inherit_graph_102.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_103.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_104.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_105.svg | 18 +-
docs/reference/api/doxygen/inherit_graph_106.svg | 18 +-
docs/reference/api/doxygen/inherit_graph_107.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_108.svg | 4 +-
docs/reference/api/doxygen/inherit_graph_109.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_110.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_111.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_112.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_113.svg | 4 +-
docs/reference/api/doxygen/inherit_graph_114.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_115.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_116.svg | 27 +-
docs/reference/api/doxygen/inherit_graph_117.svg | 10642 +------------------
docs/reference/api/doxygen/inherit_graph_118.svg | 10642 ++++++++++++++++++-
docs/reference/api/doxygen/inherit_graph_119.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_120.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_121.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_122.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_123.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_124.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_125.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_126.svg | 7595 +------------
docs/reference/api/doxygen/inherit_graph_127.svg | 7596 ++++++++++++-
docs/reference/api/doxygen/inherit_graph_128.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_129.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_130.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_131.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_132.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_133.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_134.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_135.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_136.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_137.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_138.svg | 4 +-
docs/reference/api/doxygen/inherit_graph_139.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_140.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_141.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_142.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_143.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_144.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_145.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_146.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_147.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_148.svg | 61 +-
docs/reference/api/doxygen/inherit_graph_149.svg | 62 +-
docs/reference/api/doxygen/inherit_graph_150.svg | 21 +-
docs/reference/api/doxygen/inherit_graph_151.svg | 21 +-
docs/reference/api/doxygen/inherit_graph_152.svg | 20 +-
docs/reference/api/doxygen/inherit_graph_153.svg | 19 +-
docs/reference/api/doxygen/inherit_graph_154.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_155.svg | 20 +-
docs/reference/api/doxygen/inherit_graph_156.svg | 19 +-
docs/reference/api/doxygen/inherit_graph_157.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_158.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_159.svg | 4 +-
docs/reference/api/doxygen/inherit_graph_160.svg | 4 +-
docs/reference/api/doxygen/inherit_graph_161.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_162.svg | 19 +-
docs/reference/api/doxygen/inherit_graph_163.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_164.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_165.svg | 18 +-
docs/reference/api/doxygen/inherit_graph_166.svg | 4 +-
docs/reference/api/doxygen/inherit_graph_167.svg | 4 +-
docs/reference/api/doxygen/inherit_graph_168.svg | 21 +-
docs/reference/api/doxygen/inherit_graph_169.svg | 21 +-
docs/reference/api/doxygen/inherit_graph_170.svg | 19 +-
docs/reference/api/doxygen/inherit_graph_171.svg | 18 +-
docs/reference/api/doxygen/inherit_graph_172.svg | 19 +-
docs/reference/api/doxygen/inherit_graph_173.svg | 21 +-
docs/reference/api/doxygen/inherit_graph_174.svg | 23 +-
docs/reference/api/doxygen/inherit_graph_175.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_176.svg | 18 +-
docs/reference/api/doxygen/inherit_graph_177.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_178.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_179.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_180.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_181.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_182.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_183.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_184.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_185.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_186.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_187.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_188.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_189.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_190.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_191.svg | 16 +-
docs/reference/api/doxygen/inherit_graph_192.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_193.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_194.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_195.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_196.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_197.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_198.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_199.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_200.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_201.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_202.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_203.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_204.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_205.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_206.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_207.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_208.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_209.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_210.svg | 17 +-
docs/reference/api/doxygen/inherit_graph_211.svg | 80 +-
docs/reference/api/doxygen/inherit_graph_212.svg | 70 +-
docs/reference/api/doxygen/inherit_graph_213.svg | 79 +-
docs/reference/api/doxygen/inherit_graph_214.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_215.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_216.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_217.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_218.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_219.svg | 14 +-
docs/reference/api/doxygen/inherit_graph_220.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_221.svg | 18 +-
docs/reference/api/doxygen/inherit_graph_222.svg | 19 +-
docs/reference/api/doxygen/inherit_graph_223.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_224.svg | 15 +-
docs/reference/api/doxygen/inherit_graph_225.svg | 29 +-
docs/reference/api/doxygen/inherit_graph_226.svg | 24 +-
docs/reference/api/doxygen/inherit_graph_227.svg | 30 +-
docs/reference/api/doxygen/inherit_graph_228.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_229.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_230.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_231.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_232.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_233.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_234.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_235.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_236.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_237.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_238.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_239.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_240.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_241.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_242.svg | 12 +-
docs/reference/api/doxygen/inherit_graph_243.svg | 12 +-
...inherit_graph_243.svg => inherit_graph_244.svg} | 0
docs/reference/api/doxygen/inherits.html | 274 +-
docs/reference/api/doxygen/instruction_8h.html | 2 +-
.../reference/api/doxygen/instruction_8h__incl.svg | 880 +-
.../api/doxygen/instruction_8h_source.html | 2 +-
docs/reference/api/doxygen/instrument_8h.html | 2 +-
docs/reference/api/doxygen/instrument_8h__incl.svg | 890 +-
docs/reference/api/doxygen/int__set_8h.html | 2 +-
docs/reference/api/doxygen/int__set_8h__incl.svg | 1216 +--
docs/reference/api/doxygen/int__set_8h_source.html | 2 +-
docs/reference/api/doxygen/int__solver_8h.html | 2 +-
.../reference/api/doxygen/int__solver_8h__incl.svg | 1326 +--
.../api/doxygen/int__solver_8h_source.html | 2 +-
docs/reference/api/doxygen/interpreter_8h.html | 2 +-
.../reference/api/doxygen/interpreter_8h__incl.svg | 1386 +--
.../api/doxygen/interpreter_8h_source.html | 2 +-
docs/reference/api/doxygen/ir_2adt_8h.html | 2 +-
docs/reference/api/doxygen/ir_2adt_8h__incl.svg | 1132 +-
docs/reference/api/doxygen/ir_2adt_8h_source.html | 2 +-
docs/reference/api/doxygen/ir_2attrs_8h.html | 2 +-
docs/reference/api/doxygen/ir_2attrs_8h__incl.svg | 1124 +-
.../reference/api/doxygen/ir_2attrs_8h_source.html | 2 +-
docs/reference/api/doxygen/ir_2expr_8h.html | 2 +-
docs/reference/api/doxygen/ir_2expr_8h__incl.svg | 1068 +-
docs/reference/api/doxygen/ir_2function_8h.html | 2 +-
.../api/doxygen/ir_2function_8h__incl.svg | 1162 +-
docs/reference/api/doxygen/ir_2ir_8h.html | 2 +-
docs/reference/api/doxygen/ir_2ir_8h__incl.svg | 1200 +--
docs/reference/api/doxygen/ir_2module_8h.html | 2 +-
docs/reference/api/doxygen/ir_2module_8h__incl.svg | 1396 +--
.../api/doxygen/ir_2module_8h_source.html | 2 +-
docs/reference/api/doxygen/ir_2op_8h.html | 2 +-
docs/reference/api/doxygen/ir_2op_8h__incl.svg | 1576 +--
docs/reference/api/doxygen/ir_2op_8h_source.html | 6 +-
docs/reference/api/doxygen/ir_2span_8h.html | 2 +-
docs/reference/api/doxygen/ir_2span_8h__incl.svg | 972 +-
docs/reference/api/doxygen/ir_2transform_8h.html | 2 +-
.../api/doxygen/ir_2transform_8h__incl.svg | 1532 +--
.../api/doxygen/ir_2transform_8h_source.html | 2 +-
docs/reference/api/doxygen/ir_2type_8h.html | 2 +-
docs/reference/api/doxygen/ir_2type_8h__incl.svg | 1010 +-
docs/reference/api/doxygen/ir_2type_8h_source.html | 2 +-
.../api/doxygen/ir__builder_2ir_2frame_8h.html | 2 +-
.../doxygen/ir__builder_2ir_2frame_8h__incl.svg | 1194 ++-
.../doxygen/ir__builder_2ir_2frame_8h_source.html | 2 +-
.../api/doxygen/ir__builder_2tir_2frame_8h.html | 2 +-
.../doxygen/ir__builder_2tir_2frame_8h__incl.svg | 1292 +--
.../doxygen/ir__builder_2tir_2frame_8h_source.html | 2 +-
docs/reference/api/doxygen/ir__docsifier_8h.html | 2 +-
.../api/doxygen/ir__docsifier_8h__incl.svg | 1232 +--
.../api/doxygen/ir__docsifier_8h_source.html | 12 +-
.../api/doxygen/iter__affine__map_8h.html | 2 +-
.../api/doxygen/iter__affine__map_8h__incl.svg | 1312 +--
.../api/doxygen/iter__affine__map_8h_source.html | 2 +-
.../api/doxygen/libtorch__runtime_8h.html | 2 +-
.../api/doxygen/libtorch__runtime_8h__incl.svg | 728 +-
.../api/doxygen/local__response__norm_8h.html | 2 +-
.../api/doxygen/local__response__norm_8h__incl.svg | 1696 ++-
.../doxygen/local__response__norm_8h_source.html | 4 +-
docs/reference/api/doxygen/loop__state_8h.html | 2 +-
.../reference/api/doxygen/loop__state_8h__incl.svg | 1266 +--
.../api/doxygen/loop__state_8h_source.html | 2 +-
docs/reference/api/doxygen/mapping_8h.html | 2 +-
docs/reference/api/doxygen/mapping_8h__incl.svg | 1696 ++-
docs/reference/api/doxygen/measure_8h_source.html | 2 +-
.../api/doxygen/measure__callback_8h.html | 2 +-
.../api/doxygen/measure__callback_8h__incl.svg | 1586 +--
.../api/doxygen/measure__callback_8h_source.html | 2 +-
.../api/doxygen/measure__candidate_8h.html | 2 +-
.../api/doxygen/measure__candidate_8h__incl.svg | 1520 +--
.../api/doxygen/measure__candidate_8h_source.html | 2 +-
.../api/doxygen/measure__record_8h_source.html | 2 +-
docs/reference/api/doxygen/memory__pools_8h.html | 2 +-
.../api/doxygen/memory__pools_8h__incl.svg | 1682 +--
.../api/doxygen/memory__pools_8h_source.html | 2 +-
.../doxygen/meta__schedule_2cost__model_8h.html | 2 +-
.../meta__schedule_2cost__model_8h__incl.svg | 1564 +--
.../meta__schedule_2cost__model_8h_source.html | 2 +-
docs/reference/api/doxygen/metadata_8h.html | 2 +-
docs/reference/api/doxygen/metadata_8h__incl.svg | 1070 +-
docs/reference/api/doxygen/metadata__base_8h.html | 2 +-
.../api/doxygen/metadata__base_8h__incl.svg | 1100 +-
.../api/doxygen/metadata__base_8h_source.html | 2 +-
docs/reference/api/doxygen/mutator_8h.html | 2 +-
docs/reference/api/doxygen/mutator_8h__incl.svg | 1074 +-
docs/reference/api/doxygen/name__supply_8h.html | 2 +-
.../api/doxygen/name__supply_8h__incl.svg | 1086 +-
.../api/doxygen/namespacemembers_func_s.html | 6 +-
docs/reference/api/doxygen/namespacemembers_i.html | 3 +
docs/reference/api/doxygen/namespacemembers_s.html | 6 +-
.../api/doxygen/namespacemembers_vars.html | 3 +
.../api/doxygen/namespacetvm_1_1runtime.html | 32 +
docs/reference/api/doxygen/nn_2bnn_8h.html | 2 +-
docs/reference/api/doxygen/nn_2bnn_8h__incl.svg | 1450 +--
docs/reference/api/doxygen/nn_2bnn_8h_source.html | 4 +-
docs/reference/api/doxygen/nn_2dense_8h.html | 2 +-
docs/reference/api/doxygen/nn_2dense_8h__incl.svg | 1696 ++-
docs/reference/api/doxygen/nn_2pooling_8h.html | 2 +-
.../reference/api/doxygen/nn_2pooling_8h__incl.svg | 1486 +--
.../api/doxygen/nn_2pooling_8h_source.html | 12 +-
docs/reference/api/doxygen/nn_2softmax_8h.html | 2 +-
.../reference/api/doxygen/nn_2softmax_8h__incl.svg | 1418 +--
.../api/doxygen/nn_2softmax_8h_source.html | 4 +-
docs/reference/api/doxygen/node_8h.html | 2 +-
docs/reference/api/doxygen/node_8h__incl.svg | 952 +-
docs/reference/api/doxygen/node_8h_source.html | 2 +-
.../reference/api/doxygen/object_8h__dep__incl.svg | 44 +-
docs/reference/api/doxygen/on__device_8h__incl.svg | 1298 +--
docs/reference/api/doxygen/op__strategy_8h.html | 2 +-
.../api/doxygen/op__strategy_8h__incl.svg | 1664 +--
.../api/doxygen/op__strategy_8h_source.html | 2 +-
docs/reference/api/doxygen/operation_8h.html | 2 +-
docs/reference/api/doxygen/operation_8h__incl.svg | 1380 +--
.../reference/api/doxygen/operation_8h_source.html | 2 +-
docs/reference/api/doxygen/optional_8h.html | 2 +-
.../api/doxygen/optional_8h__dep__incl.svg | 1633 +--
docs/reference/api/doxygen/packed__func_8h.html | 2 +-
.../api/doxygen/packed__func_8h__incl.svg | 942 +-
.../api/doxygen/packed__func_8h_source.html | 4 +-
docs/reference/api/doxygen/pad__utils_8h.html | 2 +-
docs/reference/api/doxygen/pad__utils_8h__incl.svg | 1410 +--
docs/reference/api/doxygen/papi_8h.html | 2 +-
docs/reference/api/doxygen/papi_8h__incl.svg | 1180 +-
docs/reference/api/doxygen/papi_8h_source.html | 2 +-
docs/reference/api/doxygen/parser_8h.html | 2 +-
docs/reference/api/doxygen/parser_8h__incl.svg | 1306 +--
docs/reference/api/doxygen/pattern_8h.html | 2 +-
docs/reference/api/doxygen/pattern_8h__incl.svg | 1204 +--
.../api/doxygen/pattern__functor_8h__incl.svg | 1412 +--
docs/reference/api/doxygen/postproc_8h.html | 2 +-
docs/reference/api/doxygen/postproc_8h__incl.svg | 1104 +-
docs/reference/api/doxygen/printer_2frame_8h.html | 2 +-
.../api/doxygen/printer_2frame_8h__incl.svg | 1146 +-
.../api/doxygen/printer_2frame_8h_source.html | 2 +-
docs/reference/api/doxygen/printer_8h.html | 2 +-
docs/reference/api/doxygen/printer_8h__incl.svg | 968 +-
docs/reference/api/doxygen/profiler_8h.html | 2 +-
docs/reference/api/doxygen/profiler_8h__incl.svg | 1518 +--
docs/reference/api/doxygen/profiling_8h.html | 2 +-
docs/reference/api/doxygen/profiling_8h__incl.svg | 720 +-
.../reference/api/doxygen/profiling_8h_source.html | 2 +-
docs/reference/api/doxygen/random_8h.html | 2 +-
docs/reference/api/doxygen/random_8h__incl.svg | 1126 +-
docs/reference/api/doxygen/random_8h_source.html | 2 +-
docs/reference/api/doxygen/ravel__unravel_8h.html | 2 +-
.../api/doxygen/ravel__unravel_8h__incl.svg | 1432 ++-
docs/reference/api/doxygen/reduce_8h.html | 2 +-
docs/reference/api/doxygen/reduce_8h__incl.svg | 1136 +-
docs/reference/api/doxygen/reduce_8h_source.html | 2 +-
docs/reference/api/doxygen/reduction_8h.html | 2 +-
docs/reference/api/doxygen/reduction_8h__incl.svg | 1442 +--
.../reference/api/doxygen/reduction_8h_source.html | 8 +-
docs/reference/api/doxygen/reflection_8h.html | 2 +-
docs/reference/api/doxygen/reflection_8h__incl.svg | 848 +-
docs/reference/api/doxygen/registry_8h.html | 2 +-
docs/reference/api/doxygen/registry_8h__incl.svg | 984 +-
docs/reference/api/doxygen/relay_2adt_8h.html | 2 +-
docs/reference/api/doxygen/relay_2adt_8h__incl.svg | 1845 ++--
.../api/doxygen/relay_2adt_8h_source.html | 2 +-
docs/reference/api/doxygen/relay_2analysis_8h.html | 2 +-
.../api/doxygen/relay_2analysis_8h__incl.svg | 1490 +--
.../api/doxygen/relay_2analysis_8h_source.html | 2 +-
.../api/doxygen/relay_2attrs_2debug_8h.html | 2 +-
.../api/doxygen/relay_2attrs_2debug_8h__incl.svg | 1154 +-
.../api/doxygen/relay_2attrs_2memory_8h__incl.svg | 1518 +--
.../doxygen/relay_2attrs_2memory_8h_source.html | 2 +-
.../reference/api/doxygen/relay_2attrs_2nn_8h.html | 2 +-
.../api/doxygen/relay_2attrs_2nn_8h__incl.svg | 1294 +--
.../api/doxygen/relay_2attrs_2nn_8h_source.html | 2 +-
.../api/doxygen/relay_2attrs_2transform_8h.html | 2 +-
.../doxygen/relay_2attrs_2transform_8h__incl.svg | 1524 +--
.../doxygen/relay_2attrs_2transform_8h_source.html | 2 +-
.../reference/api/doxygen/relay_2attrs_2vm_8h.html | 2 +-
.../api/doxygen/relay_2attrs_2vm_8h__incl.svg | 1124 +-
.../api/doxygen/relay_2attrs_2vm_8h_source.html | 2 +-
docs/reference/api/doxygen/relay_2base_8h.html | 2 +-
.../reference/api/doxygen/relay_2base_8h__incl.svg | 1228 +--
docs/reference/api/doxygen/relay_2expr_8h.html | 2 +-
.../reference/api/doxygen/relay_2expr_8h__incl.svg | 1510 +--
.../api/doxygen/relay_2expr_8h_source.html | 4 +-
.../api/doxygen/relay_2expr__functor_8h.html | 2 +-
.../api/doxygen/relay_2expr__functor_8h__incl.svg | 1438 +--
docs/reference/api/doxygen/relay_2feature_8h.html | 2 +-
.../api/doxygen/relay_2feature_8h__incl.svg | 1488 +--
.../api/doxygen/relay_2feature_8h_source.html | 4 +-
docs/reference/api/doxygen/relay_2function_8h.html | 2 +-
.../api/doxygen/relay_2function_8h__incl.svg | 1536 +--
.../api/doxygen/relay_2function_8h_source.html | 2 +-
docs/reference/api/doxygen/relay_2op_8h.html | 2 +-
docs/reference/api/doxygen/relay_2op_8h__incl.svg | 1556 +--
.../api/doxygen/relay_2op__attr__types_8h.html | 2 +-
.../doxygen/relay_2op__attr__types_8h__incl.svg | 1634 +--
.../doxygen/relay_2op__attr__types_8h_source.html | 2 +-
.../api/doxygen/relay_2qnn_2attrs_8h.html | 2 +-
.../api/doxygen/relay_2qnn_2attrs_8h__incl.svg | 1136 +-
.../reference/api/doxygen/relay_2transform_8h.html | 2 +-
.../api/doxygen/relay_2transform_8h__incl.svg | 1581 +--
.../api/doxygen/relay_2transform_8h_source.html | 2 +-
docs/reference/api/doxygen/relay_2type_8h.html | 2 +-
.../reference/api/doxygen/relay_2type_8h__incl.svg | 1751 ++-
docs/reference/api/doxygen/reorg_8h.html | 2 +-
docs/reference/api/doxygen/reorg_8h__incl.svg | 1504 +--
docs/reference/api/doxygen/reorg_8h_source.html | 2 +-
docs/reference/api/doxygen/rocblas_8h.html | 2 +-
docs/reference/api/doxygen/rocblas_8h__incl.svg | 1453 +--
docs/reference/api/doxygen/rocblas_8h_source.html | 2 +-
docs/reference/api/doxygen/rocm_2dense_8h.html | 2 +-
.../reference/api/doxygen/rocm_2dense_8h__incl.svg | 1578 +--
.../api/doxygen/rocm_2dense_8h_source.html | 2 +-
docs/reference/api/doxygen/rocm_2injective_8h.html | 2 +-
.../api/doxygen/rocm_2injective_8h__incl.svg | 1722 +--
.../api/doxygen/rocm_2injective_8h_source.html | 2 +-
docs/reference/api/doxygen/rocm_2pooling_8h.html | 2 +-
.../api/doxygen/rocm_2pooling_8h__incl.svg | 1686 +--
.../api/doxygen/rocm_2pooling_8h_source.html | 2 +-
docs/reference/api/doxygen/rocm_2reduction_8h.html | 2 +-
.../api/doxygen/rocm_2reduction_8h__incl.svg | 1722 +--
.../api/doxygen/rocm_2reduction_8h_source.html | 2 +-
docs/reference/api/doxygen/rocm_2softmax_8h.html | 2 +-
.../api/doxygen/rocm_2softmax_8h__incl.svg | 1722 +--
.../api/doxygen/rocm_2softmax_8h_source.html | 2 +-
docs/reference/api/doxygen/runner_8h.html | 2 +-
docs/reference/api/doxygen/runner_8h__incl.svg | 1530 +--
docs/reference/api/doxygen/runner_8h_source.html | 2 +-
.../api/doxygen/runtime_2container_2base_8h.html | 2 +-
.../runtime_2container_2base_8h__dep__incl.svg | 1240 +--
.../runtime_2container_2base_8h_source.html | 2 +-
.../api/doxygen/runtime_2memory_8h_source.html | 2 +-
docs/reference/api/doxygen/runtime_2module_8h.html | 2 +-
.../api/doxygen/runtime_2module_8h__incl.svg | 702 +-
docs/reference/api/doxygen/runtime_2vm_2vm_8h.html | 2 +-
.../api/doxygen/runtime_2vm_2vm_8h__incl.svg | 992 +-
docs/reference/api/doxygen/runtime_8h.html | 2 +-
docs/reference/api/doxygen/runtime_8h__incl.svg | 1582 +--
docs/reference/api/doxygen/runtime_8h_source.html | 2 +-
docs/reference/api/doxygen/schedule__pass_8h.html | 2 +-
.../api/doxygen/schedule__pass_8h__incl.svg | 1428 +--
docs/reference/api/doxygen/schedule__rule_8h.html | 2 +-
.../api/doxygen/schedule__rule_8h__incl.svg | 1146 +-
.../api/doxygen/schedule__rule_8h_source.html | 2 +-
.../api/doxygen/script_2ir__builder_2base_8h.html | 2 +-
.../doxygen/script_2ir__builder_2base_8h__incl.svg | 1172 +-
.../script_2ir__builder_2base_8h_source.html | 2 +-
docs/reference/api/doxygen/search/all_11.js | 4 +-
docs/reference/api/doxygen/search/all_13.js | 4 +-
docs/reference/api/doxygen/search/all_14.js | 14 +-
docs/reference/api/doxygen/search/all_15.js | 12 +-
docs/reference/api/doxygen/search/all_16.js | 4 +-
docs/reference/api/doxygen/search/all_17.js | 4 +-
docs/reference/api/doxygen/search/all_4.js | 2 +-
docs/reference/api/doxygen/search/all_7.js | 2 +-
docs/reference/api/doxygen/search/all_a.js | 5 +-
docs/reference/api/doxygen/search/all_d.js | 6 +-
docs/reference/api/doxygen/search/all_e.js | 6 +-
docs/reference/api/doxygen/search/all_f.js | 2 +-
docs/reference/api/doxygen/search/classes_10.js | 2 +-
docs/reference/api/doxygen/search/classes_11.js | 4 +-
docs/reference/api/doxygen/search/classes_13.js | 4 +-
docs/reference/api/doxygen/search/classes_5.js | 2 +-
docs/reference/api/doxygen/search/classes_8.js | 4 +-
docs/reference/api/doxygen/search/classes_9.js | 4 +-
docs/reference/api/doxygen/search/classes_d.js | 2 +-
docs/reference/api/doxygen/search/functions_13.js | 10 +-
docs/reference/api/doxygen/search/functions_14.js | 2 +-
docs/reference/api/doxygen/search/functions_15.js | 2 +-
docs/reference/api/doxygen/search/functions_3.js | 2 +-
docs/reference/api/doxygen/search/functions_c.js | 2 +-
docs/reference/api/doxygen/search/functions_d.js | 6 +-
docs/reference/api/doxygen/search/functions_e.js | 2 +-
docs/reference/api/doxygen/search/typedefs_e.js | 2 +-
docs/reference/api/doxygen/search/variables_9.js | 1 +
docs/reference/api/doxygen/search__policy_8h.html | 2 +-
.../api/doxygen/search__policy_8h__incl.svg | 1284 +--
.../api/doxygen/search__policy_8h_source.html | 2 +-
.../reference/api/doxygen/search__strategy_8h.html | 2 +-
.../api/doxygen/search__strategy_8h__incl.svg | 1692 +--
.../api/doxygen/search__strategy_8h_source.html | 2 +-
.../api/doxygen/search__task_8h_source.html | 2 +-
docs/reference/api/doxygen/source__map_8h.html | 2 +-
.../reference/api/doxygen/source__map_8h__incl.svg | 1056 +-
.../reference/api/doxygen/space__generator_8h.html | 2 +-
.../api/doxygen/space__generator_8h__incl.svg | 1550 +--
.../api/doxygen/space__generator_8h_source.html | 2 +-
docs/reference/api/doxygen/state_8h.html | 2 +-
docs/reference/api/doxygen/state_8h__incl.svg | 1460 +--
docs/reference/api/doxygen/stmt_8h.html | 2 +-
docs/reference/api/doxygen/stmt_8h__incl.svg | 1224 +--
docs/reference/api/doxygen/stmt_8h_source.html | 4 +-
docs/reference/api/doxygen/stmt__functor_8h.html | 2 +-
.../api/doxygen/stmt__functor_8h__incl.svg | 1334 +--
.../api/doxygen/stmt__functor_8h_source.html | 2 +-
docs/reference/api/doxygen/strided__slice_8h.html | 2 +-
.../api/doxygen/strided__slice_8h__incl.svg | 1328 +--
docs/reference/api/doxygen/string_8h_source.html | 2 +-
...tvm_1_1ConstantMemoryPoolsNode__coll__graph.svg | 4 +-
...ucttvm_1_1ConstantPoolInfoNode__coll__graph.svg | 8 +-
.../structtvm_1_1PoolInfoNode__coll__graph.svg | 4 +-
...vm_1_1WorkspaceMemoryPoolsNode__coll__graph.svg | 4 +-
...cttvm_1_1WorkspacePoolInfoNode__coll__graph.svg | 4 +-
..._1relay_1_1AdaptivePool1DAttrs__coll__graph.svg | 4 +-
..._1relay_1_1AdaptivePool2DAttrs__coll__graph.svg | 4 +-
..._1relay_1_1AdaptivePool3DAttrs__coll__graph.svg | 4 +-
...vm_1_1relay_1_1AffineGridAttrs__coll__graph.svg | 4 +-
...m_1_1relay_1_1AllocTensorAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1ArgReduceAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1AvgPool1DAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1AvgPool2DAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1AvgPool3DAttrs__coll__graph.svg | 4 +-
...m_1_1relay_1_1BatchMatmulAttrs__coll__graph.svg | 4 +-
..._1relay_1_1BatchToSpaceNDAttrs__coll__graph.svg | 8 +-
..._1_1relay_1_1BinaryConv2DAttrs__coll__graph.svg | 4 +-
..._1relay_1_1ConstructorValueObj__coll__graph.svg | 4 +-
...ucttvm_1_1relay_1_1Conv1DAttrs__coll__graph.svg | 4 +-
...1relay_1_1Conv1DTransposeAttrs__coll__graph.svg | 4 +-
...ucttvm_1_1relay_1_1Conv2DAttrs__coll__graph.svg | 4 +-
...1relay_1_1Conv2DTransposeAttrs__coll__graph.svg | 4 +-
..._1relay_1_1Conv2DWinogradAttrs__coll__graph.svg | 4 +-
...ucttvm_1_1relay_1_1Conv3DAttrs__coll__graph.svg | 4 +-
...1relay_1_1Conv3DTransposeAttrs__coll__graph.svg | 4 +-
..._1relay_1_1Conv3DWinogradAttrs__coll__graph.svg | 4 +-
...m_1_1relay_1_1CorrelationAttrs__coll__graph.svg | 4 +-
...1_1relay_1_1CropAndResizeAttrs__coll__graph.svg | 4 +-
...relay_1_1DeformableConv2DAttrs__coll__graph.svg | 4 +-
...ructtvm_1_1relay_1_1DenseAttrs__coll__graph.svg | 4 +-
...ucttvm_1_1relay_1_1DilateAttrs__coll__graph.svg | 4 +-
...vm_1_1relay_1_1Dilation2DAttrs__coll__graph.svg | 4 +-
...m_1_1relay_1_1L2NormalizeAttrs__coll__graph.svg | 4 +-
...ucttvm_1_1relay_1_1MatmulAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1MaxPool1DAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1MaxPool2DAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1MaxPool3DAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1MirrorPadAttrs__coll__graph.svg | 4 +-
...1_1relay_1_1MultiBoxPriorAttrs__coll__graph.svg | 4 +-
...y_1_1MultiBoxTransformLocAttrs__coll__graph.svg | 4 +-
...ucttvm_1_1relay_1_1NormalAttrs__coll__graph.svg | 4 +-
...structtvm_1_1relay_1_1PadAttrs__coll__graph.svg | 4 +-
...ttvm_1_1relay_1_1ProposalAttrs__coll__graph.svg | 4 +-
...ttvm_1_1relay_1_1ROIAlignAttrs__coll__graph.svg | 4 +-
...cttvm_1_1relay_1_1ROIPoolAttrs__coll__graph.svg | 4 +-
...ucttvm_1_1relay_1_1ReduceAttrs__coll__graph.svg | 4 +-
...cttvm_1_1relay_1_1ReshapeAttrs__coll__graph.svg | 4 +-
...1_1relay_1_1ReshapeTensorAttrs__coll__graph.svg | 4 +-
...ttvm_1_1relay_1_1Resize1DAttrs__coll__graph.svg | 8 +-
...ttvm_1_1relay_1_1Resize2DAttrs__coll__graph.svg | 8 +-
...ttvm_1_1relay_1_1Resize3DAttrs__coll__graph.svg | 8 +-
...tvm_1_1relay_1_1ShapeFuncAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1SliceLikeAttrs__coll__graph.svg | 4 +-
...1_1relay_1_1SlidingWindowAttrs__coll__graph.svg | 4 +-
..._1relay_1_1SpaceToBatchNDAttrs__coll__graph.svg | 8 +-
..._1_1relay_1_1SparseConv2DAttrs__coll__graph.svg | 4 +-
...1_1relay_1_1SparseToDenseAttrs__coll__graph.svg | 4 +-
...cttvm_1_1relay_1_1SqueezeAttrs__coll__graph.svg | 4 +-
...relay_1_1ThreefryGenerateAttrs__coll__graph.svg | 4 +-
...tructtvm_1_1relay_1_1TileAttrs__coll__graph.svg | 4 +-
...tvm_1_1relay_1_1TransposeAttrs__coll__graph.svg | 4 +-
...cttvm_1_1relay_1_1UniformAttrs__coll__graph.svg | 4 +-
...ttvm_1_1relay_1_1VarianceAttrs__coll__graph.svg | 4 +-
...ructtvm_1_1runtime_1_1is__valid__iterator.html} | 49 +-
..._01Optional_3_01T_01_4_00_01IterType_01_4.html} | 42 +-
..._3_01T_01_4_00_01IterType_01_4__coll__graph.svg | 66 +
...01T_01_4_00_01IterType_01_4__inherit__graph.svg | 66 +
...runtime_1_1is__valid__iterator__coll__graph.svg | 45 +
...time_1_1is__valid__iterator__inherit__graph.svg | 66 +
...1tir_1_1usmp_1_1BufferInfoNode__coll__graph.svg | 8 +-
.../api/doxygen/structural__equal_8h.html | 2 +-
.../api/doxygen/structural__equal_8h__incl.svg | 408 +-
.../api/doxygen/structural__equal_8h_source.html | 4 +-
docs/reference/api/doxygen/tag_8h.html | 2 +-
docs/reference/api/doxygen/tag_8h__incl.svg | 1186 ++-
docs/reference/api/doxygen/target_8h.html | 2 +-
docs/reference/api/doxygen/target_8h__incl.svg | 1560 +--
docs/reference/api/doxygen/target_8h_source.html | 2 +-
docs/reference/api/doxygen/target__info_8h.html | 2 +-
.../api/doxygen/target__info_8h__incl.svg | 1084 +-
docs/reference/api/doxygen/target__kind_8h.html | 2 +-
.../api/doxygen/target__kind_8h__incl.svg | 1510 +--
.../api/doxygen/target__kind_8h_source.html | 2 +-
docs/reference/api/doxygen/task__scheduler_8h.html | 2 +-
.../api/doxygen/task__scheduler_8h__incl.svg | 1722 +--
.../api/doxygen/task__scheduler_8h_source.html | 2 +-
docs/reference/api/doxygen/te_2schedule_8h.html | 2 +-
.../api/doxygen/te_2schedule_8h__incl.svg | 1310 +--
.../api/doxygen/te_2schedule_8h_source.html | 2 +-
docs/reference/api/doxygen/tensor_8h.html | 2 +-
docs/reference/api/doxygen/tensor_8h__incl.svg | 1354 +--
docs/reference/api/doxygen/tensor_8h_source.html | 2 +-
docs/reference/api/doxygen/tensor__intrin_8h.html | 2 +-
.../api/doxygen/tensor__intrin_8h__incl.svg | 1342 +--
.../api/doxygen/tensor__intrin_8h_source.html | 2 +-
docs/reference/api/doxygen/tensor__type_8h.html | 2 +-
.../api/doxygen/tensor__type_8h__incl.svg | 1084 +-
.../api/doxygen/tensor__type_8h_source.html | 2 +-
docs/reference/api/doxygen/tensor__utils_8h.html | 2 +-
.../api/doxygen/tensor__utils_8h__incl.svg | 1432 ++-
docs/reference/api/doxygen/tir_2analysis_8h.html | 2 +-
.../api/doxygen/tir_2analysis_8h__incl.svg | 1514 +--
.../api/doxygen/tir_2analysis_8h_source.html | 2 +-
docs/reference/api/doxygen/tir_2expr_8h.html | 2 +-
docs/reference/api/doxygen/tir_2expr_8h__incl.svg | 1208 +--
.../reference/api/doxygen/tir_2expr_8h_source.html | 4 +-
.../api/doxygen/tir_2expr__functor_8h.html | 2 +-
.../api/doxygen/tir_2expr__functor_8h__incl.svg | 1208 +--
docs/reference/api/doxygen/tir_2function_8h.html | 2 +-
.../api/doxygen/tir_2function_8h__incl.svg | 1298 +--
.../api/doxygen/tir_2function_8h_source.html | 2 +-
docs/reference/api/doxygen/tir_2ir_8h.html | 2 +-
docs/reference/api/doxygen/tir_2ir_8h__incl.svg | 1308 +--
docs/reference/api/doxygen/tir_2ir_8h_source.html | 2 +-
docs/reference/api/doxygen/tir_2op_8h.html | 2 +-
docs/reference/api/doxygen/tir_2op_8h__incl.svg | 1382 +--
docs/reference/api/doxygen/tir_2op_8h_source.html | 2 +-
.../api/doxygen/tir_2op__attr__types_8h.html | 2 +-
.../api/doxygen/tir_2op__attr__types_8h__incl.svg | 1090 +-
.../api/doxygen/tir_2schedule_2schedule_8h.html | 2 +-
.../doxygen/tir_2schedule_2schedule_8h__incl.svg | 1420 +--
.../doxygen/tir_2schedule_2schedule_8h_source.html | 2 +-
docs/reference/api/doxygen/tir_2transform_8h.html | 2 +-
.../api/doxygen/tir_2transform_8h__incl.svg | 1556 +--
.../api/doxygen/tir_2transform_8h_source.html | 2 +-
.../api/doxygen/tir_2usmp_2analysis_8h.html | 2 +-
.../api/doxygen/tir_2usmp_2analysis_8h__incl.svg | 1480 +--
.../api/doxygen/tir_2usmp_2transform_8h.html | 2 +-
.../api/doxygen/tir_2usmp_2transform_8h__incl.svg | 1412 +--
.../reference/api/doxygen/tir_2usmp_2utils_8h.html | 2 +-
.../api/doxygen/tir_2usmp_2utils_8h__incl.svg | 1432 +--
.../api/doxygen/tir_2usmp_2utils_8h_source.html | 2 +-
docs/reference/api/doxygen/topi_2nn_8h.html | 2 +-
docs/reference/api/doxygen/topi_2nn_8h__incl.svg | 1436 +--
docs/reference/api/doxygen/topi_2nn_8h_source.html | 6 +-
docs/reference/api/doxygen/topi_2transform_8h.html | 2 +-
.../api/doxygen/topi_2transform_8h__incl.svg | 1344 +--
.../api/doxygen/topi_2transform_8h_source.html | 14 +-
docs/reference/api/doxygen/topi_2utils_8h.html | 2 +-
.../reference/api/doxygen/topi_2utils_8h__incl.svg | 1082 +-
.../api/doxygen/topi_2utils_8h_source.html | 4 +-
docs/reference/api/doxygen/trace_8h.html | 2 +-
docs/reference/api/doxygen/trace_8h__incl.svg | 888 +-
docs/reference/api/doxygen/trace_8h_source.html | 2 +-
docs/reference/api/doxygen/traced__object_8h.html | 2 +-
.../api/doxygen/traced__object_8h__incl.svg | 896 +-
.../api/doxygen/traced__object_8h_source.html | 2 +-
.../api/doxygen/traced__object__functor_8h.html | 2 +-
.../doxygen/traced__object__functor_8h__incl.svg | 1048 +-
docs/reference/api/doxygen/transform__step_8h.html | 2 +-
.../api/doxygen/transform__step_8h__incl.svg | 1276 +--
.../api/doxygen/transform__step_8h_source.html | 2 +-
docs/reference/api/doxygen/tune__context_8h.html | 2 +-
.../api/doxygen/tune__context_8h__incl.svg | 1866 ++--
.../api/doxygen/tune__context_8h_source.html | 2 +-
.../api/doxygen/type__functor_8h__incl.svg | 1440 +--
.../api/doxygen/type__functor_8h_source.html | 2 +-
docs/reference/api/doxygen/type__relation_8h.html | 2 +-
.../api/doxygen/type__relation_8h__incl.svg | 1356 +--
.../api/doxygen/type__relation_8h_source.html | 2 +-
docs/reference/api/doxygen/var_8h.html | 2 +-
docs/reference/api/doxygen/var_8h__incl.svg | 1094 +-
docs/reference/api/doxygen/var_8h_source.html | 2 +-
docs/reference/api/doxygen/var__table_8h.html | 2 +-
docs/reference/api/doxygen/var__table_8h__incl.svg | 1172 +-
docs/reference/api/doxygen/virtual__device_8h.html | 2 +-
.../api/doxygen/virtual__device_8h__incl.svg | 1459 ++-
docs/reference/api/doxygen/vision_8h.html | 2 +-
docs/reference/api/doxygen/vision_8h__incl.svg | 1296 +--
docs/reference/api/doxygen/vision_8h_source.html | 2 +-
docs/reference/api/doxygen/x86_2bnn_8h.html | 2 +-
docs/reference/api/doxygen/x86_2bnn_8h__incl.svg | 1726 +--
docs/reference/api/doxygen/x86_2bnn_8h_source.html | 4 +-
docs/reference/api/doxygen/x86_2default_8h.html | 2 +-
.../api/doxygen/x86_2default_8h__incl.svg | 1728 +--
.../api/doxygen/x86_2default_8h_source.html | 4 +-
docs/reference/api/doxygen/x86_2injective_8h.html | 2 +-
.../api/doxygen/x86_2injective_8h__incl.svg | 1726 +--
.../api/doxygen/x86_2injective_8h_source.html | 4 +-
docs/reference/api/python/auto_scheduler.html | 4 +-
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
.../api/typedoc/classes/graphexecutor.html | 16 +-
docs/reference/api/typedoc/classes/instance.html | 40 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 112 +-
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 4 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 6 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 2 +-
docs/tutorial/autotvm_matmul_x86.html | 20 +-
docs/tutorial/autotvm_relay_x86.html | 262 +-
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 | 46 +-
1060 files changed, 160224 insertions(+), 158605 deletions(-)
diff --git a/docs/_sources/how_to/compile_models/from_darknet.rst.txt b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
index 21323b76af..a8b1a8322e 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -315,7 +315,7 @@ The process is no different from other examples.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 3.807 seconds)
+ **Total running time of the script:** ( 1 minutes 5.934 seconds)
.. _sphx_glr_download_how_to_compile_models_from_darknet.py:
diff --git a/docs/_sources/how_to/compile_models/from_keras.rst.txt b/docs/_sources/how_to/compile_models/from_keras.rst.txt
index 9ec17997a7..8154be4ed8 100644
--- a/docs/_sources/how_to/compile_models/from_keras.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_keras.rst.txt
@@ -228,7 +228,7 @@ Look up prediction top 1 index in 1000 class synset.
.. code-block:: none
Relay top-1 id: 285, class name: Egyptian cat
-
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 968ms/step
+
1/1 [==============================] - ETA: 0s
1/1 [==============================] - 1s 912ms/step
Keras top-1 id: 285, class name: Egyptian cat
diff --git a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
index f01ef38128..0acab1357e 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -115,7 +115,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip6d80a7c1-8e70-4fbe-aba5-9e9661b2a96b from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip63353ed5-629d-481e-9485-4ed76f6fd434 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 c0e8a7f1c6..5c4e4f24be 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -116,7 +116,7 @@ Load a pretrained OneFlow model and save model
.. code-block:: none
Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
0%| | 0.00/41.5M [00:00<?, ?B/s]
15%|#5 | 6.33M/41.5M [00:00<00:00, 40.8MB/s]
30%|##9 | 12.4M/41.5M [00:00<00:00, 51.5MB/s]
42%|####2 | 17.6M/41.5M [00:00<00:00, 43.1MB/s]
54%|#####4 | 22.5M/41.5M [00:00<00:00, 45.9MB/s]
65%|######5 | 27.1M/41.5M [00:00<00:00, 42.2MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 44.4MB/s]
92%|#########2| 38.3M/41.5M [00:00<00:00, 48.1MB/s]
100%|##########| 41.5M/41.5M [00:00<00:00, 43.6MB/s]
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
15%|#5 | 6.33M/41.5M [00:00<00:01, 34.9MB/s]
23%|##3 | 9.66M/41.5M [00:00<00:01, 28.2MB/s]
35%|###4 | 14.3M/41.5M [00:00<00:00, 34.8MB/s]
43%|####3 | 17.8M/41.5M [00:00<00:00, 34.4MB/s]
54%|#####3 | 22.3M/41.5M [00:00<00:00, 37.0MB/s]
63%|######2 | 26.0M/41.5M [00:00<00:00, 33.1MB/s]
77%|#######7 | 32.0M/41.5M [00:00<00:00, 39.7MB/s]
96%|#########6| 40.0M/41.5M [00:01<00:00, 46.8MB/s]
100%|##########| 41.5M/41.5M [00:01<00:00, 40.6MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
index ac877573f5..ab6887b97f 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -94,7 +94,7 @@ Load a pretrained PyTorch model
.. code-block:: none
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
0%| | 0.00/44.7M [00:00<?, ?B/s]
39%|###8 | 17.4M/44.7M [00:00<00:00, 183MB/s]
94%|#########3| 42.0M/44.7M [00:00<00:00, 226MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 212MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
10%|# | 4.51M/44.7M [00:00<00:00, 47.2MB/s]
20%|## | 9.02M/44.7M [00:00<00:00, 46.7MB/s]
84%|########4 | 37.6M/44.7M [00:00<00:00, 162MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 144MB/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 6c37c44111..43a33bc5ac 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -416,7 +416,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 4.804 seconds)
+ **Total running time of the script:** ( 1 minutes 7.471 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 286d58b72b..7d33d28354 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
Computation times
=================
-**05:11.069** total execution time for **how_to_compile_models** files:
+**05:18.170** total execution time for **how_to_compile_models** files:
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:04.804 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:07.471 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:03.807 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``) | 01:05.934 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:39.520 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``) | 00:42.285 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:28.334 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``) | 00:28.329 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:26.750 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``) | 00:26.348 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:24.968 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:25.428 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``) | 00:24.924 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``) | 00:23.215 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:20.145 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``) | 00:20.231 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:15.376 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``) | 00:16.452 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.440 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``) | 00:02.478 | 0.0 MB |
+-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index 152418be00..e31b248ace 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -434,7 +434,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.9402 15.9377 16.1041 15.8109 0.0895
+ 15.8707 15.8574 16.0833 15.7741 0.0823
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 8e93e32e2b..4d767912d9 100644
--- a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
@@ -123,7 +123,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
.. code-block:: none
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
6%|5 | 10.1M/170M [00:00<00:01, 97.2MB/s]
17%|#6 | 28.9M/170M [00:00<00:00, 154MB/s]
30%|### | 51.5M/170M [00:00<00:00, 191MB/s]
42%|####2 | 72.1M/170M [00:00<00:00, 201MB/s]
57%|#####6 | 96.4M/170M [00:00<00:00, 220MB/s]
69%|######9 | 118M/170M [00:00<00:00, 218MB/s]
84%|########3 | 142M/170M [00:00<00:00, 232MB/s]
98%|#########8| 167M/170M [00:00<00:00, 241MB/s]
100%|##########| 170M/170M [00:00<00:00, 217MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
11%|#1 | 19.2M/170M [00:00<00:00, 201MB/s]
26%|##5 | 43.9M/170M [00:00<00:00, 235MB/s]
41%|####1 | 70.3M/170M [00:00<00:00, 254MB/s]
57%|#####7 | 97.0M/170M [00:00<00:00, 264MB/s]
73%|#######2 | 124M/170M [00:00<00:00, 270MB/s]
89%|########8 | 151M/170M [00:00<00:00, 274MB/s]
100%|##########| 170M/170M [00:00<00:00, 261MB/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').
@@ -288,7 +288,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 1.844 seconds)
+ **Total running time of the script:** ( 3 minutes 2.671 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 6d215c2df1..80cec3eacf 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -232,7 +232,7 @@ training. Other models require a full post training calibration.
.. code-block:: none
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
0%| | 0.00/13.6M [00:00<?, ?B/s]
20%|## | 2.77M/13.6M [00:00<00:00, 28.9MB/s]
41%|#### | 5.52M/13.6M [00:00<00:00, 26.7MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 55.2MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
14%|#4 | 1.94M/13.6M [00:00<00:00, 20.2MB/s]
52%|#####2 | 7.06M/13.6M [00:00<00:00, 39.8MB/s]
80%|######## | 10.9M/13.6M [00:00<00:00, 39.1MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 39.3MB/s]
@@ -405,7 +405,7 @@ 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)
- 90.4286 90.2643 97.4376 90.1577 0.8035
+ 90.4631 90.3462 95.7868 90.1210 0.5973
@@ -454,7 +454,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 9.629 seconds)
+ **Total running time of the script:** ( 1 minutes 10.165 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 b25ccb6a85..4dcce1842c 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -432,7 +432,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 120.3406 120.2129 125.6875 119.4403 0.6940
+ 119.5574 119.5427 121.3536 118.7764 0.3845
@@ -469,7 +469,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 59.808 seconds)
+ **Total running time of the script:** ( 1 minutes 52.478 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 ae410f61f9..127127b562 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -253,7 +253,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 30.152 seconds)
+ **Total running time of the script:** ( 1 minutes 29.408 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 f31d16b90b..a5bb9c04dd 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
@@ -158,7 +158,7 @@ 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]
5%|5 | 6708/132723 [00:00<00:01, 67072.39KB/s]
11%|#1 | 15053/132723 [00:00<00:01, 76699.09KB/s]
18%|#7 | 23491/132723 [00:00<00:01, 80201.07KB/s]
24%|##4 | 31945/132723 [00:00<00:01, 81910.38KB/s]
30%|### | 40364/132723 [00:00<00:01, 82729.26KB/s]
37%|###6 | 48745/132723 [00:00<00:01, 83092.60KB/s]
43%|####3 | 57107/132723 [00:00<00:00, 83262.95KB/s]
49%|####9 | 65534/132723 [00:00<00:00, 83580.51KB/s]
56%|#####5 | 73974/132723 [00:00<00:00, 83834.82KB/s]
62%|######2 | 82358/132723 [00:01<00:00, 83803.26KB/s]
68%|######8 | 90739/132723 [00:01<00:00, 83778.40KB/s]
75%|#######4 | 99117/132723 [00:01<00:00, 81173.69KB/s]
81%|########1 | 107561/132723 [00:01<00:00, 82138.51KB/s]
87%|########7 | 115790/132723 [00:01<00:00, 49694.17KB/s]
94%|#########3| 124208/132723 [00:01<00:00, 56740.54KB/s]
100%|########
#9| 132650/132723 [00:01<00:00, 62994.34KB/s]
100%|##########| 132723/132723 [00:01<00:00, 72364.38KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
4%|4 | 5933/132723 [00:00<00:02, 59311.11KB/s]
10%|# | 13545/132723 [00:00<00:01, 69195.10KB/s]
15%|#5 | 20465/132723 [00:00<00:02, 48704.08KB/s]
21%|## | 27666/132723 [00:00<00:01, 56036.11KB/s]
27%|##6 | 35413/132723 [00:00<00:01, 62680.01KB/s]
32%|###2 | 43039/132723 [00:00<00:01, 66846.97KB/s]
38%|###8 | 50736/132723 [00:00<00:01, 69931.43KB/s]
44%|####3 | 58189/132723 [00:00<00:01, 71323.03KB/s]
50%|####9 | 65867/132723 [00:00<00:00, 72970.52KB/s]
55%|#####5 | 73539/132723 [00:01<00:00, 74099.47KB/s]
61%|######1 | 81224/132723 [00:01<00:00, 74925.78KB/s]
67%|######7 | 88925/132723 [00:01<00:00, 75549.19KB/s]
73%|#######2 | 96548/132723 [00:01<00:00, 75749.53KB/s]
79%|#######8 | 104243/132723 [00:01<00:00, 76105.83KB/s]
84%|########4 | 111971/132723 [00:01<00:00, 76444.11KB/s]
90%|#########
| 119631/132723 [00:01<00:00, 76244.37KB/s]
96%|#########5| 127398/132723 [00:01<00:00, 76668.13KB/s]
100%|##########| 132723/132723 [00:01<00:00, 67147.42KB/s]
@@ -234,7 +234,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 38.460 seconds)
+ **Total running time of the script:** ( 2 minutes 38.295 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 d615668c79..573fb38846 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,24 +5,24 @@
Computation times
=================
-**11:35.860** total execution time for **how_to_deploy_models** files:
+**11:27.774** total execution time for **how_to_deploy_models** files:
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:01.844 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:02.671 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:38.460 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``) | 02:38.295 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 01:59.808 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``) | 01:52.478 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:30.152 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``) | 01:29.408 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:09.629 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``) | 01:10.165 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:30.042 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``) | 00:30.331 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:23.218 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``) | 00:22.361 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:22.701 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``) | 00:22.059 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.006 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``) | 00:00.007 | 0.0 MB |
+------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
index 9e357da1ba..bfefb600c7 100644
--- a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
@@ -472,7 +472,7 @@ First let us define two helper functions to get the mobilenet model and a cat im
.. code-block:: none
- Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip84dd76bd-55c6-4606-b57a-a34b08bad007 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip7ad6995e-d264-43a5-a3a2-2bac582310c9 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 bdaddee96f..2893e9861c 100644
--- a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
Computation times
=================
-**00:41.508** total execution time for **how_to_extend_tvm** files:
+**00:41.926** total execution time for **how_to_extend_tvm** files:
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:38.367 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:38.707 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.203 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``) | 00:02.246 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:00.930 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``) | 00:00.965 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``) | 00:00.008 | 0.0 MB |
+-------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index 77a82b6126..b063ae0dc8 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -216,10 +216,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 7116us [7116us] (46.84%; 46.84%)
- FoldScaleAxis: 8077us [16us] (53.16%; 53.16%)
- FoldConstant: 8061us [1680us] (53.06%; 99.80%)
- InferType: 6381us [6381us] (42.00%; 79.16%)
+ InferType: 7069us [7069us] (46.89%; 46.89%)
+ FoldScaleAxis: 8006us [6us] (53.11%; 53.11%)
+ FoldConstant: 8001us [1614us] (53.07%; 99.93%)
+ InferType: 6386us [6386us] (42.36%; 79.82%)
@@ -258,10 +258,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 6447us [6447us] (44.86%; 44.86%)
- FoldScaleAxis: 7924us [6us] (55.14%; 55.14%)
- FoldConstant: 7919us [1661us] (55.10%; 99.93%)
- InferType: 6257us [6257us] (43.54%; 79.02%)
+ InferType: 6425us [6425us] (44.86%; 44.86%)
+ FoldScaleAxis: 7897us [5us] (55.14%; 55.14%)
+ FoldConstant: 7892us [1624us] (55.10%; 99.94%)
+ InferType: 6268us [6268us] (43.77%; 79.43%)
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 ff741f4634..5451f5d08d 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -340,7 +340,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 34.263575 ms
+ Convolution: 51.348464 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 51c82deb95..57a0464f0e 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
@@ -671,7 +671,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 10.635003 ms
+ conv2d with tensor core: 6.678476 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 5dc351adb5..97f3720ad9 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -143,8 +143,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.019158
- Baseline: 3.335783
+ Numpy running time: 0.018788
+ Baseline: 3.324655
@@ -239,7 +239,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.324760
+ Opt1: 0.309349
@@ -342,7 +342,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.349662
+ Opt2: 0.341092
@@ -438,7 +438,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.116871
+ Opt3: 0.116290
@@ -563,7 +563,7 @@ flattening.
.. code-block:: none
- Opt4: 0.109945
+ Opt4: 0.109392
@@ -685,7 +685,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.110878
+ Opt5: 0.110807
@@ -810,7 +810,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
.. code-block:: none
- Opt6: 0.146649
+ Opt6: 0.147347
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 7a5f5a70fb..48edc3912a 100644
--- a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
Computation times
=================
-**00:34.792** total execution time for **how_to_optimize_operators** files:
+**00:34.389** total execution time for **how_to_optimize_operators** files:
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.497 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``) | 00:32.203 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.285 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.183 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.010 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``) | 00:01.004 | 0.0 MB |
+-----------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
index ac7df7aa9a..b0c26b934f 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,18 +5,18 @@
Computation times
=================
-**06:22.068** total execution time for **how_to_tune_with_autoscheduler** files:
+**06:27.366** total execution time for **how_to_tune_with_autoscheduler** files:
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 03:23.108 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 03:31.859 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:23.040 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``) | 01:23.178 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 00:56.668 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``) | 00:56.682 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:21.566 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``) | 00:17.940 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:08.909 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``) | 00:08.982 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:08.778 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``) | 00:08.725 | 0.0 MB |
+----------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
index f282320c4f..6e5810e151 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -240,12 +240,12 @@ cooperative fetching, unrolling and operator fusion.
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 28;
- allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [8]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [4032]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [6144]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [8], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[2] = 0f32
conv2d_nchw_1[3] = 0f32
@@ -253,470 +253,226 @@ cooperative fetching, unrolling and operator fusion.
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 (rc.outer.outer: int32, 0, 8) {
for (ry.outer.outer: int32, 0, 3) {
- let cse_var_2: int32 = (rc.outer.outer*72)
+ let cse_var_4: int32 = (rc.outer.outer*3136)
+ let cse_var_3: int32 = (ry.outer.outer*7)
+ let cse_var_2: int32 = (rc.outer.outer*576)
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_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [4032], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) && ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 196), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 392), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 588), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 784), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 980)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 980), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1176)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1176), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1372)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1372), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1568)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1568), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1764)] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) && ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3) + floormod(threadIdx.x_1, 9)) + 1364)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1960)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1960), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2156)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2156), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2352)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2352), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2548)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2548), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2744)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2744), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2940)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2940), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3136)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3136), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3332)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3332), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3528)] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) && ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3) + floormod(threadIdx.x_1, 9)) + 2736)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3724)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3724), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ if @tir.likely((threadIdx.x_1 < 112), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 3920)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3920), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1: Buffer(kernel.shared, float32, [6144], [], scope="shared")[threadIdx.x_2] = kernel[((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 192)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 192), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 196), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 4), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 392), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 588)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 588), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 4), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 784), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 980)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 980), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 20), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1176), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1372)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1372), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 28), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1568), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 32), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1764)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1764), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 12), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1960)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1960), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 40), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2156)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2156), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 44), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2352)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2352), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2548)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2548), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 52), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2744)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2744), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 56), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2940)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2940), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 20), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3136), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 64), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3332)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3332), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 68), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3528)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3528), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 24), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3724)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3724), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 76), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3920)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3920), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 80), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4116)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4116), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 28), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4312)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4312), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 88), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4508)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4508), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 92), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4704)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4704), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4900)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4900), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 100), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5096)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5096), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 104), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5292)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5292), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 36), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5488)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5488), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 112), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5684)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5684), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 116), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5880)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5880), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ if @tir.likely((threadIdx.x_2 < 68), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 6076)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 6076), 192)*4608)) + cse_var_2) + (floordiv((threadIdx.x_2 + 124), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ }
+ for (rc.outer.inner: int32, 0, 16) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12))]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 3)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 6)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 9)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 192)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 195)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 198)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 201)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 384)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 387)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 390)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 393)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 576)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 579)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 582)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 585)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 768)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 771)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 774)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 777)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 960)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 963)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 966)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 969)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1152)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1155)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1158)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1161)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1344)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1347)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1350)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1353)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 4)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 7)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 10)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 193)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 196)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 199)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 202)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 385)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 388)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 391)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 394)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 577)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 580)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 583)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 586)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 769)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 772)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 775)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 778)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 961)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 964)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 967)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 970)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1153)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1156)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1159)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1162)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1345)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1348)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1351)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1354)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 2)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 5)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 8)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 11)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 194)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 197)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 200)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 203)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 386)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 389)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 392)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 395)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 578)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 581)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 584)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 587)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 770)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 773)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 776)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 779)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 962)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 965)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 968)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 971)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1154)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1157)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1160)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1163)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1346)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1349)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1352)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1355)]))
}
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 64), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 128), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 256), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 320), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 448), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 512), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 640), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 704), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 147456)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 832), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 896), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 184320)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1024), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1088), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 221184)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1216), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1280), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1408), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1472), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 294912)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1600), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1664), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 331776)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1792), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1856), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 368640)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1984), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2048), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 405504)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2176), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2240), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 442368)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2368), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2432), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 479232)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2560), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2624), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 516096)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2752), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2816), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 552960)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2944), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 3008), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 47)]))
}
}
}
- for (i1.inner: int32, 0, 2) {
- for (i3.inner: int32, 0, 7) {
- compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*98)) + (i1.inner*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[((i1.inner*7) + i3.inner)] + bias[(((floordiv(blockIdx.x, 7)*128) + (threadIdx.x*2)) + i1.inner)]), 0f32)
- }
+ for (i1.inner: int32, 0, 8) {
+ compute[((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*392)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*8)) + i1.inner)]), 0f32)
}
}
}
@@ -771,7 +527,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.361 ms
+ Execution time of this operator: 0.270 ms
@@ -820,19 +576,19 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
- conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=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_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=8)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=4)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
- conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
+ conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
- conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
- conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
+ conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+ conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
- conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
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)
@@ -841,14 +597,14 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=8)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
- compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+ compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
- compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
- compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
+ compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+ compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -868,12 +624,12 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=4)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
@@ -893,10 +649,10 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+ extern "C" __global__ void __launch_bounds__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[8];
+ __shared__ float pad_temp_shared[4032];
+ __shared__ float kernel_shared[6144];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -905,418 +661,169 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
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 rc_outer_outer = 0; rc_outer_outer < 8; ++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);
+ pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 196) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 392) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 588)] = (((((1 <= ((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 588) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((1 <= ((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 784) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 980)] = (((((1 <= ((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 980) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((1 <= ((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1176) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1372)] = (((((1 <= ((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1372) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1568)] = (((((1 <= ((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1568) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1764)] = (((((1 <= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) + 1364)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1960)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1960) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2156)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2156) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2352)] = (((((1 <= ((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2352) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2548)] = (((((1 <= ((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2548) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2744)] = (((((1 <= ((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2744) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2940)] = (((((1 <= ((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2940) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3136)] = (((((1 <= ((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3136) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3332)] = (((((1 <= ((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3332) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3528)] = (((((1 <= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) + 2736)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3724)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3724) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 112) {
+ pad_temp_shared[(((int)threadIdx.x) + 3920)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3920) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 5) % 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);
+ kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 192) * 4608)) + (rc_outer_outer * 576)) + (((((int)threadIdx.x) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 196) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 4) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 8) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 588) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 4) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 16) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 980) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 20) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 8) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1372) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 28) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 32) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1764) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 12) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 40) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2156) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 44) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 16) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2548) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 52) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 56) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2940) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 20) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3136) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 64) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3332)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3332) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 68) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3528) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 24) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3724)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3724) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 76) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3920) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 80) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4116)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4116) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 28) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4312) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 88) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4508)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4508) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 92) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4704)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4704) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 32) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4900)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4900) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 100) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5096)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5096) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 104) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5292)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5292) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 36) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5488)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5488) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 112) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5684)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5684) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 116) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5880)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5880) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 40) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ if (((int)threadIdx.x) < 68) {
+ kernel_shared[(((int)threadIdx.x) + 6076)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6076) / 192) * 4608)) + (rc_outer_outer * 576)) + (((((int)threadIdx.x) + 124) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
}
- 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)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12))]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 3)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 6)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 9)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 192)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 195)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 198)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 201)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 384)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 387)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 390)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 393)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 576)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 579)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 582)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 585)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 768)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 771)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 774)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 777)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 960)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 963)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 966)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 969)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1152)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1155)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1158)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1161)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1344)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1347)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1350)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1353)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 4)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 7)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 10)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 193)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 196)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 199)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 202)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 385)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 388)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 391)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 394)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 577)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 580)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 583)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 586)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 769)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 772)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 775)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 778)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 961)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 964)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 967)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 970)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1153)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1156)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1159)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1162)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1345)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1348)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1351)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1354)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 2)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 5)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 8)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 11)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 194)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 197)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 200)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 203)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 386)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 389)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 392)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 395)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 578)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 581)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 584)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 587)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 770)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 773)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 776)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 779)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 962)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 965)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 968)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 971)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1154)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1157)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1160)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1163)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1346)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1349)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1352)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1355)]));
+ }
}
}
- for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
- }
+ for (int i1_inner = 0; i1_inner < 8; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 392)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 8)) + i1_inner)]), 0.000000e+00f);
}
}
@@ -1378,7 +885,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 23.108 seconds)
+ **Total running time of the script:** ( 3 minutes 31.859 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 b3373fe393..936bbaddfc 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -643,7 +643,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1781 8.1771 8.1828 8.1743 0.0035
+ 8.1711 8.1698 8.1762 8.1674 0.0037
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 701d8b1d99..bf77597113 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -662,7 +662,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 754.2436 754.3243 754.9236 753.4831 0.5908
+ 757.3703 757.8641 757.8779 756.3690 0.7081
@@ -690,7 +690,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 23.040 seconds)
+ **Total running time of the script:** ( 1 minutes 23.178 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 855596a954..8b46f57b41 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
@@ -397,78 +397,32 @@ 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_8: placeholder_15: Buffer(placeholder_13, int32, [33], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_18: Buffer(placeholder_10, float32, [128, 256], []), placeholder_7: placeholder_19: Buffer(placeholder_12, int32, [4916], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 64) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global {
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), placeholder_9: placeholder_18: Buffer(placeholder_14, float32, [128, 512], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_5: placeholder_19: Buffer(placeholder_10, float32, [128, 256], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [512]), storage_scope = global {
for (i.outer.inner: int32, 0, 2) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 16) {
- let cse_var_1: int32 = (((i.outer.inner*512) + (i.inner.init*32)) + (nb_j.inner*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 (i.inner.init: int32, 0, 16) {
+ for (j.init: int32, 0, 16) {
+ compute_5: Buffer(compute_4, float32, [512], [])[(((i.outer.inner*256) + (i.inner.init*16)) + j.init)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_2: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_2 + 1)] - placeholder_3[cse_var_2])) {
- for (i.inner: int32, 0, 16) {
- let cse_var_21: int32 = (elem_idx*16)
- let cse_var_20: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- let cse_var_19: int32 = (((i.outer.inner*512) + (i.inner*32)) + (nb_j.inner*16))
- let cse_var_18: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i.outer.inner*4096)) + (i.inner*256))
- let cse_var_17: int32 = (cse_var_19 + 9)
- let cse_var_16: int32 = (cse_var_19 + 8)
- let cse_var_15: int32 = (cse_var_19 + 7)
- let cse_var_14: int32 = (cse_var_19 + 6)
- let cse_var_13: int32 = (cse_var_19 + 5)
- let cse_var_12: int32 = (cse_var_19 + 4)
- let cse_var_11: int32 = (cse_var_19 + 3)
- let cse_var_10: int32 = (cse_var_19 + 2)
- let cse_var_9: int32 = (cse_var_19 + 15)
- let cse_var_8: int32 = (cse_var_19 + 14)
- let cse_var_7: int32 = (cse_var_19 + 13)
- let cse_var_6: int32 = (cse_var_19 + 12)
- let cse_var_5: int32 = (cse_var_19 + 11)
- let cse_var_4: int32 = (cse_var_19 + 10)
- let cse_var_3: int32 = (cse_var_19 + 1)
- {
- compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[((placeholder_3[cse_var_20]*16) + cse_var_21)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 1)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 2)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 3)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 4)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 5)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 6)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 7)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 8)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 9)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 10)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 11)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 12)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 13)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 14)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 15)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- }
+ }
+ for (elem_idx: int32, 0, let cse_var_1: int32 = floordiv(floormod(i0.outer.i1.outer.fused, 64), 2) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+ for (i.inner: int32, 0, 16) {
+ for (j: int32, 0, 16) {
+ let cse_var_3: int32 = floordiv(floormod(i0.outer.i1.outer.fused, 64), 2)
+ let cse_var_2: int32 = (((i.outer.inner*256) + (i.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[((((floordiv(i0.outer.i1.outer.fused, 64)*8192) + (i.outer.inner*4096)) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
}
for (i0.inner: int32, 0, 32) {
- let cse_var_22: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
- compute[ramp(cse_var_22, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_22, 1, 32)]), broadcast(0f32, 32))
+ for (i1.inner: int32, 0, 8) {
+ let cse_var_5: int32 = floormod(i0.outer.i1.outer.fused, 64)
+ let cse_var_6: int32 = (cse_var_5*8)
+ let cse_var_4: int32 = ((((floordiv(i0.outer.i1.outer.fused, 64)*16384) + (i0.inner*512)) + cse_var_6) + i1.inner)
+ compute[cse_var_4] = max((compute_5[((((i0.inner*16) + cse_var_6) + i1.inner) - (floordiv(cse_var_5, 2)*16))] + placeholder_4[cse_var_4]), 0f32)
+ }
}
}
}
@@ -524,7 +478,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 1.730 ms
+ Execution time of this operator: 3.168 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 64ff6c322e..d07cdd3a46 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
Computation times
=================
-**00:46.571** total execution time for **how_to_tune_with_autotvm** files:
+**00:45.184** total execution time for **how_to_tune_with_autotvm** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:46.535 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``) | 00:45.146 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.021 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``) | 00:00.022 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``) | 00:00.005 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index fda862d477..bc64954efd 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
@@ -1156,8 +1156,8 @@ for this template
TimeoutError
[('tile_f', [-1, 2, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4909501
- No: 9 GFLOPS: 221.34/221.34 result: MeasureResult(costs=(0.0010458929724137932,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9837684631347656, timestamp=1663707309.94193) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
- No: 10 GFLOPS: 0.00/221.34 result: Traceback (most recent call last):
+ No: 9 GFLOPS: 80.80/80.80 result: MeasureResult(costs=(0.0028652725714285714,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8740947246551514, timestamp=1663712499.211997) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
+ No: 10 GFLOPS: 0.00/80.80 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1280,8 +1280,8 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5092711
- No: 11 GFLOPS: 260.39/260.39 result: MeasureResult(costs=(0.0008890432960893855,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7833216190338135, timestamp=1663707310.865398) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
- No: 12 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+ No: 11 GFLOPS: 260.30/260.30 result: MeasureResult(costs=(0.0008893510837988827,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4758203029632568, timestamp=1663712500.1406338) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
+ No: 12 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1404,7 +1404,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,183542
- No: 13 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+ No: 13 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1527,7 +1527,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2482196
- No: 14 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+ No: 14 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1650,9 +1650,9 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10306226
- No: 15 GFLOPS: 5.43/260.39 result: MeasureResult(costs=(0.042630027499999994,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8400635719299316, timestamp=1663707315.4601128) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
- No: 16 GFLOPS: 3.33/260.39 result: MeasureResult(costs=(0.0694302805,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.599075078964233, timestamp=1663707316.6981018) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
- No: 17 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+ No: 15 GFLOPS: 5.29/260.30 result: MeasureResult(costs=(0.04375817275,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8518753051757812, timestamp=1663712504.731286) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
+ No: 16 GFLOPS: 3.33/260.30 result: MeasureResult(costs=(0.069439056,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.605278015136719, timestamp=1663712505.966001) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
+ No: 17 GFLOPS: 0.00/260.30 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
@@ -1670,8 +1670,8 @@ for this template
TimeoutError
[('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10195251
- No: 18 GFLOPS: 27.41/260.39 result: MeasureResult(costs=(0.008445743117647057,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3354072570800781, timestamp=1663707327.7962105) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
- No: 19 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+ No: 18 GFLOPS: 28.40/260.30 result: MeasureResult(costs=(0.008150162214285715,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2855777740478516, timestamp=1663712517.0024183) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
+ No: 19 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1794,7 +1794,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6956993
- No: 20 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+ No: 20 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1973,7 +1973,7 @@ and measure running time.
Best config:
[('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
Finish loading 20 records
- Time cost of this operator: 0.001283
+ Time cost of this operator: 0.001228
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 1ac1f5ae67..f56280562b 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -327,10 +327,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.5 98.728 (1, 2, 10, 10, 3) 2 1 [312.5]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.074 0.971 (1, 6, 10, 10) 1 1 [3.074]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.954 0.301 (1, 1, 10, 10, 3) 1 1 [0.954]
- Total_time - 316.528 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 313.0 98.677 (1, 2, 10, 10, 3) 2 1 [313.0]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.216 1.014 (1, 6, 10, 10) 1 1 [3.216]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.981 0.309 (1, 1, 10, 10, 3) 1 1 [0.981]
+ Total_time - 317.197 - - - - -
@@ -394,10 +394,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 119.8 97.752 (1, 6, 10, 10, 1) 2 1 [119.8]
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.783 1.455 (1, 6, 10, 10) 1 1 [1.783]
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.973 0.794 (1, 1, 10, 10, 3) 1 1 [0.973]
- Total_time - 122.555 - - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 122.6 97.796 (1, 6, 10, 10, 1) 2 1 [122.6]
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.777 1.417 (1, 6, 10, 10) 1 1 [1.777]
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.986 0.787 (1, 1, 10, 10, 3) 1 1 [0.986]
+ Total_time - 125.363 - - - - -
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
index 5de2c94a7b..cade13393d 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -225,7 +225,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
.. code-block:: none
- '/tmp/tmplsnuy6r_/images/random'
+ '/tmp/tmp5yhrl4fv/images/random'
@@ -325,8 +325,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
.. code-block:: none
- /tmp/tmplsnuy6r_/images/target contains 8144 images
- /tmp/tmplsnuy6r_/images/random contains 5000 images
+ /tmp/tmp5yhrl4fv/images/target contains 8144 images
+ /tmp/tmp5yhrl4fv/images/random contains 5000 images
@@ -501,13 +501,13 @@ the time on our validation set).
.. code-block:: none
Epoch 1/3
- 328/328 - 47s - loss: 0.2198 - accuracy: 0.9264 - val_loss: 0.1292 - val_accuracy: 0.9573 - 47s/epoch - 143ms/step
+ 328/328 - 46s - loss: 0.2262 - accuracy: 0.9234 - val_loss: 0.1417 - val_accuracy: 0.9592 - 46s/epoch - 142ms/step
Epoch 2/3
- 328/328 - 43s - loss: 0.1002 - accuracy: 0.9637 - val_loss: 0.1191 - val_accuracy: 0.9619 - 43s/epoch - 133ms/step
+ 328/328 - 43s - loss: 0.1036 - accuracy: 0.9615 - val_loss: 0.1216 - val_accuracy: 0.9603 - 43s/epoch - 132ms/step
Epoch 3/3
- 328/328 - 43s - loss: 0.0696 - accuracy: 0.9729 - val_loss: 0.0979 - val_accuracy: 0.9687 - 43s/epoch - 132ms/step
+ 328/328 - 43s - loss: 0.0675 - accuracy: 0.9731 - val_loss: 0.1119 - val_accuracy: 0.9694 - 43s/epoch - 132ms/step
- <keras.callbacks.History object at 0x7f0c039f4d10>
+ <keras.callbacks.History object at 0x7ff418e26f90>
@@ -864,7 +864,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 4 minutes 41.615 seconds)
+ **Total running time of the script:** ( 4 minutes 44.758 seconds)
.. _sphx_glr_download_how_to_work_with_microtvm_micro_train.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
index 4410eea0f8..e6323bf42c 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,16 +5,16 @@
Computation times
=================
-**05:36.478** total execution time for **how_to_work_with_microtvm** files:
+**05:38.521** total execution time for **how_to_work_with_microtvm** files:
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:41.615 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``) | 04:44.758 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:43.160 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``) | 00:42.562 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:08.329 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``) | 00:07.931 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.371 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``) | 00:03.268 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``) | 00:00.001 | 0.0 MB |
+---------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index 8bbd41b594..21ccdfaeab 100644
--- a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
Computation times
=================
-**00:43.951** total execution time for **how_to_work_with_relay** files:
+**00:41.004** total execution time for **how_to_work_with_relay** files:
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:31.856 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:31.108 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:10.101 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``) | 00:08.643 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.987 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``) | 00:01.245 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``) | 00:00.007 | 0.0 MB |
+----------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
index 41f26b8749..fc90595a26 100644
--- a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
@@ -261,7 +261,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
.. code-block:: none
- <function my_cuda_math_rule at 0x7f0b869dee60>
+ <function my_cuda_math_rule at 0x7ff3ab1d6710>
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 652a3b4ea4..a293c665d1 100644
--- a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
@@ -5,20 +5,20 @@
Computation times
=================
-**00:07.975** total execution time for **how_to_work_with_schedules** files:
+**00:04.188** total execution time for **how_to_work_with_schedules** files:
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:05.671 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``) | 00:02.060 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:01.052 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``) | 00:00.949 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.547 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``) | 00:00.509 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.526 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``) | 00:00.489 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.099 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``) | 00:00.101 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.039 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.040 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``) | 00:00.027 | 0.0 MB |
+------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
index f848499dbc..c1c627d13e 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -347,7 +347,7 @@ The importing needs to happen before the tensorized GEMV being executed.
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmptk9nsssc/input0.cc'\nsource_filename = \"/tmp/tmptk9nsssc/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/tmp_tj6totc/input0.cc'\nsource_filename = \"/tmp/tmp_tj6totc/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 112d51f343..895df6dcc0 100644
--- a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:21.890** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:21.660** total execution time for **topic_vta_tutorials_autotvm** files:
+---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:21.883 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:21.654 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``) | 00:00.006 | 0.0 MB |
+---------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
index ae189286de..9ef0795a3c 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -289,7 +289,7 @@ The compilation steps are:
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
- resnet18_v1 inference graph built in 23.46s!
+ resnet18_v1 inference graph built in 23.61s!
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 f975a6bcf8..6eebc0d683 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -333,7 +333,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 16.56s!
+ yolov3-tiny inference graph built in 16.46s!
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 fc90d42f20..599a4a29e6 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**01:32.992** total execution time for **topic_vta_tutorials_frontend** files:
+**01:32.894** total execution time for **topic_vta_tutorials_frontend** files:
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:49.229 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``) | 00:49.013 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:43.763 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:43.881 | 0.0 MB |
+------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
index 7a394be9c4..92c2aa6e18 100644
--- a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:03.014** total execution time for **topic_vta_tutorials_optimize** files:
+**00:02.910** total execution time for **topic_vta_tutorials_optimize** files:
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.619 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``) | 00:02.548 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.395 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.362 | 0.0 MB |
+--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
index bfd9e928ec..e801940c4e 100644
--- a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:00.737** total execution time for **topic_vta_tutorials** files:
+**00:00.672** total execution time for **topic_vta_tutorials** files:
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.400 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.366 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.337 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.307 | 0.0 MB |
+---------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
index 9a6bee8dfd..7e0d0285ce 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -326,7 +326,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 93.402 ms
+ Execution time of this operator: 93.416 ms
diff --git a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
index c5873948be..e17a9b5c93 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -462,16 +462,16 @@ reduce variance, we take 5 measurements and average them.
waiting for device...
device available
Get devices for measurement successfully!
- No: 1 GFLOPS: 10.45/10.45 result: MeasureResult(costs=(0.025676380999999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5453996658325195, timestamp=1663706073.7865336) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
- No: 2 GFLOPS: 2.91/10.45 result: MeasureResult(costs=(0.0922461064,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6257431507110596, timestamp=1663706075.4261174) [('tile_y', [-1, 4]), ('tile_x', [-1, 8])],None,32
- No: 3 GFLOPS: 11.80/11.80 result: MeasureResult(costs=(0.0227542642,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5581610202789307, timestamp=1663706076.4931116) [('tile_y', [-1, 64]), ('tile_x', [-1, 32])],None,56
- No: 4 GFLOPS: 1.84/11.80 result: MeasureResult(costs=(0.14601579280000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.4509077072143555, timestamp=1663706079.531627) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
- No: 5 GFLOPS: 3.66/11.80 result: MeasureResult(costs=(0.0734399346,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3130671977996826, timestamp=1663706080.9727597) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
- No: 6 GFLOPS: 1.75/11.80 result: MeasureResult(costs=(0.1534541226,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5839152336120605, timestamp=1663706084.139101) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
- No: 7 GFLOPS: 0.87/11.80 result: MeasureResult(costs=(0.31012123259999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.081417083740234, timestamp=1663706089.2654505) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
- No: 8 GFLOPS: 10.36/11.80 result: MeasureResult(costs=(0.025901798600000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5589027404785156, timestamp=1663706089.8419993) [('tile_y', [-1, 4]), ('tile_x', [-1, 64])],None,62
- No: 9 GFLOPS: 1.90/11.80 result: MeasureResult(costs=(0.1414536026,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3885984420776367, timestamp=1663706092.3508565) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
- No: 10 GFLOPS: 2.76/11.80 result: MeasureResult(costs=(0.0973485894,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6855683326721191, timestamp=1663706094.0729458) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
+ No: 1 GFLOPS: 9.80/9.80 result: MeasureResult(costs=(0.0273793654,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5732645988464355, timestamp=1663711266.8987334) [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
+ No: 2 GFLOPS: 2.56/9.80 result: MeasureResult(costs=(0.104768857,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8268074989318848, timestamp=1663711268.738484) [('tile_y', [-1, 4]), ('tile_x', [-1, 8])],None,32
+ No: 3 GFLOPS: 11.76/11.76 result: MeasureResult(costs=(0.0228239882,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5408790111541748, timestamp=1663711269.8149827) [('tile_y', [-1, 64]), ('tile_x', [-1, 32])],None,56
+ No: 4 GFLOPS: 1.65/11.76 result: MeasureResult(costs=(0.163133432,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.7305753231048584, timestamp=1663711272.5885227) [('tile_y', [-1, 1]), ('tile_x', [-1, 4])],None,20
+ No: 5 GFLOPS: 3.56/11.76 result: MeasureResult(costs=(0.0752988458,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3465051651000977, timestamp=1663711274.0647159) [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
+ No: 6 GFLOPS: 1.74/11.76 result: MeasureResult(costs=(0.154619831,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.6032111644744873, timestamp=1663711277.2497196) [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
+ No: 7 GFLOPS: 0.85/11.76 result: MeasureResult(costs=(0.3159693576,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.1865622997283936, timestamp=1663711283.032408) [('tile_y', [-1, 512]), ('tile_x', [-1, 2])],None,19
+ No: 8 GFLOPS: 10.64/11.76 result: MeasureResult(costs=(0.025230924,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5501112937927246, timestamp=1663711283.6006284) [('tile_y', [-1, 4]), ('tile_x', [-1, 64])],None,62
+ No: 9 GFLOPS: 1.49/11.76 result: MeasureResult(costs=(0.1805563496,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.9932615756988525, timestamp=1663711286.714605) [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
+ No: 10 GFLOPS: 2.66/11.76 result: MeasureResult(costs=(0.10108022720000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.713489294052124, timestamp=1663711288.4851985) [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 3e8dbef1d3..1cdb9a59b5 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -320,7 +320,7 @@ standard deviation.
.. code-block:: none
- {'mean': 518.0053364100012, 'median': 518.0258245499999, 'std': 1.2164873555957136}
+ {'mean': 512.3958194400075, 'median': 513.2557159999578, 'std': 2.3685676595730016}
@@ -554,30 +554,30 @@ the tuning data to.
.. code-block:: none
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 17.48/ 17.48 GFLOPS | Progress: (4/20) | 6.52 s
[Task 1/25] Current/Best: 6.05/ 17.48 GFLOPS | Progress: (8/20) | 9.59 s
[Task 1/25] Current/Best: 10.98/ 22.08 GFLOPS | Progress: (12/20) | 12.13 s
[Task 1/25] Current/Best: 16.41/ 22.18 GFLOPS | Progress: (16/20) | 13.84 s
[Task 1/25] Current/Best: 11.32/ 23.55 GFLOPS | Progress: (20/20) | 15.62 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.22/ 12.44 GFLOPS | Progress: (4/20) | 3.98 s
[Task 2/25] Current/Best: 12.54/ 17.86 GFLOPS | Progress: (8/20) | 5.30 s
[Task 2/25] Current/Best: 20.61/ 20.61 GFLOPS | Progress: (12/20) | 6.67 s
[Task 2/25] Current/Best: 10.99/ 20.61 GFLOPS | Progress: (16/20) | 7.98 s
[Task 2/25] Current/Best: 17.90/ 20.61 GFLOPS | Progress: (20/20) | 9.61 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 3/25] Current/Best: 1.62/ 10.07 GFLOPS | Progress: (4/20) | 5.93 s
[Task 3/25] Current/Best: 15.35/ 16.81 GFLOPS | Progress: (8/20) | 7.91 s
[Task 3/25] Current/Best: 14.99/ 16.81 GFLOPS | Progress: (12/20) | 9.68 s
[Task 3/25] Current/Best: 6.83/ 23.12 GFLOPS | Progress: (16/20) | 11.66 s
[Task 3/25] Current/Best: 11.01/ 23.12 GFLOPS | Progress: (20/20) | 16.33 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 8.99/ 19.07 GFLOPS | Progress: (4/20) | 2.45 s
[Task 4/25] Current/Best: 6.33/ 19.07 GFLOPS | Progress: (8/20) | 7.28 s
[Task 4/25] Current/Best: 21.09/ 21.09 GFLOPS | Progress: (12/20) | 12.36 s
[Task 4/25] Current/Best: 16.58/ 21.09 GFLOPS | Progress: (16/20) | 14.81 s
[Task 4/25] Current/Best: 12.85/ 21.09 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: 9.07/ 9.64 GFLOPS | Progress: (4/20) | 2.66 s
[Task 5/25] Current/Best: 11.66/ 11.66 GFLOPS | Progress: (8/20) | 4.77 s
[Task 5/25] Current/Best: 11.10/ 18.00 GFLOPS | Progress: (12/20) | 7.88 s
[Task 5/25] Current/Best: 11.42/ 20.94 GFLOPS | Progress: (16/20) | 9.31 s
[Task 5/25] Current/Best: 11.92/ 20.97 GFLOPS | Progress: (20/20) | 11.22 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 12.02/ 19.96 GFLOPS | Progress: (4/20) | 4.21 s
[Task 6/25] Current/Best: 18.90/ 19.96 GFLOPS | Progress: (8/20) | 6.00 s
[Task 6/25] Current/Best: 13.23/ 19.96 GFLOPS | Progress: (12/20) | 8.02 s
[Task 6/25] Current/Best: 19.48/ 19.96 GFLOPS | Progress: (16/20) | 10.32 s
[Task 6/25] Current/Best: 3.76/ 19.96 GFLOPS | Progress: (20/20) | 12.90 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 9.77/ 11.99 GFLOPS | Progress: (4/20) | 3.80 s
[Task 7/25] Current/Best: 18.92/ 19.86 GFLOPS | Progress: (8/20) | 5.36 s
[Task 7/25] Current/Best: 13.36/ 19.86 GFLOPS | Progress: (12/20) | 7.32 s
[Task 7/25] Current/Best: 12.15/ 20.17 GFLOPS | Progress: (16/20) | 9.41 s
[Task 7/25] Current/Best: 5.82/ 20.41 GFLOPS | Progress: (20/20) | 11.95 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 10.05/ 14.18 GFLOPS | Progress: (4/20) | 2.96 s
[Task 8/25] Current/Best: 9.92/ 14.18 GFLOPS | Progress: (8/20) | 8.30 s
[Task 8/25] Current/Best: 13.30/ 14.18 GFLOPS | Progress: (12/20) | 14.92 s
[Task 8/25] Current/Best: 19.00/ 19.00 GFLOPS | Progress: (16/20) | 17.09 s
[Task 8/25] Current/Best: 18.87/ 19.00 GFLOPS | Progress: (20/20) | 24.32 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.32/ 14.32 GFLOPS | Progress: (4/20) | 12.03 s
[Task 9/25] Current/Best: 22.72/ 22.72 GFLOPS | Progress: (8/20) | 13.91 s
[Task 9/25] Current/Best: 7.74/ 22.72 GFLOPS | Progress: (12/20) | 16.52 s
[Task 9/25] Current/Best: 17.72/ 22.72 GFLOPS | Progress: (16/20) | 19.45 s
[Task 9/25] Current/Best: 8.80/ 22.72 GFLOPS | Progress: (20/20) | 28.24 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.05/ 18.05 GFLOPS | Progress: (4/20) | 2.64 s
[Task 10/25] Current/Best: 15.61/ 18.05 GFLOPS | Progress: (8/20) | 4.28 s
[Task 10/25] Current/Best: 11.36/ 18.81 GFLOPS | Progress: (12/20) | 5.84 s
[Task 10/25] Current/Best: 18.96/ 19.94 GFLOPS | Progress: (16/20) | 6.96 s
[Task 10/25] Current/Best: 7.97/ 19.94 GFLOPS | Progress: (20/20
) | 8.51 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 10.78/ 18.25 GFLOPS | Progress: (4/20) | 3.47 s
[Task 11/25] Current/Best: 14.89/ 18.25 GFLOPS | Progress: (8/20) | 6.36 s
[Task 11/25] Current/Best: 15.92/ 18.25 GFLOPS | Progress: (12/20) | 8.45 s
[Task 11/25] Current/Best: 11.84/ 20.55 GFLOPS | Progress: (16/20) | 11.46 s
[Task 11/25] Current/Best: 17.98/ 20.55 GFLOPS | Progress: (20/20) | 13.62 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.74/ 18.10 GFLOPS | Progress: (4/20) | 5.89 s
[Task 12/25] Current/Best: 4.88/ 18.10 GFLOPS | Progress: (8/20) | 9.91 s
[Task 12/25] Current/Best: 18.73/ 18.73 GFLOPS | Progress: (12/20) | 11.97 s
[Task 12/25] Current/Best: 14.94/ 18.73 GFLOPS | Progress: (16/20) | 15.01 s
[Task 12/25] Current/Best: 15.16/ 18.73 GFLOPS | Progress: (20/20) | 16.96 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.57/ 17.32 GFLOPS | Progress: (4/20) | 3.81 s
[Task 13/25] Current/Best: 15.53/ 20.67 GFLOPS | Progress: (8/20) | 6.46 s
[Task 13/25] Current/Best: 18.52/ 21.38 GFLOPS | Progress: (12/20) | 9.53 s
[Task 13/25] Current/Best: 12.18/ 21.38 GFLOPS | Progress: (16/20) | 13.01 s
[Task 13/25] Current/Best: 17.68/ 21.38 GFLOPS | Progress: (20/20) | 15.43 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 12.06/ 13.25 GFLOPS | Progress: (4/20) | 3.42 s
[Task 14/25] Current/Best: 6.03/ 13.25 GFLOPS | Progress: (8/20) | 5.68 s
[Task 14/25] Current/Best: 18.86/ 19.01 GFLOPS | Progress: (12/20) | 8.41 s
[Task 14/25] Current/Best: 15.49/ 19.01 GFLOPS | Progress: (16/20) | 10.08 s Done.
-
[Task 14/25] Current/Best: 17.12/ 19.01 GFLOPS | Progress: (20/20) | 11.84 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 15.60/ 16.52 GFLOPS | Progress: (4/20) | 2.78 s
[Task 15/25] Current/Best: 12.65/ 17.53 GFLOPS | Progress: (8/20) | 4.14 s
[Task 15/25] Current/Best: 9.89/ 19.94 GFLOPS | Progress: (12/20) | 6.54 s
[Task 15/25] Current/Best: 20.30/ 20.30 GFLOPS | Progress: (16/20) | 9.78 s
[Task 15/25] Current/Best: 9.52/ 20.30 GFLOPS | Progress: (20/20) | 10.81 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 18.74/ 18.74 GFLOPS | Progress: (4/20) | 3.07 s
[Task 16/25] Current/Best: 3.03/ 18.74 GFLOPS | Progress: (8/20) | 4.70 s
[Task 16/25] Current/Best: 17.88/ 19.25 GFLOPS | Progress: (12/20) | 5.93 s
[Task 16/25] Current/Best: 18.04/ 19.25 GFLOPS | Progress: (16/20) |
7.31 s
[Task 16/25] Current/Best: 9.94/ 21.46 GFLOPS | Progress: (20/20) | 9.49 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 11.91/ 16.00 GFLOPS | Progress: (4/20) | 4.92 s
[Task 17/25] Current/Best: 12.69/ 21.81 GFLOPS | Progress: (8/20) | 7.83 s
[Task 17/25] Current/Best: 16.36/ 21.81 GFLOPS | Progress: (12/20) | 9.95 s
[Task 17/25] Current/Best: 16.45/ 21.81 GFLOPS | Progress: (16/20) | 12.20 s
[Task 17/25] Current/Best: 9.96/ 21.81 GFLOPS | Progress: (20/20) | 14.38 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 9.83/ 15.86 GFLOPS | Progress: (4/20) | 3.94 s
[Task 18/25] Current/Best: 10.53/ 16.37 GFLOPS | Progress: (8/20) | 7.69 s
[Task 18/25] Current/Best: 18.83/ 18.83 GFLOPS | Progress: (12/20) | 9.66 s
[Task 18/25] Current/Best: 9.99/ 18.83 GFLOPS | Progress: (16/20) | 13.61 s
[Task 18/25] Current/Best: 20.38/ 20.38 GFLOPS | Progress: (20/20) | 15.18 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 7.19/ 19.77 GFLOPS | Progress: (4/20) | 6.21 s
[Task 19/25] Current/Best: 2.68/ 19.77 GFLOPS | Progress: (8/20) | 9.54 s
[Task 19/25] Current/Best: 17.94/ 20.48 GFLOPS | Progress: (12/20) | 12.53 s
[Task 19/25] Current/Best: 13.26/ 20.48 GFLOPS | Progress: (16/20) | 15.60 s
[Task 19/25] Current/Best: 2.69/ 21.15 GFLOPS | Progress: (20/20) | 18.43 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 8.61/ 15.31 GFLOPS | Progress: (4/20) | 3.40 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 1/25] Current/Best: 17.42/ 17.42 GFLOPS | Progress: (4/20) | 7.02 s
[Task 1/25] Current/Best: 6.09/ 17.42 GFLOPS | Progress: (8/20) | 9.56 s
[Task 1/25] Current/Best: 11.17/ 22.24 GFLOPS | Progress: (12/20) | 12.06 s
[Task 1/25] Current/Best: 16.46/ 22.30 GFLOPS | Progress: (16/20) | 13.77 s
[Task 1/25] Current/Best: 11.31/ 23.64 GFLOPS | Progress: (20/20) | 15.55 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 2/25] Current/Best: 12.02/ 12.52 GFLOPS | Progress: (4/20) | 3.90 s
[Task 2/25] Current/Best: 12.51/ 18.17 GFLOPS | Progress: (8/20) | 5.22 s
[Task 2/25] Current/Best: 20.89/ 20.89 GFLOPS | Progress: (12/20) | 6.56 s
[Task 2/25] Current/Best: 11.20/ 20.89 GFLOPS | Progress: (16/20) | 7.84 s
[Task 2/25] Current/Best: 17.53/ 20.89 GFLOPS | Progress: (20/20) | 9.42 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/ 10.02 GFLOPS | Progress: (4/20) | 5.95 s
[Task 3/25] Current/Best: 15.35/ 16.86 GFLOPS | Progress: (8/20) | 7.92 s
[Task 3/25] Current/Best: 14.22/ 16.86 GFLOPS | Progress: (12/20) | 9.68 s
[Task 3/25] Current/Best: 6.81/ 23.31 GFLOPS | Progress: (16/20) | 11.67 s
[Task 3/25] Current/Best: 11.02/ 23.31 GFLOPS | Progress: (20/20) | 16.28 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 4/25] Current/Best: 8.94/ 18.47 GFLOPS | Progress: (4/20) | 2.45 s
[Task 4/25] Current/Best: 6.61/ 18.47 GFLOPS | Progress: (8/20) | 6.83 s
[Task 4/25] Current/Best: 21.49/ 21.49 GFLOPS | Progress: (12/20) | 11.31 s
[Task 4/25] Current/Best: 16.51/ 21.49 GFLOPS | Progress: (16/20) | 13.56 s
[Task 4/25] Current/Best: 12.87/ 21.49 GFLOPS | Progress: (20/20) | 15.57 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 5/25] Current/Best: 9.00/ 9.76 GFLOPS | Progress: (4/20) | 2.67 s
[Task 5/25] Current/Best: 11.60/ 11.60 GFLOPS | Progress: (8/20) | 4.76 s
[Task 5/25] Current/Best: 10.98/ 18.04 GFLOPS | Progress: (12/20) | 7.89 s
[Task 5/25] Current/Best: 11.44/ 22.14 GFLOPS | Progress: (16/20) | 9.36 s
[Task 5/25] Current/Best: 12.13/ 22.14 GFLOPS | Progress: (20/20) | 11.23 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 6/25] Current/Best: 11.97/ 19.94 GFLOPS | Progress: (4/20) | 4.04 s
[Task 6/25] Current/Best: 18.81/ 19.94 GFLOPS | Progress: (8/20) | 5.83 s
[Task 6/25] Current/Best: 13.20/ 19.94 GFLOPS | Progress: (12/20) | 7.82 s
[Task 6/25] Current/Best: 19.56/ 19.94 GFLOPS | Progress: (16/20) | 10.09 s
[Task 6/25] Current/Best: 3.70/ 19.94 GFLOPS | Progress: (20/20) | 12.69 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 7/25] Current/Best: 9.77/ 11.97 GFLOPS | Progress: (4/20) | 3.71 s
[Task 7/25] Current/Best: 19.50/ 19.86 GFLOPS | Progress: (8/20) | 5.25 s
[Task 7/25] Current/Best: 15.97/ 19.86 GFLOPS | Progress: (12/20) | 7.19 s
[Task 7/25] Current/Best: 12.14/ 20.07 GFLOPS | Progress: (16/20) | 9.29 s
[Task 7/25] Current/Best: 6.01/ 20.07 GFLOPS | Progress: (20/20) | 11.84 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 8/25] Current/Best: 10.12/ 13.69 GFLOPS | Progress: (4/20) | 2.99 s
[Task 8/25] Current/Best: 9.43/ 13.69 GFLOPS | Progress: (8/20) | 7.81 s
[Task 8/25] Current/Best: 12.77/ 13.69 GFLOPS | Progress: (12/20) | 13.95 s
[Task 8/25] Current/Best: 19.01/ 19.01 GFLOPS | Progress: (16/20) | 16.08 s
[Task 8/25] Current/Best: 19.27/ 19.27 GFLOPS | Progress: (20/20) | 22.65 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 9/25] Current/Best: 14.28/ 14.28 GFLOPS | Progress: (4/20) | 12.00 s
[Task 9/25] Current/Best: 23.08/ 23.08 GFLOPS | Progress: (8/20) | 13.89 s
[Task 9/25] Current/Best: 7.98/ 23.08 GFLOPS | Progress: (12/20) | 16.30 s
[Task 9/25] Current/Best: 17.90/ 23.08 GFLOPS | Progress: (16/20) | 18.96 s
[Task 9/25] Current/Best: 8.97/ 23.08 GFLOPS | Progress: (20/20) | 26.67 s
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 10/25] Current/Best: 18.01/ 18.01 GFLOPS | Progress: (4/20) | 2.66 s
[Task 10/25] Current/Best: 15.68/ 18.01 GFLOPS | Progress: (8/20) | 4.25 s
[Task 10/25] Current/Best: 11.42/ 18.51 GFLOPS | Progress: (12/20) | 5.80 s
[Task 10/25] Current/Best: 19.16/ 20.08 GFLOPS | Progress: (16/20) | 6.92 s
[Task 10/25] Current/Best: 8.40/ 20.08 GFLOPS | Progress: (20/20
) | 8.47 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 11/25] Current/Best: 10.63/ 18.14 GFLOPS | Progress: (4/20) | 3.40 s
[Task 11/25] Current/Best: 14.84/ 18.14 GFLOPS | Progress: (8/20) | 6.18 s
[Task 11/25] Current/Best: 15.92/ 18.14 GFLOPS | Progress: (12/20) | 8.25 s
[Task 11/25] Current/Best: 11.80/ 20.60 GFLOPS | Progress: (16/20) | 11.04 s
[Task 11/25] Current/Best: 18.52/ 20.60 GFLOPS | Progress: (20/20) | 13.13 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 12/25] Current/Best: 7.77/ 18.10 GFLOPS | Progress: (4/20) | 5.38 s
[Task 12/25] Current/Best: 5.01/ 18.10 GFLOPS | Progress: (8/20) | 9.15 s
[Task 12/25] Current/Best: 18.91/ 18.91 GFLOPS | Progress: (12/20) | 11.21 s
[Task 12/25] Current/Best: 15.13/ 18.91 GFLOPS | Progress: (16/20) | 14.07 s
[Task 12/25] Current/Best: 15.00/ 18.91 GFLOPS | Progress: (20/20) | 16.02 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 13/25] Current/Best: 8.55/ 17.22 GFLOPS | Progress: (4/20) | 3.78 s
[Task 13/25] Current/Best: 14.72/ 20.66 GFLOPS | Progress: (8/20) | 6.25 s
[Task 13/25] Current/Best: 18.72/ 20.66 GFLOPS | Progress: (12/20) | 9.22 s
[Task 13/25] Current/Best: 12.22/ 20.66 GFLOPS | Progress: (16/20) | 12.62 s
[Task 13/25] Current/Best: 16.79/ 20.66 GFLOPS | Progress: (20/20) | 14.94 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 14/25] Current/Best: 12.10/ 13.30 GFLOPS | Progress: (4/20) | 3.36 s
[Task 14/25] Current/Best: 6.05/ 13.30 GFLOPS | Progress: (8/20) | 5.55 s
[Task 14/25] Current/Best: 19.18/ 19.18 GFLOPS | Progress: (12/20) | 8.15 s
[Task 14/25] Current/Best: 15.25/ 19.18 GFLOPS | Progress: (16/20) | 9.85 s Done.
+
[Task 14/25] Current/Best: 16.93/ 19.18 GFLOPS | Progress: (20/20) | 11.60 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 15/25] Current/Best: 15.08/ 17.25 GFLOPS | Progress: (4/20) | 2.80 s
[Task 15/25] Current/Best: 12.65/ 17.75 GFLOPS | Progress: (8/20) | 4.19 s
[Task 15/25] Current/Best: 9.35/ 21.06 GFLOPS | Progress: (12/20) | 6.28 s
[Task 15/25] Current/Best: 19.80/ 21.06 GFLOPS | Progress: (16/20) | 9.18 s
[Task 15/25] Current/Best: 9.51/ 21.06 GFLOPS | Progress: (20/20) | 10.17 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 16/25] Current/Best: 18.79/ 18.79 GFLOPS | Progress: (4/20) | 3.05 s
[Task 16/25] Current/Best: 3.03/ 18.79 GFLOPS | Progress: (8/20) | 4.68 s
[Task 16/25] Current/Best: 16.90/ 19.43 GFLOPS | Progress: (12/20) | 5.91 s
[Task 16/25] Current/Best: 18.27/ 19.43 GFLOPS | Progress: (16/20) |
7.29 s
[Task 16/25] Current/Best: 10.22/ 20.98 GFLOPS | Progress: (20/20) | 9.35 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 17/25] Current/Best: 12.76/ 16.07 GFLOPS | Progress: (4/20) | 4.81 s
[Task 17/25] Current/Best: 13.28/ 22.17 GFLOPS | Progress: (8/20) | 7.68 s
[Task 17/25] Current/Best: 16.43/ 22.17 GFLOPS | Progress: (12/20) | 9.81 s
[Task 17/25] Current/Best: 16.40/ 22.17 GFLOPS | Progress: (16/20) | 11.97 s
[Task 17/25] Current/Best: 9.97/ 22.17 GFLOPS | Progress: (20/20) | 14.10 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 18/25] Current/Best: 10.37/ 16.96 GFLOPS | Progress: (4/20) | 3.81 s
[Task 18/25] Current/Best: 10.46/ 17.65 GFLOPS | Progress: (8/20) | 7.28 s
[Task 18/25] Current/Best: 19.37/ 19.37 GFLOPS | Progress: (12/20) | 9.26 s
[Task 18/25] Current/Best: 10.02/ 19.37 GFLOPS | Progress: (16/20) | 12.84 s
[Task 18/25] Current/Best: 20.59/ 20.59 GFLOPS | Progress: (20/20) | 14.41 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 19/25] Current/Best: 7.18/ 19.17 GFLOPS | Progress: (4/20) | 6.17 s
[Task 19/25] Current/Best: 2.68/ 19.17 GFLOPS | Progress: (8/20) | 9.41 s
[Task 19/25] Current/Best: 17.87/ 20.11 GFLOPS | Progress: (12/20) | 12.19 s
[Task 19/25] Current/Best: 13.61/ 20.68 GFLOPS | Progress: (16/20) | 15.04 s
[Task 19/25] Current/Best: 2.69/ 22.16 GFLOPS | Progress: (20/20) | 17.88 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 20/25] Current/Best: 9.24/ 15.20 GFLOPS | Progress: (4/20) | 3.39 s Done.
Done.
-
[Task 20/25] Current/Best: 9.79/ 15.31 GFLOPS | Progress: (8/20) | 6.99 s
[Task 20/25] Current/Best: 2.32/ 15.31 GFLOPS | Progress: (12/20) | 10.92 s
[Task 20/25] Current/Best: 11.07/ 15.31 GFLOPS | Progress: (16/20) | 14.75 s
[Task 20/25] Current/Best: 11.73/ 21.62 GFLOPS | Progress: (20/20) | 16.89 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.35/ 17.67 GFLOPS | Progress: (4/20) | 3.32 s
[Task 21/25] Current/Best: 14.54/ 17.67 GFLOPS | Progress: (8/20) | 4.93 s
[Task 21/25] Current/Best: 1.61/ 17.67 GFLOPS | Progress: (12/20) | 7.09 s
[Task 21/25] Current/Best: 15.99/ 17.67 GFLOPS | Progress: (16/20) | 10.67 s
[Task 21/25] Current/Best: 4.46/ 17.67 GFLOPS | Progress: (20/20) | 18.13 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.70/ 16.82 GFLOPS | Progress: (4/20
) | 2.76 s
[Task 22/25] Current/Best: 9.34/ 20.89 GFLOPS | Progress: (8/20) | 4.80 s
[Task 22/25] Current/Best: 19.41/ 20.89 GFLOPS | Progress: (12/20) | 7.24 s
[Task 22/25] Current/Best: 15.57/ 20.89 GFLOPS | Progress: (16/20) | 9.38 s
[Task 22/25] Current/Best: 13.09/ 20.89 GFLOPS | Progress: (20/20) | 11.16 s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 23/25] Current/Best: 16.63/ 19.76 GFLOPS | Progress: (4/20) | 3.34 s
[Task 23/25] Current/Best: 14.15/ 19.76 GFLOPS | Progress: (8/20) | 6.66 s
[Task 23/25] Current/Best: 20.38/ 21.16 GFLOPS | Progress: (12/20) | 8.52 s
[Task 23/25] Current/Best: 6.36/ 21.16 GFLOPS | Progress: (16/20) | 15.71 s
[Task 23/25] Current/Best: 7.56/ 21.16 GFLOPS | Progress: (20/20) | 20.01 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.15/ 8.15 GFLOPS | Progress: (4/20) | 11.85 s
[Task 24/25] Current/Best: 1.91/ 8.15 GFLOPS | Progress: (8/20) | 22.87 s
[Task 24/25] Current/Best: 3.96/ 8.15 GFLOPS | Progress: (12/20) | 34.47 s Done.
-
[Task 24/25] Current/Best: 5.72/ 8.71 GFLOPS | Progress: (16/20) | 40.22 s
[Task 24/25] Current/Best: 2.96/ 8.71 GFLOPS | Progress: (20/20) | 46.30 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/ 2.85 GFLOPS | Progress: (4/20) | 11.65 s
[Task 25/25] Current/Best: 5.67/ 7.78 GFLOPS | Progress: (8/20) | 22.98 s
[Task 25/25] Current/Best: 5.96/ 7.78 GFLOPS | Progress: (12/20) | 34.47 s
[Task 25/25] Current/Best: 5.74/ 8.57 GFLOPS | Progress: (16/20) | 36.24 s
[Task 25/25] Current/Best: 2.87/ 8.57 GFLOPS | Progress: (20/20) | 46.96 s
+
[Task 20/25] Current/Best: 10.33/ 15.20 GFLOPS | Progress: (8/20) | 6.84 s
[Task 20/25] Current/Best: 2.32/ 15.20 GFLOPS | Progress: (12/20) | 10.79 s
[Task 20/25] Current/Best: 11.06/ 15.20 GFLOPS | Progress: (16/20) | 14.45 s
[Task 20/25] Current/Best: 11.77/ 21.38 GFLOPS | Progress: (20/20) | 16.57 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 21/25] Current/Best: 6.36/ 17.71 GFLOPS | Progress: (4/20) | 3.28 s
[Task 21/25] Current/Best: 14.62/ 17.71 GFLOPS | Progress: (8/20) | 4.86 s
[Task 21/25] Current/Best: 1.61/ 17.71 GFLOPS | Progress: (12/20) | 7.04 s
[Task 21/25] Current/Best: 15.96/ 17.71 GFLOPS | Progress: (16/20) | 10.57 s
[Task 21/25] Current/Best: 4.45/ 17.71 GFLOPS | Progress: (20/20) | 17.71 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 22/25] Current/Best: 2.70/ 16.81 GFLOPS | Progress: (4/20
) | 2.76 s
[Task 22/25] Current/Best: 8.94/ 21.10 GFLOPS | Progress: (8/20) | 4.76 s
[Task 22/25] Current/Best: 19.91/ 21.10 GFLOPS | Progress: (12/20) | 7.07 s
[Task 22/25] Current/Best: 15.50/ 21.10 GFLOPS | Progress: (16/20) | 9.14 s
[Task 22/25] Current/Best: 12.47/ 21.10 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: 16.07/ 19.49 GFLOPS | Progress: (4/20) | 3.35 s
[Task 23/25] Current/Best: 14.12/ 19.80 GFLOPS | Progress: (8/20) | 6.66 s
[Task 23/25] Current/Best: 20.40/ 21.33 GFLOPS | Progress: (12/20) | 8.51 s
[Task 23/25] Current/Best: 6.33/ 21.33 GFLOPS | Progress: (16/20) | 15.61 s
[Task 23/25] Current/Best: 7.58/ 21.33 GFLOPS | Progress: (20/20) | 19.86 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 24/25] Current/Best: 8.45/ 8.45 GFLOPS | Progress: (4/20) | 11.84 s
[Task 24/25] Current/Best: 3.34/ 8.45 GFLOPS | Progress: (8/20) | 23.11 s
[Task 24/25] Current/Best: 3.60/ 8.45 GFLOPS | Progress: (12/20) | 33.87 s Done.
+
[Task 24/25] Current/Best: 6.50/ 8.65 GFLOPS | Progress: (16/20) | 39.30 s
[Task 24/25] Current/Best: 2.97/ 8.65 GFLOPS | Progress: (20/20) | 45.21 s Done.
+
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/20) | 0.00 s
[Task 25/25] Current/Best: 1.54/ 2.86 GFLOPS | Progress: (4/20) | 11.66 s
[Task 25/25] Current/Best: 5.74/ 7.62 GFLOPS | Progress: (8/20) | 22.93 s
[Task 25/25] Current/Best: 5.89/ 7.62 GFLOPS | Progress: (12/20) | 34.25 s
[Task 25/25] Current/Best: 5.71/ 9.00 GFLOPS | Progress: (16/20) | 36.15 s
[Task 25/25] Current/Best: 2.85/ 9.00 GFLOPS | Progress: (20/20) | 46.89 s
@@ -679,8 +679,8 @@ Verify that the optimized model runs and produces the same results:
.. code-block:: none
- class='n02123045 tabby, tabby cat' with probability=0.621105
- class='n02123159 tiger cat' with probability=0.356377
+ class='n02123045 tabby, tabby cat' with probability=0.621104
+ class='n02123159 tiger cat' with probability=0.356378
class='n02124075 Egyptian cat' with probability=0.019712
class='n02129604 tiger, Panthera tigris' with probability=0.001215
class='n04040759 radiator' with probability=0.000262
@@ -737,8 +737,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 410.64014174999784, 'median': 410.4473024000072, 'std': 1.0785906497546507}
- unoptimized: {'mean': 518.0053364100012, 'median': 518.0258245499999, 'std': 1.2164873555957136}
+ optimized: {'mean': 412.8904595699896, 'median': 412.6717511499919, 'std': 1.6529257233259869}
+ unoptimized: {'mean': 512.3958194400075, 'median': 513.2557159999578, 'std': 2.3685676595730016}
@@ -761,7 +761,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 10 minutes 32.947 seconds)
+ **Total running time of the script:** ( 10 minutes 20.824 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 aa3f28280b..4523194447 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -282,7 +282,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.28e-07 secs/op
+ 1.269e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 2bdf3aba9a..8016a7abd4 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -263,7 +263,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x22db0dd0)), stage(b, placeholder(b, 0x6bccc80)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min [...]
+ [stage(a, placeholder(a, 0x11d75f00)), stage(b, placeholder(b, 0xca735e0)), 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 35de022101..d16a9fabfb 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,32 +5,32 @@
Computation times
=================
-**13:22.929** total execution time for **tutorial** files:
+**13:15.273** total execution time for **tutorial** files:
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 10:32.947 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``) | 10:20.824 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:01.534 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``) | 01:00.569 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 00:51.300 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 00:55.229 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:31.136 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``) | 00:31.623 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:23.838 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``) | 00:25.200 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:01.313 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``) | 00:00.922 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.703 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``) | 00:00.726 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.150 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.171 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``) | 00:00.005 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.001 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``) | 00:00.002 | 0.0 MB |
++------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_tutorial_install.py` (``install.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``) | 00:00.001 | 0.0 MB |
+------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_install.py` (``install.py``) | 00:00.001 | 0.0 MB |
-+------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index be7ffef199..3bd48ae6f8 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -295,7 +295,7 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
Numpy running time: 0.000007
- naive: 0.000007
+ naive: 0.000009
@@ -394,7 +394,7 @@ compile and run this new schedule with the parallel operation applied:
.. code-block:: none
- parallel: 0.000008
+ parallel: 0.000010
@@ -449,7 +449,7 @@ factor to be the number of threads on your CPU.
.. code-block:: none
- vector: 0.000026
+ vector: 0.000025
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*n: int32)], [], type="auto"),
@@ -501,10 +501,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 6.7401699993752114e-06 1.0
- naive 6.7248e-06 0.9977196421786638
- parallel 8.1308e-06 1.2063197220179454
- vector 2.64398e-05 3.922720050451379
+ numpy 6.9486299980781045e-06 1.0
+ naive 9.004e-06 1.2957949987969404
+ parallel 1.0241100000000001e-05 1.4738300935339121
+ vector 2.45658e-05 3.5353443782147798
@@ -925,7 +925,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.019187
+ Numpy running time: 0.018461
@@ -983,7 +983,7 @@ optimizations.
.. code-block:: none
- none: 3.443409
+ none: 3.390452
@@ -1086,7 +1086,7 @@ schedule.
.. code-block:: none
- blocking: 0.309998
+ blocking: 0.306090
@@ -1182,7 +1182,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.341361
+ vectorization: 0.344180
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1256,7 +1256,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.118731
+ loop permutation: 0.117610
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1355,7 +1355,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.108204
+ array packing: 0.107842
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1448,7 +1448,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.110945
+ block caching: 0.110260
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1534,7 +1534,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.145861
+ parallelization: 0.146262
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1615,13 +1615,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.4434090983000005 1.0
- blocking 0.3099975139 0.09002633873885178
- vectorization 0.34136064990000003 0.09913450309129074
- loop permutation 0.1187310283 0.03448066288685161
- array packing 0.10820351209999998 0.03142336824091558
- block caching 0.11094454249999999 0.03221939053212496
- parallelization 0.1458608874 0.04235944183106533
+ none 3.3904522782 1.0
+ blocking 0.30608994370000003 0.09027997405187015
+ vectorization 0.34417996230000003 0.10151446888458376
+ loop permutation 0.11760968150000002 0.03468849340726875
+ array packing 0.1078417876 0.03180749314579749
+ block caching 0.1102601127 0.03252076821990763
+ parallelization 0.1462624872 0.04313952098380548
@@ -1663,7 +1663,7 @@ the computation for specific platforms.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 1.534 seconds)
+ **Total running time of the script:** ( 1 minutes 0.569 seconds)
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index b7c5b55198..54955a012a 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-5dfa8da00ec658934f3fc0df8eb9f41a167e1545
+534378b935aa08b77e7529ec183133a24f121ae4
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index 1c04bcf6bd..f140f57b9f 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -572,7 +572,7 @@ class:['truck 0.9266'] left:471 top:83 right:689 bottom:169
class:['bicycle 0.9984'] left:111 top:113 right:577 bottom:447
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 3.807 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 5.934 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-darknet-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7716f96385bd5abb6e822041e285be54/from_darknet.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_darknet.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_keras.html b/docs/how_to/compile_models/from_keras.html
index 962b8fe242..5dc5e812eb 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -493,7 +493,7 @@ pip install -U tensorflow --user
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Relay top-1 id: 285, class name: Egyptian cat
1/1 [==============================] - ETA: 0s
-1/1 [==============================] - 1s 968ms/step
+1/1 [==============================] - 1s 912ms/step
Keras top-1 id: 285, class name: Egyptian cat
</pre></div>
</div>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 7732f1e856..2fad082f6e 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -427,7 +427,7 @@ to download the full example code</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"x"</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip6d80a7c1-8e70-4fbe-aba5-9e9661b2a96b from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip63353ed5-629d-481e-9485-4ed76f6fd434 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 497c90ff18..abd600b4df 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -435,14 +435,15 @@ Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdo
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
0%| | 0.00/41.5M [00:00<?, ?B/s]
- 15%|#5 | 6.33M/41.5M [00:00<00:00, 40.8MB/s]
- 30%|##9 | 12.4M/41.5M [00:00<00:00, 51.5MB/s]
- 42%|####2 | 17.6M/41.5M [00:00<00:00, 43.1MB/s]
- 54%|#####4 | 22.5M/41.5M [00:00<00:00, 45.9MB/s]
- 65%|######5 | 27.1M/41.5M [00:00<00:00, 42.2MB/s]
- 77%|#######7 | 32.0M/41.5M [00:00<00:00, 44.4MB/s]
- 92%|#########2| 38.3M/41.5M [00:00<00:00, 48.1MB/s]
-100%|##########| 41.5M/41.5M [00:00<00:00, 43.6MB/s]
+ 15%|#5 | 6.33M/41.5M [00:00<00:01, 34.9MB/s]
+ 23%|##3 | 9.66M/41.5M [00:00<00:01, 28.2MB/s]
+ 35%|###4 | 14.3M/41.5M [00:00<00:00, 34.8MB/s]
+ 43%|####3 | 17.8M/41.5M [00:00<00:00, 34.4MB/s]
+ 54%|#####3 | 22.3M/41.5M [00:00<00:00, 37.0MB/s]
+ 63%|######2 | 26.0M/41.5M [00:00<00:00, 33.1MB/s]
+ 77%|#######7 | 32.0M/41.5M [00:00<00:00, 39.7MB/s]
+ 96%|#########6| 40.0M/41.5M [00:01<00:00, 46.8MB/s]
+100%|##########| 41.5M/41.5M [00:01<00:00, 40.6MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index e3db73cfdb..4c68d4990f 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -414,9 +414,10 @@ 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%|###8 | 17.4M/44.7M [00:00<00:00, 183MB/s]
- 94%|#########3| 42.0M/44.7M [00:00<00:00, 226MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 212MB/s]
+ 10%|# | 4.51M/44.7M [00:00<00:00, 47.2MB/s]
+ 20%|## | 9.02M/44.7M [00:00<00:00, 46.7MB/s]
+ 84%|########4 | 37.6M/44.7M [00:00<00:00, 162MB/s]
+100%|##########| 44.7M/44.7M [00:00<00:00, 144MB/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 d605c83893..325b034195 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -632,7 +632,7 @@ banana (score = 0.00022)
desk (score = 0.00019)
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 4.804 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 7.471 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-tensorflow-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7f1d3d1b878694c201c614c807cdebc8/from_tensorflow.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_tensorflow.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/sg_execution_times.html b/docs/how_to/compile_models/sg_execution_times.html
index 6ce33c94cd..cf21dc1aa1 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:11.069</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:18.170</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 81%" />
@@ -336,43 +336,43 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></td>
-<td><p>01:04.804</p></td>
+<td><p>01:07.471</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></td>
-<td><p>01:03.807</p></td>
+<td><p>01:05.934</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_paddle.html#sphx-glr-how-to-compile-models-from-paddle-py"><span class="std std-ref">Compile PaddlePaddle Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_paddle.py</span></code>)</p></td>
-<td><p>00:39.520</p></td>
+<td><p>00:42.285</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></td>
-<td><p>00:28.334</p></td>
+<td><p>00:28.329</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></td>
-<td><p>00:26.750</p></td>
+<td><p>00:26.348</p></td>
<td><p>0.0 MB</p></td>
</tr>
-<tr class="row-even"><td><p><a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></td>
-<td><p>00:24.968</p></td>
+<tr class="row-even"><td><p><a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></td>
+<td><p>00:25.428</p></td>
<td><p>0.0 MB</p></td>
</tr>
-<tr class="row-odd"><td><p><a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></td>
-<td><p>00:24.924</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></td>
+<td><p>00:23.215</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></td>
-<td><p>00:20.145</p></td>
+<td><p>00:20.231</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="from_keras.html#sphx-glr-how-to-compile-models-from-keras-py"><span class="std std-ref">Compile Keras Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_keras.py</span></code>)</p></td>
-<td><p>00:15.376</p></td>
+<td><p>00:16.452</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="from_onnx.html#sphx-glr-how-to-compile-models-from-onnx-py"><span class="std std-ref">Compile ONNX Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_onnx.py</span></code>)</p></td>
-<td><p>00:02.440</p></td>
+<td><p>00:02.478</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 34c06075c3..c699758e07 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -649,7 +649,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 15.9402 15.9377 16.1041 15.8109 0.0895
+ 15.8707 15.8574 16.0833 15.7741 0.0823
</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 35d504425e..c487ce8bd4 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -436,15 +436,13 @@ 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]
- 6%|5 | 10.1M/170M [00:00<00:01, 97.2MB/s]
- 17%|#6 | 28.9M/170M [00:00<00:00, 154MB/s]
- 30%|### | 51.5M/170M [00:00<00:00, 191MB/s]
- 42%|####2 | 72.1M/170M [00:00<00:00, 201MB/s]
- 57%|#####6 | 96.4M/170M [00:00<00:00, 220MB/s]
- 69%|######9 | 118M/170M [00:00<00:00, 218MB/s]
- 84%|########3 | 142M/170M [00:00<00:00, 232MB/s]
- 98%|#########8| 167M/170M [00:00<00:00, 241MB/s]
-100%|##########| 170M/170M [00:00<00:00, 217MB/s]
+ 11%|#1 | 19.2M/170M [00:00<00:00, 201MB/s]
+ 26%|##5 | 43.9M/170M [00:00<00:00, 235MB/s]
+ 41%|####1 | 70.3M/170M [00:00<00:00, 254MB/s]
+ 57%|#####7 | 97.0M/170M [00:00<00:00, 264MB/s]
+ 73%|#######2 | 124M/170M [00:00<00:00, 270MB/s]
+ 89%|########8 | 151M/170M [00:00<00:00, 274MB/s]
+100%|##########| 170M/170M [00:00<00:00, 261MB/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').
@@ -538,7 +536,7 @@ 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 1.844 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 2.671 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-object-detection-pytorch-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7795da4b258c8feff986668b95ef57ad/deploy_object_detection_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_object_detection_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized.html b/docs/how_to/deploy_models/deploy_prequantized.html
index ac79cbbcf6..0e0bb2e3fa 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -480,9 +480,10 @@ 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]
- 20%|## | 2.77M/13.6M [00:00<00:00, 28.9MB/s]
- 41%|#### | 5.52M/13.6M [00:00<00:00, 26.7MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 55.2MB/s]
+ 14%|#4 | 1.94M/13.6M [00:00<00:00, 20.2MB/s]
+ 52%|#####2 | 7.06M/13.6M [00:00<00:00, 39.8MB/s]
+ 80%|######## | 10.9M/13.6M [00:00<00:00, 39.1MB/s]
+100%|##########| 13.6M/13.6M [00:00<00:00, 39.3MB/s]
</pre></div>
</div>
</div>
@@ -567,7 +568,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 90.4286 90.2643 97.4376 90.1577 0.8035
+ 90.4631 90.3462 95.7868 90.1210 0.5973
</pre></div>
</div>
<div class="admonition note">
@@ -606,7 +607,7 @@ 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 9.629 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 10.165 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/fb8217c13f4351224c6cf3aacf1a87fc/deploy_prequantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized_tflite.html b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
index 6925abafc9..0d58edde03 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -569,7 +569,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 120.3406 120.2129 125.6875 119.4403 0.6940
+ 119.5574 119.5427 121.3536 118.7764 0.3845
</pre></div>
</div>
<div class="admonition note">
@@ -597,7 +597,7 @@ network for ARM CPU</span></a>.</p></li>
</ul>
</div></blockquote>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 59.808 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 52.478 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-tflite-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/56691c7a27d45da61d112276334640d3/deploy_prequantized_tflite.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized_tflite.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_quantized.html b/docs/how_to/deploy_models/deploy_quantized.html
index 76b0776c53..aadb132469 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -507,7 +507,7 @@ 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 30.152 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 29.408 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-quantized-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7810ecf51bfc05f7d5e8a400ac3e815d/deploy_quantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_quantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
index 2c6ef2b870..2bf5ac9fe6 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -441,23 +441,24 @@ to your device.</p>
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
0%| | 0/132723 [00:00<?, ?KB/s]
- 5%|5 | 6708/132723 [00:00<00:01, 67072.39KB/s]
- 11%|#1 | 15053/132723 [00:00<00:01, 76699.09KB/s]
- 18%|#7 | 23491/132723 [00:00<00:01, 80201.07KB/s]
- 24%|##4 | 31945/132723 [00:00<00:01, 81910.38KB/s]
- 30%|### | 40364/132723 [00:00<00:01, 82729.26KB/s]
- 37%|###6 | 48745/132723 [00:00<00:01, 83092.60KB/s]
- 43%|####3 | 57107/132723 [00:00<00:00, 83262.95KB/s]
- 49%|####9 | 65534/132723 [00:00<00:00, 83580.51KB/s]
- 56%|#####5 | 73974/132723 [00:00<00:00, 83834.82KB/s]
- 62%|######2 | 82358/132723 [00:01<00:00, 83803.26KB/s]
- 68%|######8 | 90739/132723 [00:01<00:00, 83778.40KB/s]
- 75%|#######4 | 99117/132723 [00:01<00:00, 81173.69KB/s]
- 81%|########1 | 107561/132723 [00:01<00:00, 82138.51KB/s]
- 87%|########7 | 115790/132723 [00:01<00:00, 49694.17KB/s]
- 94%|#########3| 124208/132723 [00:01<00:00, 56740.54KB/s]
-100%|#########9| 132650/132723 [00:01<00:00, 62994.34KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 72364.38KB/s]
+ 4%|4 | 5933/132723 [00:00<00:02, 59311.11KB/s]
+ 10%|# | 13545/132723 [00:00<00:01, 69195.10KB/s]
+ 15%|#5 | 20465/132723 [00:00<00:02, 48704.08KB/s]
+ 21%|## | 27666/132723 [00:00<00:01, 56036.11KB/s]
+ 27%|##6 | 35413/132723 [00:00<00:01, 62680.01KB/s]
+ 32%|###2 | 43039/132723 [00:00<00:01, 66846.97KB/s]
+ 38%|###8 | 50736/132723 [00:00<00:01, 69931.43KB/s]
+ 44%|####3 | 58189/132723 [00:00<00:01, 71323.03KB/s]
+ 50%|####9 | 65867/132723 [00:00<00:00, 72970.52KB/s]
+ 55%|#####5 | 73539/132723 [00:01<00:00, 74099.47KB/s]
+ 61%|######1 | 81224/132723 [00:01<00:00, 74925.78KB/s]
+ 67%|######7 | 88925/132723 [00:01<00:00, 75549.19KB/s]
+ 73%|#######2 | 96548/132723 [00:01<00:00, 75749.53KB/s]
+ 79%|#######8 | 104243/132723 [00:01<00:00, 76105.83KB/s]
+ 84%|########4 | 111971/132723 [00:01<00:00, 76444.11KB/s]
+ 90%|######### | 119631/132723 [00:01<00:00, 76244.37KB/s]
+ 96%|#########5| 127398/132723 [00:01<00:00, 76668.13KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 67147.42KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -496,7 +497,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 38.460 seconds)</p>
+<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 38.295 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-ssd-gluoncv-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/cccb17d28e5e8b2e94ea8cd5ec59f6ed/deploy_ssd_gluoncv.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_ssd_gluoncv.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/sg_execution_times.html b/docs/how_to/deploy_models/sg_execution_times.html
index 279d863cd4..3505eb3c02 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>11:35.860</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>11:27.774</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 86%" />
@@ -336,39 +336,39 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>03:01.844</p></td>
+<td><p>03:02.671</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>02:38.460</p></td>
+<td><p>02:38.295</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></td>
-<td><p>01:59.808</p></td>
+<td><p>01:52.478</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_quantized.html#sphx-glr-how-to-deploy-models-deploy-quantized-py"><span class="std std-ref">Deploy a Quantized Model on Cuda</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_quantized.py</span></code>)</p></td>
-<td><p>01:30.152</p></td>
+<td><p>01:29.408</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized.html#sphx-glr-how-to-deploy-models-deploy-prequantized-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized.py</span></code>)</p></td>
-<td><p>01:09.629</p></td>
+<td><p>01:10.165</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></td>
-<td><p>00:30.042</p></td>
+<td><p>00:30.331</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:23.218</p></td>
+<td><p>00:22.361</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:22.701</p></td>
+<td><p>00:22.059</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></td>
-<td><p>00:00.006</p></td>
+<td><p>00:00.007</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
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 bcc4fa3707..dea88952da 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -608,7 +608,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
<span class="n">module</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">params</span></a> <span class="o">=</span> <span class="n">get_mobilenet</span><span class="p">()</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip84dd76bd-55c6-4606-b57a-a34b08bad007 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.zip7ad6995e-d264-43a5-a3a2-2bac582310c9 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 7e238cff60..eb11b12ae3 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:41.508</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:41.926</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -336,15 +336,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:38.367</p></td>
+<td><p>00:38.707</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="use_pass_instrument.html#sphx-glr-how-to-extend-tvm-use-pass-instrument-py"><span class="std std-ref">How to Use TVM Pass Instrument</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_instrument.py</span></code>)</p></td>
-<td><p>00:02.203</p></td>
+<td><p>00:02.246</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></td>
-<td><p>00:00.930</p></td>
+<td><p>00:00.965</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index e588566595..f39681ef28 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -512,10 +512,10 @@ profile the execution time of each passes.</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 7116us [7116us] (46.84%; 46.84%)
-FoldScaleAxis: 8077us [16us] (53.16%; 53.16%)
- FoldConstant: 8061us [1680us] (53.06%; 99.80%)
- InferType: 6381us [6381us] (42.00%; 79.16%)
+InferType: 7069us [7069us] (46.89%; 46.89%)
+FoldScaleAxis: 8006us [6us] (53.11%; 53.11%)
+ FoldConstant: 8001us [1614us] (53.07%; 99.93%)
+ InferType: 6386us [6386us] (42.36%; 79.82%)
</pre></div>
</div>
</div>
@@ -537,10 +537,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 6447us [6447us] (44.86%; 44.86%)
-FoldScaleAxis: 7924us [6us] (55.14%; 55.14%)
- FoldConstant: 7919us [1661us] (55.10%; 99.93%)
- InferType: 6257us [6257us] (43.54%; 79.02%)
+InferType: 6425us [6425us] (44.86%; 44.86%)
+FoldScaleAxis: 7897us [5us] (55.14%; 55.14%)
+ FoldConstant: 7892us [1624us] (55.10%; 99.94%)
+ InferType: 6268us [6268us] (43.77%; 79.43%)
</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 d624b73531..e4bd8f0a66 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -564,7 +564,7 @@ latency of convolution.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Convolution: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 34.263575 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 51.348464 ms
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-optimize-operators-opt-conv-cuda-py">
diff --git a/docs/how_to/optimize_operators/opt_conv_tensorcore.html b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
index ade46c5c08..e3c5ce8515 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -906,7 +906,7 @@ be able to run on our build server</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"conv2d with tensor core: </span><span class="si">%f</span><span class="s2"> ms"</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">* [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 10.635003 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 6.678476 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 056fde2a63..811d21906e 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -461,8 +461,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
<span class="nb">print</span><span class="p">(</span><span class="s2">"Baseline: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019158
-Baseline: 3.335783
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018788
+Baseline: 3.324655
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -522,7 +522,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt1: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.324760
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.309349
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -589,7 +589,7 @@ vastly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt2: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.349662
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.341092
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -650,7 +650,7 @@ the access pattern for A matrix is more cache friendly.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt3: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.116871
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.116290
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -733,7 +733,7 @@ flattening.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt4: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109945
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109392
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -819,7 +819,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt5: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.110878
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.110807
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -909,7 +909,7 @@ write to C when all the block results are ready.</p>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Opt6: </span><span class="si">%f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.146649
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147347
</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 da7b452933..533f0c8076 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:34.792</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:34.389</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -336,15 +336,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></td>
-<td><p>00:32.497</p></td>
+<td><p>00:32.203</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="opt_conv_tensorcore.html#sphx-glr-how-to-optimize-operators-opt-conv-tensorcore-py"><span class="std std-ref">How to optimize convolution using TensorCores</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_tensorcore.py</span></code>)</p></td>
-<td><p>00:01.285</p></td>
+<td><p>00:01.183</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="opt_conv_cuda.html#sphx-glr-how-to-optimize-operators-opt-conv-cuda-py"><span class="std std-ref">How to optimize convolution on GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_cuda.py</span></code>)</p></td>
-<td><p>00:01.010</p></td>
+<td><p>00:01.004</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
index df12ad8721..51431e6fb4 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>06:22.068</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>06:27.366</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 85%" />
@@ -336,27 +336,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></td>
-<td><p>03:23.108</p></td>
+<td><p>03:31.859</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-x86-py"><span class="std std-ref">Auto-scheduling a Neural Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_x86.py</span></code>)</p></td>
-<td><p>01:23.040</p></td>
+<td><p>01:23.178</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></td>
-<td><p>00:56.668</p></td>
+<td><p>00:56.682</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></td>
-<td><p>00:21.566</p></td>
+<td><p>00:17.940</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></td>
-<td><p>00:08.909</p></td>
+<td><p>00:08.982</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></td>
-<td><p>00:08.778</p></td>
+<td><p>00:08.725</p></td>
<td><p>0.0 MB</p></td>
</tr>
</tbody>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
index a35ee013af..e6930c3f1d 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
@@ -491,12 +491,12 @@ cooperative fetching, unrolling and operator fusion.</p>
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 28;
- allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [72]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [3072]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [8]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [4032]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [6144]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [8], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[2] = 0f32
conv2d_nchw_1[3] = 0f32
@@ -504,470 +504,226 @@ cooperative fetching, unrolling and operator fusion.</p>
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 (rc.outer.outer: int32, 0, 8) {
for (ry.outer.outer: int32, 0, 3) {
- let cse_var_2: int32 = (rc.outer.outer*72)
+ let cse_var_4: int32 = (rc.outer.outer*3136)
+ let cse_var_3: int32 = (ry.outer.outer*7)
+ let cse_var_2: int32 = (rc.outer.outer*576)
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_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [4032], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) && ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3) + floormod(threadIdx.x_1, 9)) - 8)], 0f [...]
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 196), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 392), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 588), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 784), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 980)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 980), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1176)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1176), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1372)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1372), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1568)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1568), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1764)] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) && ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3) + floormod(threadIdx.x_1, 9)) + 1364)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 1960)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 1960), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2156)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2156), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2352)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 21), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 3), 9))) && (floormod((threadIdx.x_1 + 3), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2352), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 3), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2548)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 28), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 1), 9))) && (floormod((threadIdx.x_1 + 1), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2548), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 1), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2744)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2744), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 2940)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 42), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 6), 9))) && (floormod((threadIdx.x_1 + 6), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 2940), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 6), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3136)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 49), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 4), 9))) && (floormod((threadIdx.x_1 + 4), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3136), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 4), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3332)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 56), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 2), 9))) && (floormod((threadIdx.x_1 + 2), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3332), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 2), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3528)] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) && ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3) + floormod(threadIdx.x_1, 9)) + 2736)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ pad_temp.shared_1[(threadIdx.x_1 + 3724)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3724), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ if @tir.likely((threadIdx.x_1 < 112), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 3920)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 14), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 5), 9))) && (floormod((threadIdx.x_1 + 5), 9) < 8)), data[((((cse_var_4 + (floordiv((threadIdx.x_1 + 3920), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 5), 9)) - 8)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1: Buffer(kernel.shared, float32, [6144], [], scope="shared")[threadIdx.x_2] = kernel[((((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 192)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 192), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 196)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 196), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 4), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 392)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 392), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 588)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 588), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 4), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 784)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 784), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 980)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 980), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 20), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1176)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1176), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1372)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1372), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 28), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1568), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 32), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1764)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1764), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 12), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 1960)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 1960), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 40), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2156)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2156), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 44), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2352)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2352), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2548)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2548), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 52), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2744)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2744), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 56), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 2940)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 2940), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 20), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3136), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 64), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3332)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3332), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 68), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3528)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3528), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 24), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3724)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3724), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 76), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 3920)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 3920), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 80), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4116)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4116), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 28), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4312)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4312), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 88), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4508)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4508), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 92), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4704)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4704), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 32), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 4900)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 4900), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 100), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5096)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5096), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 104), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5292)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5292), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 36), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5488)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5488), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 112), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5684)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5684), 192)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 116), 192), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ kernel.shared_1[(threadIdx.x_2 + 5880)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 5880), 192)*4608)) + cse_var_2) + (floormod((floordiv(threadIdx.x_2, 3) + 40), 64)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
+ if @tir.likely((threadIdx.x_2 < 68), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 6076)] = kernel[((((((blockIdx.x*147456) + (floordiv((threadIdx.x_2 + 6076), 192)*4608)) + cse_var_2) + (floordiv((threadIdx.x_2 + 124), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
+ }
+ for (rc.outer.inner: int32, 0, 16) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12))]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 3)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 6)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 9)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 192)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 195)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 198)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 201)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 384)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 387)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 390)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 393)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 576)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 579)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 582)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 585)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 768)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 771)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 774)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 777)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 960)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 963)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 966)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 969)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1152)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1155)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1158)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1161)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1344)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1347)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 126)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1350)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 189)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1353)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 4)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 7)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 10)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 193)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 196)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 199)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 202)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 385)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 388)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 391)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 394)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 577)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 580)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 583)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 586)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 769)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 772)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 775)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 778)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 961)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 964)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 967)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 970)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1153)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1156)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1159)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1162)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1345)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 64)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1348)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 127)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1351)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 190)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1354)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 2)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 5)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 8)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 11)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 194)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 197)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 200)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 203)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 386)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 389)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 392)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 395)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 578)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 581)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 584)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 587)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 770)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 773)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 776)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 779)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 962)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 965)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 968)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 971)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1154)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1157)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1160)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1163)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1346)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 65)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1349)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 128)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1352)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*252) + (floordiv(floormod(threadIdx.x, 49), 7)*9)) + floormod(threadIdx.x, 7)) + 191)]*kernel.shared_1[(((floordiv(threadIdx.x, 49)*1536) + (rc.outer.inner*12)) + 1355)]))
}
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1: Buffer(kernel.shared, float32, [3072], [], scope="shared")[threadIdx.x_2] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 64)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 64), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 128)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 128), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 192)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 36864)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 256)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 256), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 320)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 320), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 384)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 73728)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 448), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 512)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 512), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 576)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 110592)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 640)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 640), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 704)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 704), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 768)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 147456)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 832)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 832), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 896)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 896), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 960)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 184320)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1024)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1024), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1088)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1088), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1152)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 221184)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1216)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1216), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1280)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1280), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1408)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1408), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1472)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1472), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1536)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 294912)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1600)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1600), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1664)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1664), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1728)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 331776)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1792), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1856)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1856), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1920)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 368640)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 1984)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 1984), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2048)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2048), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2112)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 405504)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2176)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2176), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2240), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2304)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 442368)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2368)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2368), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2432)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2432), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2496)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 479232)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2560)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2560), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2624)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2624), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 516096)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2752)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2752), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2816)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2816), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2880)] = kernel[(((((((floordiv(blockIdx.x, 7)*589824) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 552960)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 2944)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 2944), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 64;
- kernel.shared_1[(threadIdx.x_2 + 3008)] = kernel[((((((floordiv(blockIdx.x, 7)*589824) + (floordiv((threadIdx.x_2 + 3008), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[0]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[1]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[2]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[3]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[4]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[5]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[6]*kernel.shared_1[(threadIdx.x*48)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 3)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[0]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[9]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 24)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 27)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 1)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 4)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[1]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[10]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 25)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 28)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 2)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 5)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[2]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[11]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[3]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[12]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[4]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[13]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[5]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[14]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[6]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[15]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[7]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[16]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[8]*kernel.shared_1[((threadIdx.x*48) + 26)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[17]*kernel.shared_1[((threadIdx.x*48) + 29)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 6)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 9)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[18]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[27]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 30)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 33)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 7)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 10)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[19]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[28]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 31)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 34)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 8)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 11)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[20]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[29]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[21]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[30]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[22]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[31]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[23]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[32]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[24]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[33]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[25]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[34]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[26]*kernel.shared_1[((threadIdx.x*48) + 32)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[35]*kernel.shared_1[((threadIdx.x*48) + 35)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 12)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 15)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[36]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[45]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 36)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 39)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 13)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 16)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[37]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[46]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 37)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 40)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 14)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 17)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[38]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[47]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[39]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[48]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[40]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[49]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[41]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[50]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[42]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[51]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[43]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[52]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[44]*kernel.shared_1[((threadIdx.x*48) + 38)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[53]*kernel.shared_1[((threadIdx.x*48) + 41)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 18)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 21)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[54]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[63]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 42)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 45)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 19)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 22)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[55]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[64]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 43)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 46)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 20)]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 23)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[56]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[65]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[57]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[66]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[58]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[67]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[59]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[68]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[60]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[69]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[61]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[70]*kernel.shared_1[((threadIdx.x*48) + 47)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[62]*kernel.shared_1[((threadIdx.x*48) + 44)]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[71]*kernel.shared_1[((threadIdx.x*48) + 47)]))
}
}
}
- for (i1.inner: int32, 0, 2) {
- for (i3.inner: int32, 0, 7) {
- compute[(((((floordiv(blockIdx.x, 7)*6272) + (threadIdx.x*98)) + (i1.inner*49)) + (floormod(blockIdx.x, 7)*7)) + i3.inner)] = max((conv2d_nchw_1[((i1.inner*7) + i3.inner)] + bias[(((floordiv(blockIdx.x, 7)*128) + (threadIdx.x*2)) + i1.inner)]), 0f32)
- }
+ for (i1.inner: int32, 0, 8) {
+ compute[((((blockIdx.x*1568) + (floordiv(threadIdx.x, 49)*392)) + (i1.inner*49)) + floormod(threadIdx.x, 49))] = max((conv2d_nchw_1[i1.inner] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 49)*8)) + i1.inner)]), 0f32)
}
}
}
@@ -1004,7 +760,7 @@ cooperative fetching, unrolling and operator fusion.</p>
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.361 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.270 ms
</pre></div>
</div>
</div>
@@ -1034,19 +790,19 @@ conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
-conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=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_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=8)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=4)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
-conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
+conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
-conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
-conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
+conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
+conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
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)
@@ -1055,14 +811,14 @@ s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nc
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=64)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=8)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
-compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
+compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
-compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
-compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
+compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -1082,12 +838,12 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=4)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=64)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
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)
@@ -1107,10 +863,10 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(64) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[14];
- __shared__ float pad_temp_shared[72];
- __shared__ float kernel_shared[3072];
+extern "C" __global__ void __launch_bounds__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[8];
+ __shared__ float pad_temp_shared[4032];
+ __shared__ float kernel_shared[6144];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
conv2d_nchw[2] = 0.000000e+00f;
@@ -1119,418 +875,169 @@ extern "C" __global__ void __launch_bounds__(64) default_function_kern
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 rc_outer_outer = 0; rc_outer_outer < 8; ++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);
+ pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 196) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 392) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 588)] = (((((1 <= ((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 588) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 784)] = (((((1 <= ((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 784) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 980)] = (((((1 <= ((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 980) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1176)] = (((((1 <= ((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1176) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1372)] = (((((1 <= ((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1372) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1568)] = (((((1 <= ((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1568) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1764)] = (((((1 <= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) + 1364)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 1960)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 1960) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2156)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2156) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 5) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2352)] = (((((1 <= ((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 21) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 3) % 9))) && (((((int)threadIdx.x) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2352) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 3) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2548)] = (((((1 <= ((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 28) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 1) % 9))) && (((((int)threadIdx.x) + 1) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2548) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 1) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2744)] = (((((1 <= ((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2744) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 2940)] = (((((1 <= ((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 42) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 6) % 9))) && (((((int)threadIdx.x) + 6) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 2940) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 6) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3136)] = (((((1 <= ((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 49) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 4) % 9))) && (((((int)threadIdx.x) + 4) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3136) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 4) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3332)] = (((((1 <= ((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 56) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 2) % 9))) && (((((int)threadIdx.x) + 2) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3332) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 2) % 9)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3528)] = (((((1 <= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) + 2736)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 3724)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3724) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 112) {
+ pad_temp_shared[(((int)threadIdx.x) + 3920)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 5) % 9))) && (((((int)threadIdx.x) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 3136) + (((((int)threadIdx.x) + 3920) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 5) % 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);
+ kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 192) * 4608)) + (rc_outer_outer * 576)) + (((((int)threadIdx.x) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 196)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 196) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 4) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 392)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 392) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 8) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 588)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 588) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 4) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 784)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 784) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 16) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 980)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 980) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 20) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1176) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 8) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1372)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1372) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 28) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1568) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 32) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1764)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1764) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 12) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 1960) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 40) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2156)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2156) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 44) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2352) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 16) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2548)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2548) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 52) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2744) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 56) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 2940)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 2940) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 20) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3136) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 64) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3332)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3332) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 68) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3528)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3528) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 24) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3724)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3724) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 76) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 3920)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 3920) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 80) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4116)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4116) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 28) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4312)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4312) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 88) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4508)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4508) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 92) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4704)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4704) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 32) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 4900)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 4900) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 100) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5096)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5096) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 104) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5292)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5292) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 36) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5488)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5488) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 112) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5684)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5684) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) + 116) % 192) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+ kernel_shared[(((int)threadIdx.x) + 5880)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 5880) / 192) * 4608)) + (rc_outer_outer * 576)) + ((((((int)threadIdx.x) / 3) + 40) & 63) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+ if (((int)threadIdx.x) < 68) {
+ kernel_shared[(((int)threadIdx.x) + 6076)] = kernel[((((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 6076) / 192) * 4608)) + (rc_outer_outer * 576)) + (((((int)threadIdx.x) + 124) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
}
- 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)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12))]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 3)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 6)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 9)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 192)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 195)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 198)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 201)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 384)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 387)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 390)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 393)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 576)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 579)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 582)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 585)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 768)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 771)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 774)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 777)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 960)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 963)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 966)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 969)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1152)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1155)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1158)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1161)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1344)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1347)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 126)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1350)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 189)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1353)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 4)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 7)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 10)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 193)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 196)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 199)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 202)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 385)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 388)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 391)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 394)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 577)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 580)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 583)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 586)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 769)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 772)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 775)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 778)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 961)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 964)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 967)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 970)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1153)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1156)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1159)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1162)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1345)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 64)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1348)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 127)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1351)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 190)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1354)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 2)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 5)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 8)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 11)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 194)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 197)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 200)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 203)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 386)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 389)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 392)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 395)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 578)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 581)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 584)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 587)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 770)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 773)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 776)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 779)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 962)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 965)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 968)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 971)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1154)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1157)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1160)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1163)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1346)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 65)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1349)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 128)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1352)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 252) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (((int)threadIdx.x) % 7)) + 191)] * kernel_shared[((((((int)threadIdx.x) / 49) * 1536) + (rc_outer_inner * 12)) + 1355)]));
+ }
}
}
- for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
- for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
- compute[((((((((int)blockIdx.x) / 7) * 6272) + (((int)threadIdx.x) * 98)) + (i1_inner * 49)) + ((((int)blockIdx.x) % 7) * 7)) + i3_inner)] = max((conv2d_nchw[((i1_inner * 7) + i3_inner)] + bias[((((((int)blockIdx.x) / 7) * 128) + (((int)threadIdx.x) * 2)) + i1_inner)]), 0.000000e+00f);
- }
+ for (int i1_inner = 0; i1_inner < 8; ++i1_inner) {
+ compute[((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 49) * 392)) + (i1_inner * 49)) + (((int)threadIdx.x) % 49))] = max((conv2d_nchw[i1_inner] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 49) * 8)) + i1_inner)]), 0.000000e+00f);
}
}
</pre></div>
@@ -1567,7 +1074,7 @@ 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> ( 3 minutes 23.108 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 31.859 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/e3e540f3b477c0c52d8eb73e674e8ffd/tune_conv2d_layer_cuda.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_conv2d_layer_cuda.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
index bcc61a9425..2fc3494e41 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -902,7 +902,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 8.1781 8.1771 8.1828 8.1743 0.0035
+ 8.1711 8.1698 8.1762 8.1674 0.0037
</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 76c49640a7..a6a0e84bc5 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -921,7 +921,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 754.2436 754.3243 754.9236 753.4831 0.5908
+ 757.3703 757.8641 757.8779 756.3690 0.7081
</pre></div>
</div>
</div>
@@ -943,7 +943,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.040 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 23.178 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-x86-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/e416b94ca1090b0897c0f6e0df95b911/tune_network_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_network_x86.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
index 00406033e5..1b1e87a75c 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -625,78 +625,32 @@ 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_8: placeholder_15: Buffer(placeholder_13, int32, [33], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_5: placeholder_18: Buffer(placeholder_10, float32, [128, 256], []), placeholder_7: placeholder_19: Buffer(placeholder_12, int32, [4916], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 64) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global {
+ preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), placeholder_9: placeholder_18: Buffer(placeholder_14, float32, [128, 512], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_5: placeholder_19: Buffer(placeholder_10, float32, [128, 256], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 256) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [512]), storage_scope = global {
for (i.outer.inner: int32, 0, 2) {
- for (nb_j.inner: int32, 0, 2) {
- for (i.inner.init: int32, 0, 16) {
- let cse_var_1: int32 = (((i.outer.inner*512) + (i.inner.init*32)) + (nb_j.inner*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 (i.inner.init: int32, 0, 16) {
+ for (j.init: int32, 0, 16) {
+ compute_5: Buffer(compute_4, float32, [512], [])[(((i.outer.inner*256) + (i.inner.init*16)) + j.init)] = 0f32
}
- for (elem_idx: int32, 0, let cse_var_2: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_2 + 1)] - placeholder_3[cse_var_2])) {
- for (i.inner: int32, 0, 16) {
- let cse_var_21: int32 = (elem_idx*16)
- let cse_var_20: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- let cse_var_19: int32 = (((i.outer.inner*512) + (i.inner*32)) + (nb_j.inner*16))
- let cse_var_18: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i.outer.inner*4096)) + (i.inner*256))
- let cse_var_17: int32 = (cse_var_19 + 9)
- let cse_var_16: int32 = (cse_var_19 + 8)
- let cse_var_15: int32 = (cse_var_19 + 7)
- let cse_var_14: int32 = (cse_var_19 + 6)
- let cse_var_13: int32 = (cse_var_19 + 5)
- let cse_var_12: int32 = (cse_var_19 + 4)
- let cse_var_11: int32 = (cse_var_19 + 3)
- let cse_var_10: int32 = (cse_var_19 + 2)
- let cse_var_9: int32 = (cse_var_19 + 15)
- let cse_var_8: int32 = (cse_var_19 + 14)
- let cse_var_7: int32 = (cse_var_19 + 13)
- let cse_var_6: int32 = (cse_var_19 + 12)
- let cse_var_5: int32 = (cse_var_19 + 11)
- let cse_var_4: int32 = (cse_var_19 + 10)
- let cse_var_3: int32 = (cse_var_19 + 1)
- {
- compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[((placeholder_3[cse_var_20]*16) + cse_var_21)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 1)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 2)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 3)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 4)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 5)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 6)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 7)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 8)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 9)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 10)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 11)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 12)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 13)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 14)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_20]*16) + cse_var_21) + 15)]*max(placeholder[(cse_var_18 + placeholder_2[(placeholder_3[cse_var_20] + elem_idx)])], 0f32)))
- }
+ }
+ for (elem_idx: int32, 0, let cse_var_1: int32 = floordiv(floormod(i0.outer.i1.outer.fused, 64), 2) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+ for (i.inner: int32, 0, 16) {
+ for (j: int32, 0, 16) {
+ let cse_var_3: int32 = floordiv(floormod(i0.outer.i1.outer.fused, 64), 2)
+ let cse_var_2: int32 = (((i.outer.inner*256) + (i.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[((((floordiv(i0.outer.i1.outer.fused, 64)*8192) + (i.outer.inner*4096)) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
}
for (i0.inner: int32, 0, 32) {
- let cse_var_22: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
- compute[ramp(cse_var_22, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_22, 1, 32)]), broadcast(0f32, 32))
+ for (i1.inner: int32, 0, 8) {
+ let cse_var_5: int32 = floormod(i0.outer.i1.outer.fused, 64)
+ let cse_var_6: int32 = (cse_var_5*8)
+ let cse_var_4: int32 = ((((floordiv(i0.outer.i1.outer.fused, 64)*16384) + (i0.inner*512)) + cse_var_6) + i1.inner)
+ compute[cse_var_4] = max((compute_5[((((i0.inner*16) + cse_var_6) + i1.inner) - (floordiv(cse_var_5, 2)*16))] + placeholder_4[cse_var_4]), 0f32)
+ }
}
}
}
@@ -734,7 +688,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
<span class="p">)</span>
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.730 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 3.168 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 44c908c851..20ea8b2808 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:46.571</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:45.184</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -336,11 +336,11 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></td>
-<td><p>00:46.535</p></td>
+<td><p>00:45.146</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></td>
-<td><p>00:00.021</p></td>
+<td><p>00:00.022</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></td>
diff --git a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
index 545202e301..6aa4b2ab09 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -1436,8 +1436,8 @@ No: 8 GFLOPS: 0.00/0.00 result: Traceback (most recent call last):
TimeoutError
[('tile_f', [-1, 2, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4909501
-No: 9 GFLOPS: 221.34/221.34 result: MeasureResult(costs=(0.0010458929724137932,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9837684631347656, timestamp=1663707309.94193) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
-No: 10 GFLOPS: 0.00/221.34 result: Traceback (most recent call last):
+No: 9 GFLOPS: 80.80/80.80 result: MeasureResult(costs=(0.0028652725714285714,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8740947246551514, timestamp=1663712499.211997) [('tile_f', [-1, 1, 4, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5072689
+No: 10 GFLOPS: 0.00/80.80 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1560,8 +1560,8 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5092711
-No: 11 GFLOPS: 260.39/260.39 result: MeasureResult(costs=(0.0008890432960893855,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7833216190338135, timestamp=1663707310.865398) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
-No: 12 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+No: 11 GFLOPS: 260.30/260.30 result: MeasureResult(costs=(0.0008893510837988827,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4758203029632568, timestamp=1663712500.1406338) [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
+No: 12 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1684,7 +1684,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 128, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,183542
-No: 13 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+No: 13 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1807,7 +1807,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2482196
-No: 14 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+No: 14 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1930,9 +1930,9 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10306226
-No: 15 GFLOPS: 5.43/260.39 result: MeasureResult(costs=(0.042630027499999994,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8400635719299316, timestamp=1663707315.4601128) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
-No: 16 GFLOPS: 3.33/260.39 result: MeasureResult(costs=(0.0694302805,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.599075078964233, timestamp=1663707316.6981018) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
-No: 17 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+No: 15 GFLOPS: 5.29/260.30 result: MeasureResult(costs=(0.04375817275,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8518753051757812, timestamp=1663712504.731286) [('tile_f', [-1, 2, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5330964
+No: 16 GFLOPS: 3.33/260.30 result: MeasureResult(costs=(0.069439056,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.605278015136719, timestamp=1663712505.966001) [('tile_f', [-1, 8, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2140058
+No: 17 GFLOPS: 0.00/260.30 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
@@ -1950,8 +1950,8 @@ No: 17 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
TimeoutError
[('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10195251
-No: 18 GFLOPS: 27.41/260.39 result: MeasureResult(costs=(0.008445743117647057,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3354072570800781, timestamp=1663707327.7962105) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
-No: 19 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+No: 18 GFLOPS: 28.40/260.30 result: MeasureResult(costs=(0.008150162214285715,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2855777740478516, timestamp=1663712517.0024183) [('tile_f', [-1, 4, 8, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6068603
+No: 19 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2074,7 +2074,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 871, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 16, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6956993
-No: 20 GFLOPS: 0.00/260.39 result: Traceback (most recent call last):
+No: 20 GFLOPS: 0.00/260.30 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2237,7 +2237,7 @@ and measure running time.</p>
Best config:
[('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4264713
Finish loading 20 records
-Time cost of this operator: 0.001283
+Time cost of this operator: 0.001228
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autotvm-tune-conv2d-cuda-py">
diff --git a/docs/how_to/work_with_microtvm/micro_autotune.html b/docs/how_to/work_with_microtvm/micro_autotune.html
index 447124c2a3..9305d92873 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -582,10 +582,10 @@ the tuned operator.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.5 98.728 (1, 2, 10, 10, 3) 2 1 [312.5]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.074 0.971 (1, 6, 10, 10) 1 1 [3.074]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.954 0.301 (1, 1, 10, 10, 3) 1 1 [0.954]
-Total_time - 316.528 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 313.0 98.677 (1, 2, 10, 10, 3) 2 1 [313.0]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.216 1.014 (1, 6, 10, 10) 1 1 [3.216]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.981 0.309 (1, 1, 10, 10, 3) 1 1 [0.981]
+Total_time - 317.197 - - - - -
</pre></div>
</div>
</div>
@@ -636,10 +636,10 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs Measurements(us)
--------- --- -------- ------- ----- ------ ------- ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 119.8 97.752 (1, 6, 10, 10, 1) 2 1 [119.8]
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.783 1.455 (1, 6, 10, 10) 1 1 [1.783]
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.973 0.794 (1, 1, 10, 10, 3) 1 1 [0.973]
-Total_time - 122.555 - - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 122.6 97.796 (1, 6, 10, 10, 1) 2 1 [122.6]
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.777 1.417 (1, 6, 10, 10) 1 1 [1.777]
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.986 0.787 (1, 1, 10, 10, 3) 1 1 [0.986]
+Total_time - 125.363 - - - - -
</pre></div>
</div>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index c76a86375b..f36b594fb8 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -516,7 +516,7 @@ take about <strong>2 minutes</strong> to download the Stanford Cars, while COCO
<a href="https://docs.python.org/3/library/shutil.html#shutil.move" title="shutil.move" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">move</span></a><span class="p">(</span><span class="sa">f</span><span class="s2">"</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-typ [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmplsnuy6r_/images/random'
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>'/tmp/tmp5yhrl4fv/images/random'
</pre></div>
</div>
</div>
@@ -576,8 +576,8 @@ objects to other stuff? We can display some examples from our datasets using <co
<span class="n">plt</span><span class="o">.</span><span class="n">axis</span><span class="p">(</span><span class="s2">"off"</span><span class="p">)</span>
</pre></div>
</div>
-<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmplsnuy6r_/images/target contains 8144 images
-/tmp/tmplsnuy6r_/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp5yhrl4fv/images/target contains 8144 images
+/tmp/tmp5yhrl4fv/images/random contains 5000 images
</pre></div>
</div>
</div>
@@ -689,13 +689,13 @@ the time on our validation set).</p>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 47s - loss: 0.2198 - accuracy: 0.9264 - val_loss: 0.1292 - val_accuracy: 0.9573 - 47s/epoch - 143ms/step
+328/328 - 46s - loss: 0.2262 - accuracy: 0.9234 - val_loss: 0.1417 - val_accuracy: 0.9592 - 46s/epoch - 142ms/step
Epoch 2/3
-328/328 - 43s - loss: 0.1002 - accuracy: 0.9637 - val_loss: 0.1191 - val_accuracy: 0.9619 - 43s/epoch - 133ms/step
+328/328 - 43s - loss: 0.1036 - accuracy: 0.9615 - val_loss: 0.1216 - val_accuracy: 0.9603 - 43s/epoch - 132ms/step
Epoch 3/3
-328/328 - 43s - loss: 0.0696 - accuracy: 0.9729 - val_loss: 0.0979 - val_accuracy: 0.9687 - 43s/epoch - 132ms/step
+328/328 - 43s - loss: 0.0675 - accuracy: 0.9731 - val_loss: 0.1119 - val_accuracy: 0.9694 - 43s/epoch - 132ms/step
-<keras.callbacks.History object at 0x7f0c039f4d10>
+<keras.callbacks.History object at 0x7ff418e26f90>
</pre></div>
</div>
</div>
@@ -957,7 +957,7 @@ as intended.</p>
<p>From here, we could modify the model to read live images from the camera - we have another
Arduino tutorial for how to do that <a class="reference external" href="https://github.com/guberti/tvm-arduino-demos/tree/master/examples/person_detection">on GitHub</a>. Alternatively, we could also
<a class="reference external" href="https://tvm.apache.org/docs/how_to/work_with_microtvm/micro_autotune.html">use TVM’s autotuning capabilities</a> to dramatically improve the model’s performance.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 41.615 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes 44.758 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-train-py">
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/b52cec46baf4f78d6bcd94cbe269c8a6/micro_train.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_train.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/sg_execution_times.html b/docs/how_to/work_with_microtvm/sg_execution_times.html
index 2e3bb7e8dd..b804128d3f 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:36.478</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>05:38.521</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -336,19 +336,19 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>04:41.615</p></td>
+<td><p>04:44.758</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>00:43.160</p></td>
+<td><p>00:42.562</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_aot.html#sphx-glr-how-to-work-with-microtvm-micro-aot-py"><span class="std std-ref">microTVM Host-Driven AoT</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:08.329</p></td>
+<td><p>00:07.931</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:03.371</p></td>
+<td><p>00:03.268</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index 569d31600d..14f9b6273f 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:43.951</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:41.004</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 84%" />
@@ -336,15 +336,15 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:31.856</p></td>
+<td><p>00:31.108</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></td>
-<td><p>00:10.101</p></td>
+<td><p>00:08.643</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></td>
-<td><p>00:01.987</p></td>
+<td><p>00:01.245</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index 594e680348..6a65f2374b 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -522,7 +522,7 @@ The following example customizes CUDA lowering rule for <code class="code docuti
<a href="../../reference/api/python/ir.html#tvm.ir.register_intrin_lowering" title="tvm.ir.register_intrin_lowering" class="sphx-glr-backref-module-tvm-ir sphx-glr-backref-type-py-function"><span class="n">register_intrin_lowering</span></a><span class="p">(</span><span class="s2">"tir.exp"</span><span class="p">,</span> <span class="n">target</span><span class="o">=</span><span class="s2">"cuda"</span><span class="p">,</span> <span class="n">f</span><span class="o">= [...]
</pre></div>
</div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7f0b869dee60>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><function my_cuda_math_rule at 0x7ff3ab1d6710>
</pre></div>
</div>
<p>Register the rule to TVM with override option to override existing rule.
diff --git a/docs/how_to/work_with_schedules/sg_execution_times.html b/docs/how_to/work_with_schedules/sg_execution_times.html
index c31ae67c4f..f271820da8 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -327,7 +327,7 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:07.975</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:04.188</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 83%" />
@@ -336,27 +336,27 @@
</colgroup>
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></td>
-<td><p>00:05.671</p></td>
+<td><p>00:02.060</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></td>
-<td><p>00:01.052</p></td>
+<td><p>00:00.949</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></td>
-<td><p>00:00.547</p></td>
+<td><p>00:00.509</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></td>
-<td><p>00:00.526</p></td>
+<td><p>00:00.489</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></td>
-<td><p>00:00.099</p></td>
+<td><p>00:00.101</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.039</p></td>
+<td><p>00:00.040</p></td>
<td><p>0.0 MB</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index 692ecbbe96..01a98ac8df 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -577,7 +577,7 @@ 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/tmptk9nsssc/input0.cc'\nsource_filename = \"/tmp/tmptk9nsssc/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/tmp_tj6totc/input0.cc'\nsource_filename = \"/tmp/tmp_tj6totc/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index aa2238b85b..3153785d75 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -224,17 +224,7 @@
<p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal" href="index.html">Installing TVM</a><ul class="current">
-<li class="toctree-l2 current"><a class="reference internal" href="from_source.html">Install from Source</a><ul class="current">
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#developers-get-source-from-github">Developers: Get Source from Github</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#build-the-shared-library">Build the Shared Library</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#python-package-installation">Python Package Installation</a></li>
-<li class="toctree-l3 current"><a class="reference internal" href="from_source.html#install-contrib-libraries">Install Contrib Libraries</a><ul class="current">
-<li class="toctree-l4 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a></li>
-</ul>
-</li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#enable-c-tests">Enable C++ Tests</a></li>
-</ul>
-</li>
+<li class="toctree-l2"><a class="reference internal" href="from_source.html">Install from Source</a></li>
<li class="toctree-l2"><a class="reference internal" href="docker.html">Docker Images</a></li>
<li class="toctree-l2 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#conditions">Conditions</a></li>
diff --git a/docs/reference/api/doxygen/affine__type_8h.html b/docs/reference/api/doxygen/affine__type_8h.html
index c51f064f4d..7c2e09615b 100644
--- a/docs/reference/api/doxygen/affine__type_8h.html
+++ b/docs/reference/api/doxygen/affine__type_8h.html
@@ -77,7 +77,7 @@ $(function() {
</div><div class="textblock"><div class="dynheader">
Include dependency graph for affine_type.h:</div>
<div class="dyncontent">
-<div class="center"><iframe scrolling="no" frameborder="0" src="affine__type_8h__incl.svg" width="4043" height="1082"><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="affine__type_8h__incl.svg" width="4286" height="1082"><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/affine__type_8h__incl.svg b/docs/reference/api/doxygen/affine__type_8h__incl.svg
index 52439ff7c6..c4977735bd 100644
--- a/docs/reference/api/doxygen/affine__type_8h__incl.svg
+++ b/docs/reference/api/doxygen/affine__type_8h__incl.svg
@@ -4,1352 +4,1364 @@
<!-- Generated by graphviz version 2.40.1 (20161225.0304)
-->
<!-- Title: include/tvm/ir/affine_type.h Pages: 1 -->
-<svg width="3032pt" height="811pt"
- viewBox="0.00 0.00 3032.00 811.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+<svg width="3214pt" height="811pt"
+ viewBox="0.00 0.00 3214.00 811.00" 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(4 807)">
<title>include/tvm/ir/affine_type.h</title>
-<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-807 3028,-807 3028,4 -4,4"/>
+<polygon fill="#ffffff" stroke="transparent" points="-4,4 -4,-807 3210,-807 3210,4 -4,4"/>
<!-- Node0 -->
<g id="node1" class="node">
<title>Node0</title>
-<polygon fill="#bfbfbf" stroke="#000000" points="1525.5,-772.5 1525.5,-802.5 1640.5,-802.5 1640.5,-772.5 1525.5,-772.5"/>
-<text text-anchor="start" x="1533.5" y="-790.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">include/tvm/ir/affine</text>
-<text text-anchor="middle" x="1583" y="-779.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_type.h</text>
+<polygon fill="#bfbfbf" stroke="#000000" points="299.5,-772.5 299.5,-802.5 414.5,-802.5 414.5,-772.5 299.5,-772.5"/>
+<text text-anchor="start" x="307.5" y="-790.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">include/tvm/ir/affine</text>
+<text text-anchor="middle" x="357" y="-779.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_type.h</text>
</g>
<!-- Node1 -->
<g id="node2" class="node">
<title>Node1</title>
<g id="a_node2"><a xlink:href="ir_2expr_8h.html" target="_top" xlink:title="Base expr nodes in TVM. ">
-<polygon fill="#ffffff" stroke="#000000" points="1509.5,-716.5 1509.5,-735.5 1588.5,-735.5 1588.5,-716.5 1509.5,-716.5"/>
-<text text-anchor="middle" x="1549" y="-723.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/expr.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="283.5,-716.5 283.5,-735.5 362.5,-735.5 362.5,-716.5 283.5,-716.5"/>
+<text text-anchor="middle" x="323" y="-723.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/expr.h</text>
</a>
</g>
</g>
<!-- Node0->Node1 -->
<g id="edge1" class="edge">
<title>Node0->Node1</title>
-<path fill="none" stroke="#191970" d="M1574.5955,-772.2977C1569.9743,-763.9388 1564.2031,-753.4997 1559.3111,-744.6509"/>
-<polygon fill="#191970" stroke="#191970" points="1562.3452,-742.9051 1554.4438,-735.8469 1556.2191,-746.292 1562.3452,-742.9051"/>
+<path fill="none" stroke="#191970" d="M348.5955,-772.2977C343.9743,-763.9388 338.2031,-753.4997 333.3111,-744.6509"/>
+<polygon fill="#191970" stroke="#191970" points="336.3452,-742.9051 328.4438,-735.8469 330.2191,-746.292 336.3452,-742.9051"/>
</g>
<!-- Node49 -->
<g id="node50" class="node">
<title>Node49</title>
<g id="a_node50"><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="2397,-660.5 2397,-679.5 2477,-679.5 2477,-660.5 2397,-660.5"/>
-<text text-anchor="middle" x="2437" y="-667.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="374,-660.5 374,-679.5 454,-679.5 454,-660.5 374,-660.5"/>
+<text text-anchor="middle" x="414" y="-667.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
</a>
</g>
</g>
<!-- Node0->Node49 -->
-<g id="edge160" class="edge">
+<g id="edge162" class="edge">
<title>Node0->Node49</title>
-<path fill="none" stroke="#191970" d="M1640.693,-779.5621C1798.769,-757.8128 2236.9838,-697.5198 2386.3597,-676.9675"/>
-<polygon fill="#191970" stroke="#191970" points="2387.1782,-680.3879 2396.6078,-675.5575 2386.224,-673.4533 2387.1782,-680.3879"/>
+<path fill="none" stroke="#191970" d="M364.4518,-772.1389C374.8613,-750.6806 393.7353,-711.7738 404.9429,-688.6702"/>
+<polygon fill="#191970" stroke="#191970" points="408.147,-690.0844 409.3626,-679.5595 401.8489,-687.0291 408.147,-690.0844"/>
</g>
<!-- Node2 -->
<g id="node3" class="node">
<title>Node2</title>
<g id="a_node3"><a xlink:href="ir_2span_8h.html" target="_top" xlink:title="Span information for debugging purposes. ">
-<polygon fill="#ffffff" stroke="#000000" points="2711.5,-604.5 2711.5,-623.5 2792.5,-623.5 2792.5,-604.5 2711.5,-604.5"/>
-<text text-anchor="middle" x="2752" y="-611.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/span.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="194.5,-604.5 194.5,-623.5 275.5,-623.5 275.5,-604.5 194.5,-604.5"/>
+<text text-anchor="middle" x="235" y="-611.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/span.h</text>
</a>
</g>
</g>
<!-- Node1->Node2 -->
<g id="edge2" class="edge">
<title>Node1->Node2</title>
-<path fill="none" stroke="#191970" d="M1588.5975,-725.0809C1743.5093,-721.3265 2308.6523,-706.1021 2486,-680 2567.4228,-668.0162 2660.2506,-642.0963 2711.9008,-626.5432"/>
-<polygon fill="#191970" stroke="#191970" points="2713.1145,-629.8325 2721.6662,-623.578 2711.0807,-623.1345 2713.1145,-629.8325"/>
+<path fill="none" stroke="#191970" d="M315.4972,-716.4509C300.6899,-697.6054 267.5167,-655.3849 248.7043,-631.4419"/>
+<polygon fill="#191970" stroke="#191970" points="251.4142,-629.2257 242.4838,-623.5249 245.91,-633.5505 251.4142,-629.2257"/>
</g>
<!-- Node3 -->
<g id="node4" class="node">
<title>Node3</title>
<g id="a_node4"><a xlink:href="node_8h.html" target="_top" xlink:title="Definitions and helper macros for IR/AST nodes. ">
-<polygon fill="#ffffff" stroke="#000000" points="2222.5,-548.5 2222.5,-567.5 2321.5,-567.5 2321.5,-548.5 2222.5,-548.5"/>
-<text text-anchor="middle" x="2272" y="-555.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/node.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1843.5,-548.5 1843.5,-567.5 1942.5,-567.5 1942.5,-548.5 1843.5,-548.5"/>
+<text text-anchor="middle" x="1893" y="-555.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/node.h</text>
</a>
</g>
</g>
<!-- Node1->Node3 -->
-<g id="edge153" class="edge">
+<g id="edge155" class="edge">
<title>Node1->Node3</title>
-<path fill="none" stroke="#191970" d="M1588.6429,-716.7884C1711.2484,-688.2991 2084.2435,-601.6281 2221.0203,-569.8459"/>
-<polygon fill="#191970" stroke="#191970" points="2221.8891,-573.2374 2230.8374,-567.5648 2220.3047,-566.419 2221.8891,-573.2374"/>
+<path fill="none" stroke="#191970" d="M362.6345,-725.6977C567.3637,-723.7643 1506.1856,-710.2621 1782,-624 1816.6379,-613.1669 1851.8526,-589.6448 1873.0652,-573.8123"/>
+<polygon fill="#191970" stroke="#191970" points="1875.4331,-576.408 1881.2566,-567.5571 1871.1847,-570.8446 1875.4331,-576.408"/>
</g>
<!-- Node8 -->
<g id="node9" class="node">
<title>Node8</title>
<g id="a_node9"><a xlink:href="object_8h.html" target="_top" xlink:title="A managed object in the TVM runtime. ">
-<polygon fill="#ffffff" stroke="#000000" points="1350.5,-123.5 1350.5,-142.5 1469.5,-142.5 1469.5,-123.5 1350.5,-123.5"/>
-<text text-anchor="middle" x="1410" 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="1738.5,-123.5 1738.5,-142.5 1857.5,-142.5 1857.5,-123.5 1738.5,-123.5"/>
+<text text-anchor="middle" x="1798" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/object.h</text>
</a>
</g>
</g>
<!-- Node1->Node8 -->
-<g id="edge155" class="edge">
+<g id="edge157" class="edge">
<title>Node1->Node8</title>
-<path fill="none" stroke="#191970" d="M1509.2205,-725.1433C1338.4362,-720.9488 676,-699.1886 676,-614 676,-614 676,-614 676,-502 676,-405.3346 751.268,-398.5869 839,-358 940.8829,-310.8665 984.4019,-362.7544 1089,-322 1141.4358,-301.5695 1248.4762,-205.7266 1298,-179 1322.0099,-166.0425 1350.4635,-154.3996 1372.6831,-146.069"/>
-<polygon fill="#191970" stroke="#191970" points="1373.9321,-149.3389 1382.1032,-142.5947 1371.5098,-142.7713 1373.9321,-149.3389"/>
+<path fill="none" stroke="#191970" d="M298.7478,-716.4281C265.9481,-701.9927 208.8146,-671.2157 185,-624 162.2513,-578.8976 47.833,-741.1189 214,-425 286.5617,-286.9573 362.3448,-288.0401 509,-235 599.8366,-202.1476 623.5131,-193.5882 719,-179 914.7113,-149.0999 1521.8109,-137.2272 1728.3701,-133.9866"/>
+<polygon fill="#191970" stroke="#191970" points="1728.5405,-137.4845 1738.485,-133.8298 1728.432,-130.4853 1728.5405,-137.4845"/>
</g>
<!-- Node14 -->
<g id="node15" class="node">
<title>Node14</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2470,-62 2470,-81 2514,-81 2514,-62 2470,-62"/>
-<text text-anchor="middle" x="2492" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="206,-62 206,-81 250,-81 250,-62 206,-62"/>
+<text text-anchor="middle" x="228" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
</g>
<!-- Node1->Node14 -->
-<g id="edge158" class="edge">
+<g id="edge160" class="edge">
<title>Node1->Node14</title>
-<path fill="none" stroke="#191970" d="M1588.6158,-724.957C1816.285,-718.7097 2948,-684.1281 2948,-614 2948,-614 2948,-614 2948,-245 2948,-200.5978 2909.5501,-204.212 2873,-179 2829.1409,-148.7462 2817.3094,-140.5465 2767,-123 2682.2576,-93.4442 2576.5848,-79.7245 2524.3405,-74.3779"/>
-<polygon fill="#191970" stroke="#191970" points="2524.6231,-70.8889 2514.3277,-73.3897 2523.9355,-77.8551 2524.6231,-70.8889"/>
+<path fill="none" stroke="#191970" d="M283.3939,-720.7868C197.1436,-708.3258 0,-673.3519 0,-614 0,-614 0,-614 0,-189 0,-102.0732 132.5702,-79.3944 195.5337,-73.5253"/>
+<polygon fill="#191970" stroke="#191970" points="196.0527,-76.994 205.7233,-72.6663 195.4646,-70.0188 196.0527,-76.994"/>
</g>
<!-- Node15 -->
<g id="node16" class="node">
<title>Node15</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2761.5,-62 2761.5,-81 2830.5,-81 2830.5,-62 2761.5,-62"/>
-<text text-anchor="middle" x="2796" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2801.5,-62 2801.5,-81 2870.5,-81 2870.5,-62 2801.5,-62"/>
+<text text-anchor="middle" x="2836" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
</g>
<!-- Node1->Node15 -->
-<g id="edge159" class="edge">
+<g id="edge161" class="edge">
<title>Node1->Node15</title>
-<path fill="none" stroke="#191970" d="M1588.5174,-725.148C1761.8018,-721.294 2455.459,-704.6053 2672,-680 2830.153,-662.0293 3024,-773.1707 3024,-614 3024,-614 3024,-614 3024,-189 3024,-107.4203 2907.2378,-82.4276 2840.8286,-74.8102"/>
-<polygon fill="#191970" stroke="#191970" points="2840.9411,-71.3027 2830.6283,-73.7309 2840.2044,-78.2638 2840.9411,-71.3027"/>
+<path fill="none" stroke="#191970" d="M362.6578,-725.5351C708.9311,-721.3503 3206,-688.5869 3206,-614 3206,-614 3206,-614 3206,-189 3206,-151.6561 3186.1823,-141.9437 3154,-123 3108.5431,-96.2423 2957.4269,-80.9561 2881.0362,-74.7635"/>
+<polygon fill="#191970" stroke="#191970" points="2881.0171,-71.2511 2870.7718,-73.9498 2880.4639,-78.2292 2881.0171,-71.2511"/>
</g>
<!-- Node24 -->
<g id="node25" class="node">
<title>Node24</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="222,-179.5 222,-198.5 286,-198.5 286,-179.5 222,-179.5"/>
-<text text-anchor="middle" x="254" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1370,-179.5 1370,-198.5 1434,-198.5 1434,-179.5 1370,-179.5"/>
+<text text-anchor="middle" x="1402" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
</g>
<!-- Node1->Node24 -->
-<g id="edge156" class="edge">
+<g id="edge158" class="edge">
<title>Node1->Node24</title>
-<path fill="none" stroke="#191970" d="M1509.1533,-725.2628C1266.6991,-720.5224 0,-692.0229 0,-614 0,-614 0,-614 0,-373.5 0,-269.9115 140.0031,-218.441 212.126,-198.7985"/>
-<polygon fill="#191970" stroke="#191970" points="213.0557,-202.173 221.8307,-196.236 211.2685,-195.405 213.0557,-202.173"/>
+<path fill="none" stroke="#191970" d="M324.7319,-716.3665C334.7782,-663.4939 390.6498,-406.7354 547,-291 645.5031,-218.0849 695.6708,-252.2823 817,-235 1018.9614,-206.2324 1262.8994,-194.3243 1359.578,-190.5018"/>
+<polygon fill="#191970" stroke="#191970" points="1359.9627,-193.9896 1369.8199,-190.1056 1359.692,-186.9949 1359.9627,-193.9896"/>
</g>
<!-- Node26 -->
<g id="node27" class="node">
<title>Node26</title>
<g id="a_node27"><a xlink:href="string_8h.html" target="_top" xlink:title="Runtime String container types. ">
-<polygon fill="#ffffff" stroke="#000000" points="954,-291.5 954,-321.5 1080,-321.5 1080,-291.5 954,-291.5"/>
-<text text-anchor="start" x="962" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="1017" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/string.h</text>
+<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">/string.h</text>
</a>
</g>
</g>
<!-- Node1->Node26 -->
-<g id="edge154" class="edge">
+<g id="edge156" class="edge">
<title>Node1->Node26</title>
-<path fill="none" stroke="#191970" d="M1509.4333,-725.1717C1347.8991,-721.6241 743.118,-706.6651 662,-680 611.4113,-663.3705 562,-667.2518 562,-614 562,-614 562,-614 562,-440.5 562,-320.3399 871.3503,-327.1004 943.5165,-321.521"/>
-<polygon fill="#191970" stroke="#191970" points="944.2896,-324.9559 953.832,-320.3524 943.5015,-318.0004 944.2896,-324.9559"/>
+<path fill="none" stroke="#191970" d="M327.557,-716.1101C334.2706,-702.3182 347.8938,-677.0829 365,-660 542.2805,-482.9605 604.4601,-440.5848 841,-358 914.5587,-332.3179 1133.8651,-316.8449 1249.6023,-310.2483"/>
+<polygon fill="#191970" stroke="#191970" points="1250.004,-313.7313 1259.7916,-309.6752 1249.6108,-306.7424 1250.004,-313.7313"/>
</g>
<!-- Node45 -->
<g id="node46" class="node">
<title>Node45</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="848,-364 848,-383 892,-383 892,-364 848,-364"/>
-<text text-anchor="middle" x="870" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="370,-364 370,-383 414,-383 414,-364 370,-364"/>
+<text text-anchor="middle" x="392" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
</g>
<!-- Node1->Node45 -->
-<g id="edge157" class="edge">
+<g id="edge159" class="edge">
<title>Node1->Node45</title>
-<path fill="none" stroke="#191970" d="M1509.2615,-724.372C1353.0916,-717.4553 790,-687.4253 790,-614 790,-614 790,-614 790,-502 790,-455.9602 827.4675,-412.7534 851.1203,-390.0646"/>
-<polygon fill="#191970" stroke="#191970" points="853.5878,-392.5496 858.538,-383.1823 848.8267,-387.4182 853.5878,-392.5496"/>
+<path fill="none" stroke="#191970" d="M283.3696,-722.698C244.1486,-717.9814 183.5414,-706.694 138,-680 103.2788,-659.6482 76,-654.2462 76,-614 76,-614 76,-614 76,-502 76,-441.212 279.1955,-395.2857 359.7066,-379.4745"/>
+<polygon fill="#191970" stroke="#191970" points="360.6251,-382.8618 369.7795,-377.5283 359.2971,-375.9889 360.6251,-382.8618"/>
</g>
<!-- Node1->Node49 -->
-<g id="edge146" class="edge">
+<g id="edge148" class="edge">
<title>Node1->Node49</title>
-<path fill="none" stroke="#191970" d="M1588.5909,-723.5033C1732.4288,-714.4324 2227.2514,-683.2274 2386.909,-673.1589"/>
-<polygon fill="#191970" stroke="#191970" points="2387.1735,-676.6493 2396.9334,-672.5267 2386.7329,-669.6631 2387.1735,-676.6493"/>
+<path fill="none" stroke="#191970" d="M338.851,-716.2455C352.8132,-707.6534 373.3728,-695.0014 389.4596,-685.1018"/>
+<polygon fill="#191970" stroke="#191970" points="391.6484,-687.8645 398.3307,-679.6427 387.9797,-681.9029 391.6484,-687.8645"/>
</g>
<!-- Node2->Node3 -->
<g id="edge3" class="edge">
<title>Node2->Node3</title>
-<path fill="none" stroke="#191970" d="M2711.4892,-609.2737C2626.9004,-599.405 2431.1499,-576.5675 2331.9458,-564.9937"/>
-<polygon fill="#191970" stroke="#191970" points="2332.1238,-561.4908 2321.7855,-563.8083 2331.3125,-568.4436 2332.1238,-561.4908"/>
+<path fill="none" stroke="#191970" d="M275.5728,-612.6296C497.8145,-605.1233 1567.2462,-569.0025 1833.0795,-560.0239"/>
+<polygon fill="#191970" stroke="#191970" points="1833.4235,-563.5143 1843.2996,-559.6787 1833.1872,-556.5183 1833.4235,-563.5143"/>
</g>
<!-- Node2->Node8 -->
-<g id="edge144" class="edge">
+<g id="edge146" class="edge">
<title>Node2->Node8</title>
-<path fill="none" stroke="#191970" d="M2747.2998,-604.4593C2729.6931,-569.2703 2664.7634,-444.3316 2589,-358 2524.3602,-284.3437 2498.4507,-267.5026 2406,-235 2273.5045,-188.419 2234.3909,-196.176 2095,-179 1871.2043,-151.4235 1604.0909,-139.4836 1479.8991,-135.1294"/>
-<polygon fill="#191970" stroke="#191970" points="1479.8829,-131.6269 1469.7681,-134.78 1479.6415,-138.6227 1479.8829,-131.6269"/>
+<path fill="none" stroke="#191970" d="M233.9971,-604.4242C232.0615,-585.0891 228,-539.9907 228,-502 228,-502 228,-502 228,-440.5 228,-415.0679 479.9309,-262.6202 556,-235 774.1827,-155.7793 1499.1937,-137.7012 1728.1069,-133.925"/>
+<polygon fill="#191970" stroke="#191970" points="1728.364,-137.4214 1738.3064,-133.7608 1728.2513,-130.4223 1728.364,-137.4214"/>
</g>
<!-- Node2->Node14 -->
-<g id="edge145" class="edge">
+<g id="edge147" class="edge">
<title>Node2->Node14</title>
-<path fill="none" stroke="#191970" d="M2775.9954,-604.4657C2811.0144,-588.7839 2872,-554.0082 2872,-502 2872,-502 2872,-502 2872,-245 2872,-168.7085 2615.3387,-100.63 2523.8862,-78.794"/>
-<polygon fill="#191970" stroke="#191970" points="2524.6095,-75.3687 2514.0729,-76.4776 2523.0013,-82.1814 2524.6095,-75.3687"/>
+<path fill="none" stroke="#191970" d="M207.1925,-604.4384C152.8894,-583.7971 38,-529.5983 38,-440.5 38,-440.5 38,-440.5 38,-189 38,-157.7875 39.4067,-144.5352 62,-123 81.3711,-104.5361 153.5905,-86.9348 196.1246,-77.8647"/>
+<polygon fill="#191970" stroke="#191970" points="196.923,-81.2735 205.994,-75.7995 195.4891,-74.4219 196.923,-81.2735"/>
</g>
<!-- Node4 -->
<g id="node5" class="node">
<title>Node4</title>
<g id="a_node5"><a xlink:href="reflection_8h.html" target="_top" xlink:title="Reflection and serialization of compiler IR/AST nodes. ">
-<polygon fill="#ffffff" stroke="#000000" points="1989.5,-492.5 1989.5,-511.5 2110.5,-511.5 2110.5,-492.5 1989.5,-492.5"/>
-<text text-anchor="middle" x="2050" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/reflection.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1788.5,-492.5 1788.5,-511.5 1909.5,-511.5 1909.5,-492.5 1788.5,-492.5"/>
+<text text-anchor="middle" x="1849" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/reflection.h</text>
</a>
</g>
</g>
<!-- Node3->Node4 -->
<g id="edge4" class="edge">
<title>Node3->Node4</title>
-<path fill="none" stroke="#191970" d="M2234.0891,-548.4369C2196.335,-538.9133 2138.2735,-524.2672 2097.4736,-513.9753"/>
-<polygon fill="#191970" stroke="#191970" points="2098.3069,-510.576 2087.7546,-511.5237 2096.5947,-517.3633 2098.3069,-510.576"/>
+<path fill="none" stroke="#191970" d="M1885.3358,-548.2455C1879.1527,-540.3761 1870.2937,-529.101 1862.8711,-519.6542"/>
+<polygon fill="#191970" stroke="#191970" points="1865.5068,-517.3435 1856.5764,-511.6427 1860.0025,-521.6683 1865.5068,-517.3435"/>
</g>
<!-- Node5 -->
<g id="node6" class="node">
<title>Node5</title>
<g id="a_node6"><a xlink:href="structural__equal_8h.html" target="_top" xlink:title="Structural equality comparison. ">
-<polygon fill="#ffffff" stroke="#000000" points="1597.5,-425.5 1597.5,-455.5 1710.5,-455.5 1710.5,-425.5 1597.5,-425.5"/>
-<text text-anchor="start" x="1605.5" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
-<text text-anchor="middle" x="1654" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_equal.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1277.5,-425.5 1277.5,-455.5 1390.5,-455.5 1390.5,-425.5 1277.5,-425.5"/>
+<text text-anchor="start" x="1285.5" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
+<text text-anchor="middle" x="1334" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_equal.h</text>
</a>
</g>
</g>
<!-- Node3->Node5 -->
-<g id="edge135" class="edge">
+<g id="edge137" class="edge">
<title>Node3->Node5</title>
-<path fill="none" stroke="#191970" d="M2222.4186,-551.0641C2164.3801,-542.7368 2065.4464,-527.8927 1981,-512 1890.284,-494.9274 1786.2469,-471.5134 1720.3465,-456.2007"/>
-<polygon fill="#191970" stroke="#191970" points="1721.0394,-452.7685 1710.506,-453.9085 1719.4512,-459.586 1721.0394,-452.7685"/>
+<path fill="none" stroke="#191970" d="M1844.6243,-548.4362C1757.8784,-531.1896 1569.9211,-493.3797 1400.3565,-456.0475"/>
+<polygon fill="#191970" stroke="#191970" points="1401.0223,-452.6103 1390.5031,-453.8741 1399.5144,-459.446 1401.0223,-452.6103"/>
</g>
<!-- Node3->Node8 -->
-<g id="edge139" class="edge">
+<g id="edge141" class="edge">
<title>Node3->Node8</title>
-<path fill="none" stroke="#191970" d="M2320.0145,-548.4579C2396.0773,-532.2581 2538.6901,-497.1014 2568,-456 2614.2346,-391.1651 2583.3471,-363.3909 2473,-291 2270.6038,-158.2222 2179.4571,-214.4184 1940,-179 1853.132,-166.1512 1602.8926,-147.0104 1479.9115,-138.0141"/>
-<polygon fill="#191970" stroke="#191970" points="1480.076,-134.5168 1469.8478,-137.2797 1479.5665,-141.4983 1480.076,-134.5168"/>
+<path fill="none" stroke="#191970" d="M1899.4474,-548.4207C1911.8903,-528.975 1938,-483.0961 1938,-440.5 1938,-440.5 1938,-440.5 1938,-306.5 1938,-233.9934 1863.0326,-174.6965 1822.815,-148.1024"/>
+<polygon fill="#191970" stroke="#191970" points="1824.6357,-145.112 1814.3338,-142.6381 1820.8445,-150.9964 1824.6357,-145.112"/>
</g>
<!-- Node9 -->
<g id="node10" class="node">
<title>Node9</title>
<g id="a_node10"><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="1921.5,-56.5 1921.5,-86.5 2050.5,-86.5 2050.5,-56.5 1921.5,-56.5"/>
-<text text-anchor="start" x="1929.5" y="-74.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/c_runtime</text>
-<text text-anchor="middle" x="1986" y="-63.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="600.5,-56.5 600.5,-86.5 729.5,-86.5 729.5,-56.5 600.5,-56.5"/>
+<text text-anchor="start" x="608.5" y="-74.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/c_runtime</text>
+<text text-anchor="middle" x="665" y="-63.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
</a>
</g>
</g>
<!-- Node3->Node9 -->
-<g id="edge137" class="edge">
+<g id="edge139" class="edge">
<title>Node3->Node9</title>
-<path fill="none" stroke="#191970" d="M2321.7269,-553.2756C2369.1402,-547.5625 2441.8747,-535.6106 2501,-512 2544.4018,-494.6683 2567.4065,-496.9101 2590,-456 2594.75,-447.3992 2597.7283,-390.0156 2589,-358 2543.8966,-192.5602 2446.4396,-174.8857 2283,-123 2209.3264,-99.6115 2121.7975,-86.0678 2060.9274,-78.7946"/>
-<polygon fill="#191970" stroke="#191970" points="2060.9768,-75.2767 2050.6386,-77.5945 2060.1658,-82.2296 2060.9768,-75.2767"/>
+<path fill="none" stroke="#191970" d="M1843.3748,-556.5962C1590.7422,-548.9259 455.7681,-507.9755 361,-389 287.2447,-296.4048 62.5773,-476.0146 518,-123 539.038,-106.6927 565.6307,-95.3119 590.3962,-87.48"/>
+<polygon fill="#191970" stroke="#191970" points="591.578,-90.7795 600.1446,-84.5455 589.5602,-84.0766 591.578,-90.7795"/>
</g>
<!-- Node3->Node14 -->
-<g id="edge140" class="edge">
+<g id="edge142" class="edge">
<title>Node3->Node14</title>
-<path fill="none" stroke="#191970" d="M2321.58,-554.8533C2423.1381,-548.0513 2647.7604,-531.0462 2677,-512 2736.7514,-473.0789 2758,-444.8098 2758,-373.5 2758,-373.5 2758,-373.5 2758,-306.5 2758,-208.0696 2711.6246,-183.5224 2634,-123 2601.348,-97.5419 2554.9596,-83.8969 2524.1979,-77.1772"/>
-<polygon fill="#191970" stroke="#191970" points="2524.5174,-73.6697 2514.017,-75.0812 2523.1058,-80.5259 2524.5174,-73.6697"/>
+<path fill="none" stroke="#191970" d="M1843.308,-557.7343C1689.0632,-556.5161 1204.8903,-549.7289 806,-512 636.0456,-495.9249 593.5133,-488.8968 426,-456 275.1181,-426.3694 114,-460.2638 114,-306.5 114,-306.5 114,-306.5 114,-189 114,-140.5473 164.5515,-104.5212 198.3292,-85.8573"/>
+<polygon fill="#191970" stroke="#191970" points="200.1693,-88.843 207.3555,-81.0577 196.8828,-82.6624 200.1693,-88.843"/>
</g>
<!-- Node3->Node15 -->
-<g id="edge141" class="edge">
+<g id="edge143" class="edge">
<title>Node3->Node15</title>
-<path fill="none" stroke="#191970" d="M2321.7609,-556.6799C2420.415,-553.4198 2637.0489,-542.8741 2704,-512 2860.7359,-439.7222 2986,-417.5984 2986,-245 2986,-245 2986,-245 2986,-189 2986,-121.0882 2897.1098,-91.1069 2840.9296,-78.94"/>
-<polygon fill="#191970" stroke="#191970" points="2841.3485,-75.4531 2830.8489,-76.8703 2839.9407,-82.31 2841.3485,-75.4531"/>
+<path fill="none" stroke="#191970" d="M1942.777,-557.0747C2162.7672,-552.8602 3039.29,-534.7027 3093,-512 3135.4198,-494.0695 3168,-486.5536 3168,-440.5 3168,-440.5 3168,-440.5 3168,-189 3168,-154.6999 3156.3041,-142.3744 3128,-123 3088.4744,-95.9444 2952.1079,-81.0096 2880.5835,-74.8707"/>
+<polygon fill="#191970" stroke="#191970" points="2880.7776,-71.3748 2870.5205,-74.0284 2880.1937,-78.3504 2880.7776,-71.3748"/>
</g>
<!-- Node16 -->
<g id="node17" class="node">
<title>Node16</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="564.5,-62 564.5,-81 609.5,-81 609.5,-62 564.5,-62"/>
-<text text-anchor="middle" x="587" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2386.5,-62 2386.5,-81 2431.5,-81 2431.5,-62 2386.5,-62"/>
+<text text-anchor="middle" x="2409" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
</g>
<!-- Node3->Node16 -->
-<g id="edge142" class="edge">
+<g id="edge144" class="edge">
<title>Node3->Node16</title>
-<path fill="none" stroke="#191970" d="M2222.2197,-557.6925C2020.5047,-555.8665 1244.6196,-543.2304 614,-456 452.3548,-433.6404 396.5272,-466.6471 253,-389 146.9749,-331.6412 101.3767,-192.3167 200,-123 256.9761,-82.9547 471.8404,-74.0153 554.4055,-72.0466"/>
-<polygon fill="#191970" stroke="#191970" points="554.5548,-75.5443 564.4757,-71.8259 554.4014,-68.546 554.5548,-75.5443"/>
+<path fill="none" stroke="#191970" d="M1942.5902,-557.6665C2160.6979,-555.9968 3025.8762,-547.1712 3071,-512 3097.1224,-491.6392 3092,-473.6201 3092,-440.5 3092,-440.5 3092,-440.5 3092,-189 3092,-122.2269 2578.4286,-82.9049 2441.9641,-73.6351"/>
+<polygon fill="#191970" stroke="#191970" points="2441.9365,-70.1255 2431.7246,-72.9481 2441.4679,-77.1098 2441.9365,-70.1255"/>
</g>
<!-- Node18 -->
<g id="node19" class="node">
<title>Node18</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1384.5,-235.5 1384.5,-254.5 1431.5,-254.5 1431.5,-235.5 1384.5,-235.5"/>
-<text text-anchor="middle" x="1408" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2504.5,-235.5 2504.5,-254.5 2551.5,-254.5 2551.5,-235.5 2504.5,-235.5"/>
+<text text-anchor="middle" x="2528" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
</g>
<!-- Node3->Node18 -->
-<g id="edge143" class="edge">
+<g id="edge145" class="edge">
<title>Node3->Node18</title>
-<path fill="none" stroke="#191970" d="M2222.3317,-555.9133C2048.5058,-547.9595 1478,-516.0965 1478,-440.5 1478,-440.5 1478,-440.5 1478,-373.5 1478,-329.1535 1445.4523,-285.5371 1424.7445,-262.2749"/>
-<polygon fill="#191970" stroke="#191970" points="1427.0821,-259.6466 1417.7332,-254.6618 1421.9331,-264.3887 1427.0821,-259.6466"/>
+<path fill="none" stroke="#191970" d="M1942.7819,-553.0713C2110.2289,-535.2093 2646.3393,-467.1367 2713,-322 2718.7505,-309.4797 2721.5774,-301.7822 2713,-291 2670.692,-237.8166 2629.2884,-271.2851 2561.4284,-255.9002"/>
+<polygon fill="#191970" stroke="#191970" points="2562.2314,-252.4929 2551.6747,-253.391 2560.4873,-259.2721 2562.2314,-252.4929"/>
</g>
<!-- Node22 -->
<g id="node23" class="node">
<title>Node22</title>
<g id="a_node23"><a xlink:href="runtime_2memory_8h.html" target="_top" xlink:title="Runtime memory management. ">
-<polygon fill="#ffffff" stroke="#000000" points="723.5,-179.5 723.5,-198.5 852.5,-198.5 852.5,-179.5 723.5,-179.5"/>
-<text text-anchor="middle" x="788" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2118.5,-179.5 2118.5,-198.5 2247.5,-198.5 2247.5,-179.5 2118.5,-179.5"/>
+<text text-anchor="middle" x="2183" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
</a>
</g>
</g>
<!-- Node3->Node22 -->
-<g id="edge138" class="edge">
+<g id="edge140" class="edge">
<title>Node3->Node22</title>
-<path fill="none" stroke="#191970" d="M2222.291,-554.7138C2109.6117,-547.2512 1826.1236,-528.4001 1589,-512 1238.289,-487.7439 1149.104,-497.3876 800,-456 716.7759,-446.1335 697.0265,-436.4108 614,-425 540.4808,-414.8958 337.3352,-438.4487 282,-389 230.9544,-343.3846 231.5761,-277.8055 285,-235 317.6393,-208.848 580.9834,-196.1275 713.0475,-191.3435"/>
-<polygon fill="#191970" stroke="#191970" points="713.2263,-194.8394 723.0954,-190.9856 712.9771,-187.8438 713.2263,-194.8394"/>
+<path fill="none" stroke="#191970" d="M1942.7675,-553.846C2019.3722,-547.1145 2170.6682,-532.4812 2298,-512 2591.3698,-464.8118 2842.7124,-603.1072 2939,-322 2956.6183,-270.5642 2904.5602,-252.2509 2853,-235 2797.8215,-216.5385 2421.3228,-198.8991 2258.1552,-192.0246"/>
+<polygon fill="#191970" stroke="#191970" points="2257.8596,-188.5092 2247.7218,-191.5875 2257.5665,-195.5031 2257.8596,-188.5092"/>
</g>
<!-- Node34 -->
<g id="node35" class="node">
<title>Node34</title>
<g id="a_node35"><a xlink:href="structural__hash_8h.html" target="_top" xlink:title="tvm/node/structural\l_hash.h">
-<polygon fill="#ffffff" stroke="#000000" points="2113.5,-425.5 2113.5,-455.5 2226.5,-455.5 2226.5,-425.5 2113.5,-425.5"/>
-<text text-anchor="start" x="2121.5" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
-<text text-anchor="middle" x="2170" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_hash.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1966.5,-425.5 1966.5,-455.5 2079.5,-455.5 2079.5,-425.5 1966.5,-425.5"/>
+<text text-anchor="start" x="1974.5" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
+<text text-anchor="middle" x="2023" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_hash.h</text>
</a>
</g>
</g>
<!-- Node3->Node34 -->
-<g id="edge136" class="edge">
+<g id="edge138" class="edge">
<title>Node3->Node34</title>
-<path fill="none" stroke="#191970" d="M2263.6529,-548.3845C2247.6819,-529.9865 2212.3982,-489.341 2189.9896,-463.5272"/>
-<polygon fill="#191970" stroke="#191970" points="2192.4058,-460.9715 2183.2073,-455.7143 2187.1197,-465.5603 2192.4058,-460.9715"/>
+<path fill="none" stroke="#191970" d="M1906.5086,-548.2326C1918.7247,-539.2229 1937.0359,-525.2725 1952,-512 1969.5576,-496.4271 1988.2117,-477.5404 2001.9911,-463.0897"/>
+<polygon fill="#191970" stroke="#191970" points="2004.7777,-465.2371 2009.107,-455.5671 1999.6924,-460.4267 2004.7777,-465.2371"/>
</g>
<!-- Node47 -->
<g id="node48" class="node">
<title>Node47</title>
<g id="a_node48"><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="2253.5,-492.5 2253.5,-511.5 2384.5,-511.5 2384.5,-492.5 2253.5,-492.5"/>
-<text text-anchor="middle" x="2319" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/repr_printer.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2930.5,-492.5 2930.5,-511.5 3061.5,-511.5 3061.5,-492.5 2930.5,-492.5"/>
+<text text-anchor="middle" x="2996" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/repr_printer.h</text>
</a>
</g>
</g>
<!-- Node3->Node47 -->
-<g id="edge132" class="edge">
+<g id="edge134" class="edge">
<title>Node3->Node47</title>
-<path fill="none" stroke="#191970" d="M2280.1868,-548.2455C2286.8589,-540.2958 2296.4479,-528.8706 2304.4253,-519.3656"/>
-<polygon fill="#191970" stroke="#191970" points="2307.1592,-521.5525 2310.907,-511.6427 2301.7974,-517.0524 2307.1592,-521.5525"/>
+<path fill="none" stroke="#191970" d="M1942.7558,-557.1036C2090.8357,-554.1468 2541.5662,-543.0536 2914,-512 2916.0656,-511.8278 2918.1594,-511.6431 2920.2715,-511.448"/>
+<polygon fill="#191970" stroke="#191970" points="2920.6443,-514.9283 2930.2534,-510.4657 2919.9587,-507.9619 2920.6443,-514.9283"/>
</g>
<!-- Node4->Node5 -->
<g id="edge5" class="edge">
<title>Node4->Node5</title>
-<path fill="none" stroke="#191970" d="M1989.2872,-494.7763C1923.8383,-486.6701 1816.7037,-472.4762 1725,-456 1723.5254,-455.7351 1722.0348,-455.4613 1720.5329,-455.1802"/>
-<polygon fill="#191970" stroke="#191970" points="1720.9896,-451.7036 1710.5056,-453.2324 1719.6547,-458.5752 1720.9896,-451.7036"/>
+<path fill="none" stroke="#191970" d="M1788.4414,-493.7322C1783.5581,-493.1244 1778.6981,-492.5391 1774,-492 1611.3386,-473.3336 1566.5078,-482.7641 1400.8308,-455.9915"/>
+<polygon fill="#191970" stroke="#191970" points="1401.2179,-452.5084 1390.784,-454.348 1400.0877,-459.4166 1401.2179,-452.5084"/>
</g>
<!-- Node4->Node8 -->
-<g id="edge95" class="edge">
+<g id="edge97" class="edge">
<title>Node4->Node8</title>
-<path fill="none" stroke="#191970" d="M2050.8168,-492.4238C2053.0699,-459.4244 2055.8392,-349.2738 1999,-291 1926.52,-216.6906 1619.4896,-163.5169 1479.7608,-142.6996"/>
-<polygon fill="#191970" stroke="#191970" points="1479.9953,-139.1964 1469.5913,-141.1977 1478.9726,-146.1213 1479.9953,-139.1964"/>
+<path fill="none" stroke="#191970" d="M1848.3184,-492.4153C1845.4332,-453.2508 1833.2896,-301.6371 1810,-179 1808.3297,-170.2049 1805.9206,-160.6046 1803.6985,-152.4604"/>
+<polygon fill="#191970" stroke="#191970" points="1807.0512,-151.454 1800.963,-142.7832 1800.3152,-153.3582 1807.0512,-151.454"/>
</g>
<!-- Node4->Node9 -->
-<g id="edge91" class="edge">
+<g id="edge93" class="edge">
<title>Node4->Node9</title>
-<path fill="none" stroke="#191970" d="M2110.6837,-494.8962C2158.6974,-487.9041 2220.2054,-475.4194 2236,-456 2282.2935,-399.0823 2271.587,-357.6869 2241,-291 2196.6172,-194.235 2086.1883,-123.9994 2026.2458,-91.5625"/>
-<polygon fill="#191970" stroke="#191970" points="2027.5488,-88.2911 2017.0768,-86.6828 2024.2602,-94.4706 2027.5488,-88.2911"/>
+<path fill="none" stroke="#191970" d="M1788.3432,-500.7053C1576.8915,-496.01 879.8181,-478.9393 782,-456 632.1533,-420.8595 523.2844,-407.1405 500,-255 490.7064,-194.2756 494.1304,-162.7123 541,-123 557.8121,-108.7552 579.1156,-98.0084 599.3012,-90.1185"/>
+<polygon fill="#191970" stroke="#191970" points="600.6382,-93.3554 608.7853,-86.5823 598.1926,-86.7965 600.6382,-93.3554"/>
</g>
<!-- Node4->Node14 -->
-<g id="edge129" class="edge">
+<g id="edge131" class="edge">
<title>Node4->Node14</title>
-<path fill="none" stroke="#191970" d="M2110.7464,-499.3223C2198.8399,-494.579 2356.9903,-482.5767 2407,-456 2492.8715,-410.3652 2592.6764,-238.1406 2620,-143 2632.4688,-99.5838 2566.1421,-82.0021 2524.2714,-75.3006"/>
-<polygon fill="#191970" stroke="#191970" points="2524.6188,-71.8142 2514.216,-73.8222 2523.6005,-78.7397 2524.6188,-71.8142"/>
+<path fill="none" stroke="#191970" d="M1788.3255,-500.8894C1577.3541,-496.8365 876.4601,-481.7188 653,-456 422.155,-429.4312 152,-538.8689 152,-306.5 152,-306.5 152,-306.5 152,-189 152,-147.4563 185.9099,-108.986 208.4281,-88.0193"/>
+<polygon fill="#191970" stroke="#191970" points="211.0192,-90.3968 216.125,-81.1133 206.3444,-85.1865 211.0192,-90.3968"/>
</g>
<!-- Node4->Node15 -->
-<g id="edge130" class="edge">
+<g id="edge132" class="edge">
<title>Node4->Node15</title>
-<path fill="none" stroke="#191970" d="M2110.8601,-498.9535C2205.4366,-493.5321 2382.8666,-480.4169 2441,-456 2482.9649,-438.3741 2727.3288,-230.6911 2760,-199 2783.5296,-176.1762 2796.6269,-173.7444 2808,-143 2814.3252,-125.9012 2809.6514,-105.2624 2804.4001,-90.5631"/>
-<polygon fill="#191970" stroke="#191970" points="2807.5648,-89.049 2800.6232,-81.045 2801.0584,-91.6309 2807.5648,-89.049"/>
+<path fill="none" stroke="#191970" d="M1909.6034,-499.7238C2065.0689,-493.6622 2475.2593,-476.2051 2610,-456 2735.2607,-437.2165 2765.3527,-424.2843 2887,-389 2996.5397,-357.2275 3130,-420.5546 3130,-306.5 3130,-306.5 3130,-306.5 3130,-189 3130,-157.7875 3129.7667,-143.2327 3106,-123 3072.716,-94.6652 2948.7985,-80.5105 2881.0114,-74.7514"/>
+<polygon fill="#191970" stroke="#191970" points="2880.8896,-71.2296 2870.6364,-73.8983 2880.3159,-78.2061 2880.8896,-71.2296"/>
</g>
<!-- Node4->Node18 -->
-<g id="edge131" class="edge">
+<g id="edge133" class="edge">
<title>Node4->Node18</title>
-<path fill="none" stroke="#191970" d="M1989.4027,-494.3709C1937.8701,-487.0712 1862.2289,-474.4292 1798,-456 1762.1426,-445.7115 1755.6476,-435.9935 1720,-425 1649.9337,-403.392 1619.6994,-430.1312 1559,-389 1515.3216,-359.4026 1531.1027,-326.4229 1492,-291 1477.0639,-277.4694 1457.4402,-266.4393 1440.9164,-258.5752"/>
-<polygon fill="#191970" stroke="#191970" points="1442.1542,-255.2926 1431.6045,-254.3158 1439.2424,-261.6583 1442.1542,-255.2926"/>
+<path fill="none" stroke="#191970" d="M1909.5557,-498.8194C2086.7994,-488.1063 2598.1837,-446.7252 2696,-322 2704.5024,-311.1586 2704.4331,-301.8954 2696,-291 2658.9231,-243.0971 2622.0462,-270.2914 2561.5977,-256.0056"/>
+<polygon fill="#191970" stroke="#191970" points="2562.256,-252.5583 2551.69,-253.3388 2560.4366,-259.3177 2562.256,-252.5583"/>
</g>
<!-- Node4->Node22 -->
-<g id="edge93" class="edge">
+<g id="edge95" class="edge">
<title>Node4->Node22</title>
-<path fill="none" stroke="#191970" d="M1989.2753,-498.4239C1684.4784,-480.428 335.2323,-400.1932 317,-389 258.6264,-353.1632 252.9894,-271.4215 311,-235 344.1961,-214.1581 587.0063,-199.0738 712.8304,-192.5646"/>
-<polygon fill="#191970" stroke="#191970" points="713.3494,-196.0427 723.1574,-192.036 712.9915,-189.0518 713.3494,-196.0427"/>
+<path fill="none" stroke="#191970" d="M1909.7861,-500.3566C2012.3447,-496.8278 2225.9882,-486.2181 2404,-456 2638.0225,-416.2739 2775.7279,-509.7251 2921,-322 3004.1529,-214.5473 2461.1067,-193.8999 2257.7103,-189.9385"/>
+<polygon fill="#191970" stroke="#191970" points="2257.7679,-186.4391 2247.7043,-189.7518 2257.6372,-193.4378 2257.7679,-186.4391"/>
</g>
<!-- Node33 -->
<g id="node34" class="node">
<title>Node33</title>
<g id="a_node34"><a xlink:href="data__type_8h.html" target="_top" xlink:title="tvm/runtime/data_type.h">
-<polygon fill="#ffffff" stroke="#000000" points="2326,-297 2326,-316 2464,-316 2464,-297 2326,-297"/>
-<text text-anchor="middle" x="2395" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/data_type.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="556,-297 556,-316 694,-316 694,-297 556,-297"/>
+<text text-anchor="middle" x="625" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/data_type.h</text>
</a>
</g>
</g>
<!-- Node4->Node33 -->
-<g id="edge92" class="edge">
+<g id="edge94" class="edge">
<title>Node4->Node33</title>
-<path fill="none" stroke="#191970" d="M2110.5357,-497.3455C2190.274,-490.5292 2325.6928,-476.3887 2371,-456 2391.9915,-446.5536 2440.3266,-409.8879 2450,-389 2455.7899,-376.4978 2455.4809,-370.6407 2450,-358 2443.6351,-343.3206 2430.7971,-330.9121 2419.1244,-321.9554"/>
-<polygon fill="#191970" stroke="#191970" points="2421.0999,-319.065 2410.94,-316.0606 2417.0088,-324.7451 2421.0999,-319.065"/>
+<path fill="none" stroke="#191970" d="M1788.1923,-500.0524C1557.4996,-492.1536 744.8302,-459.5048 655,-389 635.758,-373.8975 628.8515,-345.3039 626.3765,-326.072"/>
+<polygon fill="#191970" stroke="#191970" points="629.8475,-325.6061 625.3715,-316.0032 622.8821,-326.3014 629.8475,-325.6061"/>
</g>
<!-- Node4->Node34 -->
-<g id="edge67" class="edge">
+<g id="edge69" class="edge">
<title>Node4->Node34</title>
-<path fill="none" stroke="#191970" d="M2068.75,-492.3906C2085.5169,-483.7976 2110.6134,-470.9356 2131.6252,-460.1671"/>
-<polygon fill="#191970" stroke="#191970" points="2133.3417,-463.2203 2140.6447,-455.5446 2130.149,-456.9908 2133.3417,-463.2203"/>
+<path fill="none" stroke="#191970" d="M1876.1875,-492.3906C1901.3716,-483.4893 1939.5159,-470.0073 1970.6118,-459.0165"/>
+<polygon fill="#191970" stroke="#191970" points="1972.1728,-462.1771 1980.4348,-455.5446 1969.84,-455.5772 1972.1728,-462.1771"/>
</g>
<!-- Node35 -->
<g id="node36" class="node">
<title>Node35</title>
<g id="a_node36"><a xlink:href="ndarray_8h.html" target="_top" xlink:title="A device-independent managed NDArray abstraction. ">
-<polygon fill="#ffffff" stroke="#000000" points="1567.5,-364 1567.5,-383 1692.5,-383 1692.5,-364 1567.5,-364"/>
-<text text-anchor="middle" x="1630" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/ndarray.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1970.5,-364 1970.5,-383 2095.5,-383 2095.5,-364 1970.5,-364"/>
+<text text-anchor="middle" x="2033" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/ndarray.h</text>
</a>
</g>
</g>
<!-- Node4->Node35 -->
-<g id="edge94" class="edge">
+<g id="edge96" class="edge">
<title>Node4->Node35</title>
-<path fill="none" stroke="#191970" d="M2018.7749,-492.4466C1943.9661,-469.5587 1754.5031,-411.592 1670.8123,-385.9866"/>
-<polygon fill="#191970" stroke="#191970" points="1671.7317,-382.6078 1661.1453,-383.029 1669.6837,-389.3016 1671.7317,-382.6078"/>
+<path fill="none" stroke="#191970" d="M1862.2292,-492.4144C1882.5187,-477.748 1922.5877,-448.9218 1957,-425 1974.6655,-412.7198 1994.8316,-399.0715 2009.9498,-388.9121"/>
+<polygon fill="#191970" stroke="#191970" points="2012.2883,-391.558 2018.6432,-383.0807 2008.3888,-385.7447 2012.2883,-391.558"/>
</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. ">
-<polygon fill="#ffffff" stroke="#000000" points="1301,-425.5 1301,-455.5 1417,-455.5 1417,-425.5 1301,-425.5"/>
-<text text-anchor="start" x="1309" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/packed</text>
-<text text-anchor="middle" x="1359" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_func.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1409,-425.5 1409,-455.5 1525,-455.5 1525,-425.5 1409,-425.5"/>
+<text text-anchor="start" x="1417" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/packed</text>
+<text text-anchor="middle" x="1467" y="-432.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_func.h</text>
</a>
</g>
</g>
<!-- Node4->Node41 -->
-<g id="edge96" class="edge">
+<g id="edge98" class="edge">
<title>Node4->Node41</title>
-<path fill="none" stroke="#191970" d="M1989.284,-496.5962C1860.8969,-485.1696 1562.6136,-458.6219 1427.5892,-446.6045"/>
-<polygon fill="#191970" stroke="#191970" points="1427.6454,-443.0958 1417.3745,-445.6954 1427.0248,-450.0682 1427.6454,-443.0958"/>
+<path fill="none" stroke="#191970" d="M1789.7316,-492.4581C1720.8801,-481.3734 1607.5004,-463.1198 1535.0905,-451.4622"/>
+<polygon fill="#191970" stroke="#191970" points="1535.5831,-447.9965 1525.1539,-449.8625 1534.4704,-454.9075 1535.5831,-447.9965"/>
</g>
<!-- Node6 -->
<g id="node7" class="node">
<title>Node6</title>
<g id="a_node7"><a xlink:href="functor_8h.html" target="_top" xlink:title="Defines the Functor data structures. ">
-<polygon fill="#ffffff" stroke="#000000" points="2122.5,-297 2122.5,-316 2231.5,-316 2231.5,-297 2122.5,-297"/>
-<text text-anchor="middle" x="2177" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/functor.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2802.5,-297 2802.5,-316 2911.5,-316 2911.5,-297 2802.5,-297"/>
+<text text-anchor="middle" x="2857" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/functor.h</text>
</a>
</g>
</g>
<!-- Node5->Node6 -->
<g id="edge6" class="edge">
<title>Node5->Node6</title>
-<path fill="none" stroke="#191970" d="M1710.7187,-437.1869C1808.9208,-430.8388 2006.2247,-415.215 2070,-389 2089.1997,-381.1079 2132.997,-344.6936 2158.2021,-322.9686"/>
-<polygon fill="#191970" stroke="#191970" points="2160.714,-325.423 2165.9748,-316.2266 2156.1272,-320.1351 2160.714,-325.423"/>
+<path fill="none" stroke="#191970" d="M1390.7588,-426.4875C1393.8748,-425.9318 1396.9701,-425.4296 1400,-425 1710.6349,-380.9573 1792.2655,-414.1179 2105,-389 2359.9307,-368.5247 2662.1342,-331.4649 2792.1942,-314.8974"/>
+<polygon fill="#191970" stroke="#191970" points="2792.9014,-318.3356 2802.3776,-313.5972 2792.0148,-311.3919 2792.9014,-318.3356"/>
</g>
<!-- Node5->Node14 -->
-<g id="edge66" class="edge">
+<g id="edge68" class="edge">
<title>Node5->Node14</title>
-<path fill="none" stroke="#191970" d="M1710.8585,-436.4513C1815.0185,-428.6305 2032.4121,-410.3235 2105,-389 2169.6496,-370.0085 2186.2019,-361.2098 2241,-322 2343.1132,-248.9347 2441.8321,-133.2975 2477.7544,-89.2945"/>
-<polygon fill="#191970" stroke="#191970" points="2480.6902,-91.2308 2484.2653,-81.2573 2475.251,-86.8245 2480.6902,-91.2308"/>
+<path fill="none" stroke="#191970" d="M1277.4834,-440.164C1082.0884,-438.6395 442.6136,-430.4979 361,-389 265.8287,-340.6085 245.3581,-301.0587 214,-199 202.5544,-161.7489 213.2993,-115.8462 221.2446,-90.5278"/>
+<polygon fill="#191970" stroke="#191970" points="224.5711,-91.616 224.4055,-81.0225 217.9288,-89.4071 224.5711,-91.616"/>
</g>
<!-- Node19 -->
<g id="node20" class="node">
<title>Node19</title>
<g id="a_node20"><a xlink:href="object__path_8h.html" target="_top" xlink:title="tvm/node/object_path.h">
-<polygon fill="#ffffff" stroke="#000000" points="1030,-364 1030,-383 1162,-383 1162,-364 1030,-364"/>
-<text text-anchor="middle" x="1096" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/object_path.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="850,-364 850,-383 982,-383 982,-364 850,-364"/>
+<text text-anchor="middle" x="916" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/object_path.h</text>
</a>
</g>
</g>
<!-- Node5->Node19 -->
<g id="edge21" class="edge">
<title>Node5->Node19</title>
-<path fill="none" stroke="#191970" d="M1597.2529,-436.2552C1506.7531,-429.0804 1324.3058,-412.9562 1171,-389 1163.7749,-387.871 1156.1925,-386.5116 1148.7398,-385.073"/>
-<polygon fill="#191970" stroke="#191970" points="1149.1939,-381.5949 1138.7032,-383.0764 1147.8281,-388.4603 1149.1939,-381.5949"/>
+<path fill="none" stroke="#191970" d="M1277.399,-432.9695C1209.2713,-423.6821 1091.4821,-406.9128 991,-389 984.2282,-387.7928 977.1304,-386.4405 970.115,-385.0511"/>
+<polygon fill="#191970" stroke="#191970" points="970.4374,-381.5458 959.943,-383.0014 969.0546,-388.4078 970.4374,-381.5458"/>
</g>
<!-- Node32 -->
<g id="node33" class="node">
<title>Node32</title>
<g id="a_node33"><a xlink:href="array_8h.html" target="_top" xlink:title="Runtime Array container types. ">
-<polygon fill="#ffffff" stroke="#000000" points="294,-291.5 294,-321.5 420,-321.5 420,-291.5 294,-291.5"/>
-<text text-anchor="start" x="302" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="357" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/array.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1548,-358.5 1548,-388.5 1674,-388.5 1674,-358.5 1548,-358.5"/>
+<text text-anchor="start" x="1556" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
+<text text-anchor="middle" x="1611" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/array.h</text>
</a>
</g>
</g>
<!-- Node5->Node32 -->
<g id="edge55" class="edge">
<title>Node5->Node32</title>
-<path fill="none" stroke="#191970" d="M1597.3993,-436.4538C1452.7538,-426.0343 1076.9934,-398.4384 1021,-389 966.6821,-379.844 955.3273,-367.1003 901,-358 701.3702,-324.5603 647.8359,-347.2033 447,-322 441.5922,-321.3214 436.014,-320.5455 430.4086,-319.7108"/>
-<polygon fill="#191970" stroke="#191970" points="430.7376,-316.2202 420.3205,-318.153 429.6692,-323.1382 430.7376,-316.2202"/>
+<path fill="none" stroke="#191970" d="M1390.7235,-426.7799C1433.4873,-416.4363 1492.456,-402.1731 1538.6198,-391.0071"/>
+<polygon fill="#191970" stroke="#191970" points="1539.6665,-394.355 1548.5633,-388.602 1538.0207,-387.5512 1539.6665,-394.355"/>
</g>
<!-- Node5->Node33 -->
-<g id="edge61" class="edge">
+<g id="edge63" class="edge">
<title>Node5->Node33</title>
-<path fill="none" stroke="#191970" d="M1710.9554,-439.544C1835.948,-436.7707 2130.435,-426.3558 2222,-389 2242.1811,-380.7667 2241.1326,-368.9125 2260,-358 2288.9126,-341.2776 2324.2693,-328.0187 2351.5455,-319.1441"/>
-<polygon fill="#191970" stroke="#191970" points="2352.7322,-322.4395 2361.2021,-316.0746 2350.6117,-315.7684 2352.7322,-322.4395"/>
+<path fill="none" stroke="#191970" d="M1277.4688,-438.4436C1130.6619,-432.7533 745.1464,-415.4511 694,-389 666.6166,-374.8383 645.854,-344.6276 634.5623,-324.9598"/>
+<polygon fill="#191970" stroke="#191970" points="637.5394,-323.1094 629.6639,-316.0223 631.4009,-326.4738 637.5394,-323.1094"/>
</g>
<!-- Node7 -->
<g id="node8" class="node">
<title>Node7</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1086.5,-179.5 1086.5,-198.5 1175.5,-198.5 1175.5,-179.5 1086.5,-179.5"/>
-<text text-anchor="middle" x="1131" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/logging.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1592.5,-179.5 1592.5,-198.5 1681.5,-198.5 1681.5,-179.5 1592.5,-179.5"/>
+<text text-anchor="middle" x="1637" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/logging.h</text>
</g>
<!-- Node6->Node7 -->
<g id="edge7" class="edge">
<title>Node6->Node7</title>
-<path fill="none" stroke="#191970" d="M2128.9754,-296.9515C2050.3069,-281.6463 1888.9222,-251.6549 1751,-235 1543.1252,-209.8979 1294.2731,-196.4249 1185.9162,-191.3708"/>
-<polygon fill="#191970" stroke="#191970" points="1186.0253,-187.8722 1175.8747,-190.9079 1185.7029,-194.8648 1186.0253,-187.8722"/>
+<path fill="none" stroke="#191970" d="M2816.9583,-296.9966C2773.7645,-286.7246 2702.9774,-269.831 2642,-255 2605.9693,-246.2366 2597.6872,-240.3912 2561,-235 2391.6464,-210.1133 1863.7619,-194.7797 1691.7725,-190.3434"/>
+<polygon fill="#191970" stroke="#191970" points="1691.77,-186.8423 1681.6837,-190.0851 1691.5908,-193.84 1691.77,-186.8423"/>
</g>
<!-- Node6->Node8 -->
<g id="edge8" class="edge">
<title>Node6->Node8</title>
-<path fill="none" stroke="#191970" d="M2161.0551,-296.8737C2119.4945,-272.3139 2003.725,-207.2744 1898,-179 1820.9927,-158.4057 1595.3433,-143.3183 1479.8521,-136.7071"/>
-<polygon fill="#191970" stroke="#191970" points="1479.8854,-133.2035 1469.7033,-136.1318 1479.4891,-140.1923 1479.8854,-133.2035"/>
+<path fill="none" stroke="#191970" d="M2831.3037,-296.9001C2787.5819,-281.0239 2695.6112,-249.6042 2615,-235 2393.5902,-194.8876 2332.8183,-230.3582 2110,-199 2015.1569,-185.6523 1906.023,-160.2059 1845.2214,-145.1072"/>
+<polygon fill="#191970" stroke="#191970" points="1845.7167,-141.6234 1835.1665,-142.595 1844.0198,-148.4147 1845.7167,-141.6234"/>
</g>
<!-- Node6->Node15 -->
<g id="edge18" class="edge">
<title>Node6->Node15</title>
-<path fill="none" stroke="#191970" d="M2202.0322,-296.9967C2298.8561,-260.238 2648.9872,-127.3126 2761.1633,-84.7256"/>
-<polygon fill="#191970" stroke="#191970" points="2762.5496,-87.9431 2770.6563,-81.1216 2760.0651,-81.3988 2762.5496,-87.9431"/>
+<path fill="none" stroke="#191970" d="M2856.1508,-296.9967C2853.018,-261.9391 2842.0685,-139.4096 2837.7604,-91.2"/>
+<polygon fill="#191970" stroke="#191970" points="2841.2361,-90.7703 2836.8598,-81.1216 2834.2639,-91.3935 2841.2361,-90.7703"/>
</g>
<!-- Node6->Node16 -->
<g id="edge19" class="edge">
<title>Node6->Node16</title>
-<path fill="none" stroke="#191970" d="M2136.3904,-296.9671C2072.5261,-282.2117 1944.5887,-253.6066 1835,-235 1354.7051,-153.4528 764.4616,-89.9392 620.0306,-74.8956"/>
-<polygon fill="#191970" stroke="#191970" points="619.9679,-71.3704 609.66,-73.8191 619.2451,-78.333 619.9679,-71.3704"/>
+<path fill="none" stroke="#191970" d="M2835.3753,-296.955C2798.5639,-280.1532 2722.2518,-242.8856 2666,-199 2629.3148,-170.3795 2633.8612,-147.0004 2594,-123 2545.7016,-93.9196 2480.4894,-80.7865 2441.7824,-75.2213"/>
+<polygon fill="#191970" stroke="#191970" points="2441.9827,-71.7166 2431.604,-73.8453 2441.0449,-78.6535 2441.9827,-71.7166"/>
</g>
<!-- Node6->Node18 -->
<g id="edge20" class="edge">
<title>Node6->Node18</title>
-<path fill="none" stroke="#191970" d="M2122.4817,-301.5777C2087.2587,-298.4407 2040.4311,-294.351 1999,-291 1784.8803,-273.6819 1731.079,-272.8135 1517,-255 1491.7745,-252.901 1463.2895,-250.2789 1441.7664,-248.2464"/>
-<polygon fill="#191970" stroke="#191970" points="1442.0907,-244.7616 1431.8046,-247.3009 1441.4291,-251.7302 1442.0907,-244.7616"/>
+<path fill="none" stroke="#191970" d="M2815.8364,-296.9286C2806.6889,-294.8922 2797.0283,-292.8117 2788,-291 2707.1878,-274.7834 2611.6317,-258.654 2561.9095,-250.4894"/>
+<polygon fill="#191970" stroke="#191970" points="2562.1625,-246.9843 2551.7286,-248.8239 2561.0323,-253.8925 2562.1625,-246.9843"/>
</g>
<!-- Node8->Node9 -->
<g id="edge9" class="edge">
<title>Node8->Node9</title>
-<path fill="none" stroke="#191970" d="M1469.7421,-126.6213C1575.4789,-115.3317 1795.3091,-91.8602 1911.0367,-79.5039"/>
-<polygon fill="#191970" stroke="#191970" points="1911.6527,-82.9581 1921.2246,-78.4161 1910.9095,-75.9977 1911.6527,-82.9581"/>
+<path fill="none" stroke="#191970" d="M1738.2832,-129.7585C1547.1001,-119.381 950.3351,-86.9882 739.6073,-75.5497"/>
+<polygon fill="#191970" stroke="#191970" points="739.7431,-72.052 729.568,-75.0048 739.3636,-79.0418 739.7431,-72.052"/>
</g>
<!-- Node13 -->
<g id="node14" class="node">
<title>Node13</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="894.5,-62 894.5,-81 1019.5,-81 1019.5,-62 894.5,-62"/>
-<text text-anchor="middle" x="957" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/logging.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1241.5,-62 1241.5,-81 1366.5,-81 1366.5,-62 1241.5,-62"/>
+<text text-anchor="middle" x="1304" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/logging.h</text>
</g>
<!-- Node8->Node13 -->
<g id="edge13" class="edge">
<title>Node8->Node13</title>
-<path fill="none" stroke="#191970" d="M1350.3123,-124.8967C1267.7252,-113.6845 1118.743,-93.4585 1029.7471,-81.3763"/>
-<polygon fill="#191970" stroke="#191970" points="1030.1121,-77.8938 1019.7321,-80.0166 1029.1703,-84.8302 1030.1121,-77.8938"/>
+<path fill="none" stroke="#191970" d="M1738.2339,-124.8593C1733.4191,-124.223 1728.6288,-123.5965 1724,-123 1602.2992,-107.3162 1460.6084,-90.1742 1376.8242,-80.1539"/>
+<polygon fill="#191970" stroke="#191970" points="1376.993,-76.6493 1366.6483,-78.9379 1376.1624,-83.5998 1376.993,-76.6493"/>
</g>
<!-- Node8->Node14 -->
<g id="edge14" class="edge">
<title>Node8->Node14</title>
-<path fill="none" stroke="#191970" d="M1469.7142,-129.6059C1669.1625,-118.2694 2308.243,-81.9446 2459.6265,-73.3401"/>
-<polygon fill="#191970" stroke="#191970" points="2459.8853,-76.8311 2469.6705,-72.7692 2459.488,-69.8424 2459.8853,-76.8311"/>
+<path fill="none" stroke="#191970" d="M1738.3411,-130.8232C1562.5978,-124.3898 1031.5363,-104.7999 591,-87 468.8609,-82.0649 323.6271,-75.7195 260.253,-72.9268"/>
+<polygon fill="#191970" stroke="#191970" points="260.241,-69.423 250.0964,-72.4788 259.9324,-76.4162 260.241,-69.423"/>
</g>
<!-- Node8->Node15 -->
<g id="edge15" class="edge">
<title>Node8->Node15</title>
-<path fill="none" stroke="#191970" d="M1469.7368,-130.8552C1636.4171,-124.8008 2120.7403,-106.7309 2523,-87 2603.2353,-83.0644 2696.6811,-77.5593 2750.8153,-74.2791"/>
-<polygon fill="#191970" stroke="#191970" points="2751.3331,-77.7542 2761.1024,-73.6539 2750.9084,-70.7671 2751.3331,-77.7542"/>
+<path fill="none" stroke="#191970" d="M1857.5356,-129.4726C2045.643,-118.3275 2624.5933,-84.0255 2790.898,-74.1722"/>
+<polygon fill="#191970" stroke="#191970" points="2791.4714,-77.6445 2801.2469,-73.5591 2791.0574,-70.6567 2791.4714,-77.6445"/>
</g>
<!-- Node8->Node16 -->
<g id="edge16" class="edge">
<title>Node8->Node16</title>
-<path fill="none" stroke="#191970" d="M1350.4808,-128.5523C1188.6729,-116.461 744.5357,-83.2721 619.9063,-73.959"/>
-<polygon fill="#191970" stroke="#191970" points="620.0712,-70.4616 609.8382,-73.2066 619.5495,-77.4422 620.0712,-70.4616"/>
+<path fill="none" stroke="#191970" d="M1857.6977,-126.9911C1984.6129,-114.2165 2277.6759,-84.7184 2376.0905,-74.8125"/>
+<polygon fill="#191970" stroke="#191970" points="2376.7196,-78.2669 2386.3188,-73.783 2376.0185,-71.3021 2376.7196,-78.2669"/>
</g>
<!-- Node17 -->
<g id="node18" class="node">
<title>Node17</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1429,-62 1429,-81 1479,-81 1479,-62 1429,-62"/>
-<text text-anchor="middle" x="1454" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">atomic</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2012,-62 2012,-81 2062,-81 2062,-62 2012,-62"/>
+<text text-anchor="middle" x="2037" y="-69" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">atomic</text>
</g>
<!-- Node8->Node17 -->
<g id="edge17" class="edge">
<title>Node8->Node17</title>
-<path fill="none" stroke="#191970" d="M1416.875,-123.3906C1423.3593,-114.3273 1433.2414,-100.5149 1441.1801,-89.4188"/>
-<polygon fill="#191970" stroke="#191970" points="1444.0559,-91.4143 1447.0281,-81.2449 1438.3629,-87.3412 1444.0559,-91.4143"/>
+<path fill="none" stroke="#191970" d="M1835.0815,-123.4581C1880.6963,-111.7204 1957.5492,-91.9445 2002.1272,-80.4736"/>
+<polygon fill="#191970" stroke="#191970" points="2003.1145,-83.8336 2011.9268,-77.9519 2001.3701,-77.0544 2003.1145,-83.8336"/>
</g>
<!-- Node10 -->
<g id="node11" class="node">
<title>Node10</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1846.5,-.5 1846.5,-19.5 1939.5,-19.5 1939.5,-.5 1846.5,-.5"/>
-<text text-anchor="middle" x="1893" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dlpack/dlpack.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="525.5,-.5 525.5,-19.5 618.5,-19.5 618.5,-.5 525.5,-.5"/>
+<text text-anchor="middle" x="572" 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="M1963.0112,-56.2977C1948.8034,-46.9022 1930.6215,-34.8787 1916.3475,-25.4395"/>
-<polygon fill="#191970" stroke="#191970" points="1917.8845,-22.2598 1907.6128,-19.6633 1914.0233,-28.0986 1917.8845,-22.2598"/>
+<path fill="none" stroke="#191970" d="M642.0112,-56.2977C627.8034,-46.9022 609.6215,-34.8787 595.3475,-25.4395"/>
+<polygon fill="#191970" stroke="#191970" points="596.8845,-22.2598 586.6128,-19.6633 593.0233,-28.0986 596.8845,-22.2598"/>
</g>
<!-- Node11 -->
<g id="node12" class="node">
<title>Node11</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1958,-.5 1958,-19.5 2014,-19.5 2014,-.5 1958,-.5"/>
-<text text-anchor="middle" x="1986" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stddef.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="637,-.5 637,-19.5 693,-19.5 693,-.5 637,-.5"/>
+<text text-anchor="middle" x="665" 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="M1986,-56.2977C1986,-48.3834 1986,-38.6043 1986,-30.0759"/>
-<polygon fill="#191970" stroke="#191970" points="1989.5001,-29.8469 1986,-19.8469 1982.5001,-29.847 1989.5001,-29.8469"/>
+<path fill="none" stroke="#191970" d="M665,-56.2977C665,-48.3834 665,-38.6043 665,-30.0759"/>
+<polygon fill="#191970" stroke="#191970" points="668.5001,-29.8469 665,-19.8469 661.5001,-29.847 668.5001,-29.8469"/>
</g>
<!-- Node12 -->
<g id="node13" class="node">
<title>Node12</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2032.5,-.5 2032.5,-19.5 2085.5,-19.5 2085.5,-.5 2032.5,-.5"/>
-<text text-anchor="middle" x="2059" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">stdint.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="711.5,-.5 711.5,-19.5 764.5,-19.5 764.5,-.5 711.5,-.5"/>
+<text text-anchor="middle" x="738" 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="M2004.0449,-56.2977C2014.8114,-47.2274 2028.4851,-35.7077 2039.4995,-26.4285"/>
-<polygon fill="#191970" stroke="#191970" points="2041.9191,-28.9667 2047.3118,-19.8469 2037.4089,-23.6132 2041.9191,-28.9667"/>
+<path fill="none" stroke="#191970" d="M683.0449,-56.2977C693.8114,-47.2274 707.4851,-35.7077 718.4995,-26.4285"/>
+<polygon fill="#191970" stroke="#191970" points="720.9191,-28.9667 726.3118,-19.8469 716.4089,-23.6132 720.9191,-28.9667"/>
</g>
<!-- Node19->Node8 -->
<g id="edge53" class="edge">
<title>Node19->Node8</title>
-<path fill="none" stroke="#191970" d="M1105.9669,-363.9267C1123.0013,-347.8233 1159.1616,-314.7496 1193,-291 1219.3964,-272.4736 1230.9992,-275.3706 1256,-255 1290.9087,-226.5564 1287.6299,-206.8677 1323,-179 1339.3792,-166.095 1359.9149,-155.0396 1376.9212,-147.0034"/>
-<polygon fill="#191970" stroke="#191970" points="1378.8347,-149.9766 1386.4682,-142.6295 1375.9191,-143.6127 1378.8347,-149.9766"/>
+<path fill="none" stroke="#191970" d="M899.8673,-363.8848C862.6363,-340.5589 776.8736,-279.6072 817,-235 877.8419,-167.3639 1513.7299,-141.7307 1727.9802,-134.9744"/>
+<polygon fill="#191970" stroke="#191970" points="1728.2768,-138.467 1738.163,-134.6576 1728.059,-131.4704 1728.2768,-138.467"/>
</g>
<!-- Node19->Node14 -->
<g id="edge54" class="edge">
<title>Node19->Node14</title>
-<path fill="none" stroke="#191970" d="M1136.6864,-363.9593C1147.7617,-361.6726 1159.7944,-359.47 1171,-358 1449.0743,-321.5211 1529.5943,-392.6751 1801,-322 1822.4,-316.4274 1959.9626,-244.355 1980,-235 2035.1603,-209.2469 2048.4426,-201.5197 2105,-179 2174.9962,-151.1293 2192.5909,-143.8126 2265,-123 2333.6762,-103.2604 2415.7807,-86.2195 2459.9981,-77.5669"/>
-<polygon fill="#191970" stroke="#191970" points="2460.7254,-80.9911 2469.8757,-75.6506 2459.3921,-74.1193 2460.7254,-80.9911"/>
+<path fill="none" stroke="#191970" d="M890.0332,-363.8473C863.2916,-353.7949 820.4902,-337.3943 784,-322 586.9971,-238.8892 541.3217,-209.9884 346,-123 316.7477,-109.9722 283.2814,-95.3973 259.4786,-85.0867"/>
+<polygon fill="#191970" stroke="#191970" points="260.7942,-81.8425 250.2266,-81.0831 258.0142,-88.2668 260.7942,-81.8425"/>
</g>
<!-- Node20 -->
<g id="node21" class="node">
<title>Node20</title>
<g id="a_node21"><a xlink:href="optional_8h.html" target="_top" xlink:title="Runtime Optional container types. ">
-<polygon fill="#ffffff" stroke="#000000" points="552,-291.5 552,-321.5 678,-321.5 678,-291.5 552,-291.5"/>
-<text text-anchor="start" x="560" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="615" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/optional.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1632,-291.5 1632,-321.5 1758,-321.5 1758,-291.5 1632,-291.5"/>
+<text text-anchor="start" x="1640" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
+<text text-anchor="middle" x="1695" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/optional.h</text>
</a>
</g>
</g>
<!-- Node19->Node20 -->
<g id="edge22" class="edge">
<title>Node19->Node20</title>
-<path fill="none" stroke="#191970" d="M1029.7592,-365.6431C932.4143,-353.9965 755.4057,-332.4093 692,-322 690.8899,-321.8178 689.7713,-321.6312 688.6461,-321.4407"/>
-<polygon fill="#191970" stroke="#191970" points="688.8383,-317.9208 678.383,-319.6342 687.6248,-324.8148 688.8383,-317.9208"/>
+<path fill="none" stroke="#191970" d="M957.0939,-363.9981C968.0607,-361.7394 979.9382,-359.5409 991,-358 1111.2079,-341.2549 1466.8508,-319.5915 1621.6416,-310.6451"/>
+<polygon fill="#191970" stroke="#191970" points="1622.2061,-314.1184 1631.9881,-310.0487 1621.8033,-307.13 1622.2061,-314.1184"/>
</g>
<!-- Node19->Node26 -->
<g id="edge36" class="edge">
<title>Node19->Node26</title>
-<path fill="none" stroke="#191970" d="M1084.6812,-363.9005C1073.7678,-354.6448 1056.8657,-340.3101 1042.6828,-328.2816"/>
-<polygon fill="#191970" stroke="#191970" points="1044.6026,-325.3206 1034.7122,-321.5218 1040.0749,-330.6592 1044.6026,-325.3206"/>
+<path fill="none" stroke="#191970" d="M960.3463,-363.9533C970.4065,-361.894 981.0562,-359.7976 991,-358 1079.7173,-341.9619 1182.3318,-326.5167 1249.7212,-316.795"/>
+<polygon fill="#191970" stroke="#191970" points="1250.4142,-320.2314 1259.8143,-315.3437 1249.4179,-313.3027 1250.4142,-320.2314"/>
</g>
<!-- Node20->Node16 -->
<g id="edge23" class="edge">
<title>Node20->Node16</title>
-<path fill="none" stroke="#191970" d="M551.7699,-297.7116C502.7125,-289.4455 440.5226,-275.2901 425,-255 374.1241,-188.4983 497.5251,-115.6639 557.0746,-85.6381"/>
-<polygon fill="#191970" stroke="#191970" points="558.9754,-88.6026 566.3853,-81.03 555.8704,-82.329 558.9754,-88.6026"/>
+<path fill="none" stroke="#191970" d="M1749.0424,-291.439C1808.782,-275.303 1908.0733,-249.9958 1995,-235 2068.3478,-222.3467 2264.8425,-238.3844 2328,-199 2361.8995,-177.8606 2354.5562,-156.7078 2376,-123 2382.9975,-112.0006 2390.8727,-99.7073 2397.1746,-89.8906"/>
+<polygon fill="#191970" stroke="#191970" points="2400.1941,-91.6658 2402.6545,-81.3606 2394.3047,-87.8823 2400.1941,-91.6658"/>
</g>
<!-- Node21 -->
<g id="node22" class="node">
<title>Node21</title>
<g id="a_node22"><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="736.5,-235.5 736.5,-254.5 791.5,-254.5 791.5,-235.5 736.5,-235.5"/>
-<text text-anchor="middle" x="764" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">./base.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1626.5,-235.5 1626.5,-254.5 1681.5,-254.5 1681.5,-235.5 1626.5,-235.5"/>
+<text text-anchor="middle" x="1654" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">./base.h</text>
</a>
</g>
</g>
<!-- Node20->Node21 -->
<g id="edge24" class="edge">
<title>Node20->Node21</title>
-<path fill="none" stroke="#191970" d="M651.4495,-291.4554C675.8744,-281.374 707.7581,-268.2139 731.3521,-258.4755"/>
-<polygon fill="#191970" stroke="#191970" points="733.0021,-261.5809 740.9103,-254.5303 730.3313,-255.1104 733.0021,-261.5809"/>
+<path fill="none" stroke="#191970" d="M1684.8652,-291.2977C1679.2332,-282.8498 1672.1849,-272.2773 1666.246,-263.369"/>
+<polygon fill="#191970" stroke="#191970" points="1669.0238,-261.226 1660.5646,-254.8469 1663.1995,-265.1089 1669.0238,-261.226"/>
</g>
<!-- Node21->Node7 -->
<g id="edge25" class="edge">
<title>Node21->Node7</title>
-<path fill="none" stroke="#191970" d="M791.7822,-241.3199C845.0084,-234.1858 965.2632,-217.6648 1076.4503,-199.211"/>
-<polygon fill="#191970" stroke="#191970" points="1077.2034,-202.6338 1086.4906,-197.5351 1076.0508,-195.7294 1077.2034,-202.6338"/>
+<path fill="none" stroke="#191970" d="M1651.0388,-235.2455C1648.7962,-227.8579 1645.6423,-217.4689 1642.8919,-208.4087"/>
+<polygon fill="#191970" stroke="#191970" points="1646.1812,-207.1948 1639.9272,-198.6427 1639.483,-209.2282 1646.1812,-207.1948"/>
</g>
<!-- Node21->Node8 -->
<g id="edge32" class="edge">
<title>Node21->Node8</title>
-<path fill="none" stroke="#191970" d="M789.155,-235.4475C828.5832,-220.8782 907.6109,-193.2941 977,-179 1103.4685,-152.9475 1253.9343,-141.1822 1340.0743,-136.2367"/>
-<polygon fill="#191970" stroke="#191970" points="1340.4656,-139.7204 1350.2543,-135.6669 1340.0743,-132.7313 1340.4656,-139.7204"/>
+<path fill="none" stroke="#191970" d="M1670.0233,-235.4072C1684.5145,-226.5151 1706.238,-212.6392 1724,-199 1744.2756,-183.4307 1765.9287,-163.7158 1780.5386,-149.9038"/>
+<polygon fill="#191970" stroke="#191970" points="1783.0583,-152.3372 1787.8771,-142.9016 1778.226,-147.2728 1783.0583,-152.3372"/>
</g>
<!-- Node21->Node13 -->
<g id="edge26" class="edge">
<title>Node21->Node13</title>
-<path fill="none" stroke="#191970" d="M791.8216,-236.1182C812.6474,-228.531 841.0346,-216.1158 862,-199 900.6605,-167.4383 932.0973,-116.8633 947.231,-89.8729"/>
-<polygon fill="#191970" stroke="#191970" points="950.3158,-91.5267 952.0521,-81.0751 944.177,-88.1627 950.3158,-91.5267"/>
+<path fill="none" stroke="#191970" d="M1666.1611,-235.2794C1681.4148,-221.9052 1703.9736,-197.3969 1690,-179 1652.2697,-129.326 1477.966,-96.4925 1376.9587,-81.2827"/>
+<polygon fill="#191970" stroke="#191970" points="1377.2253,-77.784 1366.8198,-79.778 1376.1976,-84.7082 1377.2253,-77.784"/>
</g>
<!-- Node21->Node16 -->
<g id="edge35" class="edge">
<title>Node21->Node16</title>
-<path fill="none" stroke="#191970" d="M751.468,-235.3456C740.684,-226.6262 725.1471,-213.0423 714,-199 689.8706,-168.6035 697.9565,-150.9199 671,-123 655.8621,-107.3211 634.9863,-94.4926 617.8333,-85.5607"/>
-<polygon fill="#191970" stroke="#191970" points="619.2551,-82.3583 608.7469,-81.006 616.1182,-88.6161 619.2551,-82.3583"/>
+<path fill="none" stroke="#191970" d="M1681.5448,-236.2655C1683.0491,-235.8281 1684.5414,-235.4035 1686,-235 1890.4666,-178.4415 1942.8106,-168.5849 2150,-123 2231.1336,-105.1494 2327.1743,-86.7665 2376.3692,-77.5484"/>
+<polygon fill="#191970" stroke="#191970" points="2377.2301,-80.9482 2386.4172,-75.6711 2375.9444,-74.0673 2377.2301,-80.9482"/>
</g>
<!-- Node21->Node22 -->
<g id="edge27" class="edge">
<title>Node21->Node22</title>
-<path fill="none" stroke="#191970" d="M768.1805,-235.2455C771.381,-227.7776 775.896,-217.2427 779.8084,-208.1137"/>
-<polygon fill="#191970" stroke="#191970" points="783.1452,-209.2129 783.8674,-198.6427 776.7112,-206.4554 783.1452,-209.2129"/>
+<path fill="none" stroke="#191970" d="M1681.7495,-241.3771C1697.1349,-239.4073 1716.6158,-236.9768 1734,-235 1865.5696,-220.0391 2019.0869,-204.7857 2108.246,-196.1451"/>
+<polygon fill="#191970" stroke="#191970" points="2108.6463,-199.6228 2118.2627,-195.1759 2107.9721,-192.6553 2108.6463,-199.6228"/>
</g>
<!-- Node21->Node24 -->
<g id="edge33" class="edge">
<title>Node21->Node24</title>
-<path fill="none" stroke="#191970" d="M736.218,-241.9494C651.7993,-232.6799 397.8966,-204.8004 296.5011,-193.6668"/>
-<polygon fill="#191970" stroke="#191970" points="296.6457,-190.1617 286.3234,-192.5492 295.8816,-197.1199 296.6457,-190.1617"/>
+<path fill="none" stroke="#191970" d="M1626.3019,-239.1883C1587.7404,-231.0482 1515.4483,-215.6105 1444.2471,-199.2496"/>
+<polygon fill="#191970" stroke="#191970" points="1444.7419,-195.7719 1434.2112,-196.9356 1443.1691,-202.593 1444.7419,-195.7719"/>
</g>
<!-- Node25 -->
<g id="node26" class="node">
<title>Node25</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="985.5,-179.5 985.5,-198.5 1068.5,-198.5 1068.5,-179.5 985.5,-179.5"/>
-<text text-anchor="middle" x="1027" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">initializer_list</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1452.5,-179.5 1452.5,-198.5 1535.5,-198.5 1535.5,-179.5 1452.5,-179.5"/>
+<text text-anchor="middle" x="1494" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">initializer_list</text>
</g>
<!-- Node21->Node25 -->
<g id="edge34" class="edge">
<title>Node21->Node25</title>
-<path fill="none" stroke="#191970" d="M791.6485,-239.1129C835.103,-229.8602 920.0973,-211.7625 975.17,-200.036"/>
-<polygon fill="#191970" stroke="#191970" points="976.0366,-203.4301 985.0884,-197.9242 974.5787,-196.5836 976.0366,-203.4301"/>
+<path fill="none" stroke="#191970" d="M1626.495,-235.3733C1600.1198,-226.1419 1560.1149,-212.1402 1530.9642,-201.9375"/>
+<polygon fill="#191970" stroke="#191970" points="1531.7409,-198.5012 1521.1461,-198.5011 1529.4284,-205.1082 1531.7409,-198.5012"/>
</g>
<!-- Node22->Node8 -->
<g id="edge28" class="edge">
<title>Node22->Node8</title>
-<path fill="none" stroke="#191970" d="M852.5132,-183.1917C970.1308,-172.6024 1218.4813,-150.2428 1340.0219,-139.3003"/>
-<polygon fill="#191970" stroke="#191970" points="1340.6737,-142.7558 1350.3195,-138.3732 1340.0459,-135.784 1340.6737,-142.7558"/>
+<path fill="none" stroke="#191970" d="M2118.1254,-179.5637C2048.7806,-169.4772 1939.2209,-153.5412 1867.9765,-143.1784"/>
+<polygon fill="#191970" stroke="#191970" points="1868.1521,-139.6672 1857.7524,-141.6913 1867.1444,-146.5943 1868.1521,-139.6672"/>
</g>
<!-- Node22->Node15 -->
<g id="edge30" class="edge">
<title>Node22->Node15</title>
-<path fill="none" stroke="#191970" d="M852.5355,-185.2236C1161.9066,-167.1205 2488.9636,-89.4665 2750.9684,-74.1351"/>
-<polygon fill="#191970" stroke="#191970" points="2751.4657,-77.612 2761.2442,-73.5338 2751.0567,-70.624 2751.4657,-77.612"/>
+<path fill="none" stroke="#191970" d="M2223.556,-179.4144C2282.1484,-165.7636 2393.9881,-140.4474 2490,-123 2597.4812,-103.4684 2724.7436,-85.9583 2790.9245,-77.2719"/>
+<polygon fill="#191970" stroke="#191970" points="2791.5469,-80.7204 2801.0093,-75.9545 2790.6402,-73.7793 2791.5469,-80.7204"/>
</g>
<!-- Node22->Node16 -->
<g id="edge31" class="edge">
<title>Node22->Node16</title>
-<path fill="none" stroke="#191970" d="M778.5076,-179.4323C764.3629,-165.5757 736.4635,-139.7516 709,-123 680.2368,-105.4556 644.4254,-91.1213 619.0609,-82.0712"/>
-<polygon fill="#191970" stroke="#191970" points="620.148,-78.7436 609.5532,-78.7513 617.8404,-85.3523 620.148,-78.7436"/>
+<path fill="none" stroke="#191970" d="M2199.0752,-179.4431C2221.4418,-166.2589 2263.3381,-141.9701 2300,-123 2326.0696,-109.5107 2356.2891,-95.3448 2378.2873,-85.2922"/>
+<polygon fill="#191970" stroke="#191970" points="2379.819,-88.4406 2387.4732,-81.1151 2376.9213,-82.0685 2379.819,-88.4406"/>
</g>
<!-- Node23 -->
<g id="node24" class="node">
<title>Node23</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="763.5,-123.5 763.5,-142.5 812.5,-142.5 812.5,-123.5 763.5,-123.5"/>
-<text text-anchor="middle" x="788" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstdlib</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2158.5,-123.5 2158.5,-142.5 2207.5,-142.5 2207.5,-123.5 2158.5,-123.5"/>
+<text text-anchor="middle" x="2183" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstdlib</text>
</g>
<!-- Node22->Node23 -->
<g id="edge29" class="edge">
<title>Node22->Node23</title>
-<path fill="none" stroke="#191970" d="M788,-179.2455C788,-171.9382 788,-161.6944 788,-152.7046"/>
-<polygon fill="#191970" stroke="#191970" points="791.5001,-152.6426 788,-142.6427 784.5001,-152.6427 791.5001,-152.6426"/>
+<path fill="none" stroke="#191970" d="M2183,-179.2455C2183,-171.9382 2183,-161.6944 2183,-152.7046"/>
+<polygon fill="#191970" stroke="#191970" points="2186.5001,-152.6426 2183,-142.6427 2179.5001,-152.6427 2186.5001,-152.6426"/>
</g>
<!-- Node26->Node7 -->
<g id="edge37" class="edge">
<title>Node26->Node7</title>
-<path fill="none" stroke="#191970" d="M1073.9843,-291.4414C1092.8191,-283.6261 1112.0045,-272.0198 1124,-255 1133.4281,-241.6229 1134.5493,-222.7736 1133.679,-208.6418"/>
-<polygon fill="#191970" stroke="#191970" points="1137.1445,-208.1202 1132.6747,-198.5144 1130.1787,-208.811 1137.1445,-208.1202"/>
+<path fill="none" stroke="#191970" d="M1363.0873,-291.4992C1424.9786,-268.3392 1542.7919,-224.253 1601.6685,-202.2212"/>
+<polygon fill="#191970" stroke="#191970" points="1603.1337,-205.41 1611.2728,-198.6272 1600.6804,-198.854 1603.1337,-205.41"/>
</g>
<!-- Node26->Node8 -->
<g id="edge41" class="edge">
<title>Node26->Node8</title>
-<path fill="none" stroke="#191970" d="M1063.6179,-291.4973C1087.9275,-282.6521 1117.5982,-270.2063 1142,-255 1185.2909,-228.0227 1182.8896,-202.8102 1228,-179 1262.7585,-160.6538 1305.1574,-149.3845 1340.2849,-142.5959"/>
-<polygon fill="#191970" stroke="#191970" points="1341.2638,-145.9743 1350.4605,-140.714 1339.9908,-139.091 1341.2638,-145.9743"/>
+<path fill="none" stroke="#191970" d="M1330.9865,-291.0801C1335.9136,-281.0303 1342.0291,-267.5062 1346,-255 1356.4191,-222.185 1335.5844,-202.2256 1361,-179 1387.5125,-154.772 1611.4902,-141.2971 1728.0714,-135.8654"/>
+<polygon fill="#191970" stroke="#191970" points="1728.496,-139.3497 1738.3252,-135.3955 1728.1754,-132.3571 1728.496,-139.3497"/>
</g>
<!-- Node26->Node13 -->
<g id="edge39" class="edge">
<title>Node26->Node13</title>
-<path fill="none" stroke="#191970" d="M956.4515,-291.4417C923.5253,-281.8966 887.5919,-268.783 878,-255 855.22,-222.2667 890.6145,-237.8484 934,-143 941.7801,-125.9912 948.0624,-105.5959 952.0828,-90.9488"/>
-<polygon fill="#191970" stroke="#191970" points="955.5355,-91.5819 954.7038,-81.0198 948.7674,-89.7953 955.5355,-91.5819"/>
+<path fill="none" stroke="#191970" d="M1321.7826,-291.4422C1318.4754,-250.5381 1309.3247,-137.3587 1305.6042,-91.3419"/>
+<polygon fill="#191970" stroke="#191970" points="1309.0768,-90.86 1304.7822,-81.1746 1302.0996,-91.4242 1309.0768,-90.86"/>
</g>
<!-- Node26->Node14 -->
<g id="edge47" class="edge">
<title>Node26->Node14</title>
-<path fill="none" stroke="#191970" d="M1080.254,-300.9519C1116.2272,-297.8661 1162.1228,-294.0457 1203,-291 1446.4059,-272.8644 1513.1689,-309.8789 1751,-255 1902.1098,-220.1318 1923.9449,-162.1592 2074,-123 2214.6449,-86.2965 2388.918,-75.5412 2459.7804,-72.5672"/>
-<polygon fill="#191970" stroke="#191970" points="2460.0037,-76.0612 2469.8574,-72.1683 2459.7267,-69.0667 2460.0037,-76.0612"/>
+<path fill="none" stroke="#191970" d="M1259.8767,-293.9656C1210.9297,-284.0292 1141.4732,-269.4558 1081,-255 1025.4289,-241.7161 639.9792,-134.4428 584,-123 465.7218,-98.8226 323.072,-81.7926 260.3055,-74.9023"/>
+<polygon fill="#191970" stroke="#191970" points="260.5538,-71.4088 250.2344,-73.8089 259.7982,-78.3679 260.5538,-71.4088"/>
</g>
<!-- Node26->Node15 -->
<g id="edge49" class="edge">
<title>Node26->Node15</title>
-<path fill="none" stroke="#191970" d="M1080.2482,-300.8724C1116.2195,-297.7613 1162.1152,-293.9421 1203,-291 1466.8023,-272.0167 1534.4524,-286.9496 1797,-255 2168.8606,-209.748 2611.4788,-113.2398 2751.3591,-81.7027"/>
-<polygon fill="#191970" stroke="#191970" points="2752.3037,-85.0776 2761.2859,-79.459 2750.7603,-78.2498 2752.3037,-85.0776"/>
+<path fill="none" stroke="#191970" d="M1386.1272,-303.1119C1588.363,-292.2319 2208.2386,-258.6746 2229,-255 2286.8539,-244.7604 2670.2874,-124.005 2796.0953,-84.1638"/>
+<polygon fill="#191970" stroke="#191970" points="2797.2621,-87.4656 2805.738,-81.1088 2795.1479,-80.7925 2797.2621,-87.4656"/>
</g>
<!-- Node26->Node16 -->
<g id="edge51" class="edge">
<title>Node26->Node16</title>
-<path fill="none" stroke="#191970" d="M953.819,-292.7177C950.8403,-292.1248 947.8891,-291.5492 945,-291 848.5265,-272.6611 811.1591,-305.6034 727,-255 709.4469,-244.4456 630.6316,-133.703 599.6058,-89.537"/>
-<polygon fill="#191970" stroke="#191970" points="602.3559,-87.3626 593.7499,-81.1831 596.6239,-91.3806 602.3559,-87.3626"/>
+<path fill="none" stroke="#191970" d="M1386.2995,-298.9974C1468.8654,-289.1602 1617.7923,-271.2421 1745,-255 1811.715,-246.4817 1828.1941,-242.7734 1895,-235 2055.599,-216.3131 2107.989,-261.7451 2257,-199 2287.7272,-186.0615 2358.9063,-119.7775 2391.7836,-88.2297"/>
+<polygon fill="#191970" stroke="#191970" points="2394.5004,-90.4721 2399.2691,-81.0111 2389.6413,-85.4333 2394.5004,-90.4721"/>
</g>
<!-- Node26->Node18 -->
<g id="edge52" class="edge">
<title>Node26->Node18</title>
-<path fill="none" stroke="#191970" d="M1080.2555,-296.5506C1163.9933,-283.3796 1308.6679,-260.6238 1374.1092,-250.3307"/>
-<polygon fill="#191970" stroke="#191970" points="1374.943,-253.7426 1384.2777,-248.7313 1373.8553,-246.8276 1374.943,-253.7426"/>
+<path fill="none" stroke="#191970" d="M1386.2076,-303.2101C1446.9191,-300.0559 1541.2264,-295.1711 1623,-291 1940.7672,-274.7913 2020.2406,-271.3616 2338,-255 2392.728,-252.182 2456.3179,-248.8167 2494.2993,-246.7969"/>
+<polygon fill="#191970" stroke="#191970" points="2494.5044,-250.2911 2504.3042,-246.2645 2494.1323,-243.301 2494.5044,-250.2911"/>
</g>
<!-- Node26->Node21 -->
<g id="edge38" class="edge">
<title>Node26->Node21</title>
-<path fill="none" stroke="#191970" d="M955.1092,-291.4554C906.9639,-279.7521 841.7507,-263.8999 801.3045,-254.0681"/>
-<polygon fill="#191970" stroke="#191970" points="802.1244,-250.6655 791.5807,-251.7044 800.471,-257.4675 802.1244,-250.6655"/>
+<path fill="none" stroke="#191970" d="M1386.0589,-296.3344C1444.74,-286.6209 1534.5358,-271.1102 1612,-255 1613.4677,-254.6948 1614.9624,-254.3762 1616.471,-254.0482"/>
+<polygon fill="#191970" stroke="#191970" points="1617.503,-257.4035 1626.4881,-251.7892 1615.963,-250.5749 1617.503,-257.4035"/>
</g>
<!-- Node26->Node22 -->
<g id="edge40" class="edge">
<title>Node26->Node22</title>
-<path fill="none" stroke="#191970" d="M953.9435,-291.7456C912.387,-281.2677 862.3696,-266.9895 844,-255 825.0336,-242.621 808.9892,-221.9412 799.0003,-207.0255"/>
-<polygon fill="#191970" stroke="#191970" points="801.9422,-205.1293 793.5914,-198.609 796.0534,-208.9138 801.9422,-205.1293"/>
+<path fill="none" stroke="#191970" d="M1386.3077,-299.5627C1459.7051,-291.0702 1584.1569,-275.2816 1690,-255 1726.8425,-247.9403 1735.0624,-241.544 1772,-235 1834.7123,-223.8896 2007.8864,-206.0869 2108.3056,-196.2031"/>
+<polygon fill="#191970" stroke="#191970" points="2108.7768,-199.6737 2118.3869,-195.2134 2108.0928,-192.7072 2108.7768,-199.6737"/>
</g>
<!-- Node26->Node24 -->
<g id="edge42" class="edge">
<title>Node26->Node24</title>
-<path fill="none" stroke="#191970" d="M953.8331,-292.6415C950.8506,-292.069 947.8948,-291.5185 945,-291 834.0685,-271.1291 805.5035,-271.3589 694,-255 547.7171,-233.5386 374.2084,-207.274 295.9623,-195.3862"/>
-<polygon fill="#191970" stroke="#191970" points="296.4226,-191.9161 286.0102,-193.8737 295.3707,-198.8366 296.4226,-191.9161"/>
+<path fill="none" stroke="#191970" d="M1347.2513,-291.3181C1359.919,-282.2258 1374.6809,-269.6049 1384,-255 1392.973,-240.9375 1397.5182,-222.4389 1399.7932,-208.6048"/>
+<polygon fill="#191970" stroke="#191970" points="1403.262,-209.0708 1401.1723,-198.6841 1396.3287,-208.107 1403.262,-209.0708"/>
</g>
<!-- Node26->Node25 -->
<g id="edge45" class="edge">
<title>Node26->Node25</title>
-<path fill="none" stroke="#191970" d="M1064.6358,-291.4168C1081.7276,-283.445 1099.1873,-271.7385 1109,-255 1122.5172,-231.9424 1093.8247,-213.6887 1066.7404,-202.2759"/>
-<polygon fill="#191970" stroke="#191970" points="1067.8851,-198.9646 1057.2992,-198.5276 1065.3021,-205.4706 1067.8851,-198.9646"/>
+<path fill="none" stroke="#191970" d="M1347.3849,-291.2955C1363.3516,-281.2195 1384.5922,-267.5861 1403,-255 1427.3672,-238.3392 1454.7064,-218.3269 1472.9999,-204.7392"/>
+<polygon fill="#191970" stroke="#191970" points="1475.2739,-207.4096 1481.2011,-198.6278 1471.0912,-201.7967 1475.2739,-207.4096"/>
</g>
<!-- Node27 -->
<g id="node28" class="node">
<title>Node27</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="886.5,-235.5 886.5,-254.5 939.5,-254.5 939.5,-235.5 886.5,-235.5"/>
-<text text-anchor="middle" x="913" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstddef</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1090.5,-235.5 1090.5,-254.5 1143.5,-254.5 1143.5,-235.5 1090.5,-235.5"/>
+<text text-anchor="middle" x="1117" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstddef</text>
</g>
<!-- Node26->Node27 -->
<g id="edge43" class="edge">
<title>Node26->Node27</title>
-<path fill="none" stroke="#191970" d="M991.5587,-291.4554C975.2716,-281.8241 954.2327,-269.3828 938.027,-259.7996"/>
-<polygon fill="#191970" stroke="#191970" points="939.5055,-256.6078 929.1163,-254.5303 935.9424,-262.6331 939.5055,-256.6078"/>
+<path fill="none" stroke="#191970" d="M1272.6067,-291.4554C1235.6891,-280.4339 1186.4461,-265.7327 1153.3283,-255.8456"/>
+<polygon fill="#191970" stroke="#191970" points="1154.1829,-252.4481 1143.5996,-252.9411 1152.1804,-259.1556 1154.1829,-252.4481"/>
</g>
<!-- Node28 -->
<g id="node29" class="node">
<title>Node28</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="958,-235.5 958,-254.5 1008,-254.5 1008,-235.5 958,-235.5"/>
-<text text-anchor="middle" x="983" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstring</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1162,-235.5 1162,-254.5 1212,-254.5 1212,-235.5 1162,-235.5"/>
+<text text-anchor="middle" x="1187" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">cstring</text>
</g>
<!-- Node26->Node28 -->
<g id="edge44" class="edge">
<title>Node26->Node28</title>
-<path fill="none" stroke="#191970" d="M1008.5955,-291.2977C1003.9743,-282.9388 998.2031,-272.4997 993.3111,-263.6509"/>
-<polygon fill="#191970" stroke="#191970" points="996.3452,-261.9051 988.4438,-254.8469 990.2191,-265.292 996.3452,-261.9051"/>
+<path fill="none" stroke="#191970" d="M1289.7307,-291.4554C1267.6359,-281.464 1238.854,-268.4487 1217.3781,-258.7372"/>
+<polygon fill="#191970" stroke="#191970" points="1218.629,-255.4616 1208.0752,-254.5303 1215.7447,-261.8398 1218.629,-255.4616"/>
</g>
<!-- Node29 -->
<g id="node30" class="node">
<title>Node29</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="320,-235.5 320,-254.5 378,-254.5 378,-235.5 320,-235.5"/>
-<text text-anchor="middle" x="349" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">memory</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="826,-235.5 826,-254.5 884,-254.5 884,-235.5 826,-235.5"/>
+<text text-anchor="middle" x="855" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">memory</text>
</g>
<!-- Node26->Node29 -->
<g id="edge46" class="edge">
<title>Node26->Node29</title>
-<path fill="none" stroke="#191970" d="M953.8755,-292.3769C950.8817,-291.8752 947.9119,-291.4121 945,-291 715.6211,-258.5396 655.5075,-278.1218 425,-255 413.0173,-253.798 400.0097,-252.1963 388.199,-250.6276"/>
-<polygon fill="#191970" stroke="#191970" points="388.5637,-247.1451 378.1838,-249.2685 387.6223,-254.0815 388.5637,-247.1451"/>
+<path fill="none" stroke="#191970" d="M1259.9151,-301.691C1177.7862,-294.8207 1029.704,-280.3535 894.4924,-255.0749"/>
+<polygon fill="#191970" stroke="#191970" points="894.8435,-251.5792 884.3668,-253.1567 893.5406,-258.4569 894.8435,-251.5792"/>
</g>
<!-- Node30 -->
<g id="node31" class="node">
<title>Node30</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1026.5,-235.5 1026.5,-254.5 1099.5,-254.5 1099.5,-235.5 1026.5,-235.5"/>
-<text text-anchor="middle" x="1063" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string_view</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1230.5,-235.5 1230.5,-254.5 1303.5,-254.5 1303.5,-235.5 1230.5,-235.5"/>
+<text text-anchor="middle" x="1267" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string_view</text>
</g>
<!-- Node26->Node30 -->
<g id="edge48" class="edge">
<title>Node26->Node30</title>
-<path fill="none" stroke="#191970" d="M1028.3708,-291.2977C1034.7561,-282.7609 1042.7641,-272.0545 1049.4705,-263.0883"/>
-<polygon fill="#191970" stroke="#191970" points="1052.4479,-264.9511 1055.6348,-254.8469 1046.8425,-260.7584 1052.4479,-264.9511"/>
+<path fill="none" stroke="#191970" d="M1309.1573,-291.2977C1301.141,-282.4941 1291.0238,-271.3833 1282.71,-262.2529"/>
+<polygon fill="#191970" stroke="#191970" points="1285.2869,-259.8845 1275.9663,-254.8469 1280.1112,-264.5974 1285.2869,-259.8845"/>
</g>
<!-- Node31 -->
<g id="node32" class="node">
<title>Node31</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="434.5,-235.5 434.5,-254.5 527.5,-254.5 527.5,-235.5 434.5,-235.5"/>
-<text text-anchor="middle" x="481" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">unordered_map</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="902.5,-235.5 902.5,-254.5 995.5,-254.5 995.5,-235.5 902.5,-235.5"/>
+<text text-anchor="middle" x="949" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">unordered_map</text>
</g>
<!-- Node26->Node31 -->
<g id="edge50" class="edge">
<title>Node26->Node31</title>
-<path fill="none" stroke="#191970" d="M953.8598,-292.4824C950.8701,-291.9525 947.9055,-291.4545 945,-291 783.9493,-265.8061 742.2033,-271.1911 580,-255 566.3706,-253.6395 551.6869,-252.1625 537.9096,-250.7716"/>
-<polygon fill="#191970" stroke="#191970" points="537.8843,-247.2513 527.5831,-249.7282 537.1805,-254.2159 537.8843,-247.2513"/>
+<path fill="none" stroke="#191970" d="M1259.9789,-296.1369C1188.8922,-284.4475 1074.3018,-265.6044 1005.8555,-254.3492"/>
+<polygon fill="#191970" stroke="#191970" points="1006.1268,-250.8469 995.6914,-252.6779 1004.991,-257.7542 1006.1268,-250.8469"/>
</g>
-<!-- Node32->Node16 -->
+<!-- Node32->Node15 -->
<g id="edge58" class="edge">
+<title>Node32->Node15</title>
+<path fill="none" stroke="#191970" d="M1674.0646,-370.6429C1866.0319,-361.824 2432.4796,-334.8533 2470,-322 2622.4949,-269.76 2770.3361,-135.4434 2819.1574,-88.2218"/>
+<polygon fill="#191970" stroke="#191970" points="2821.6651,-90.665 2826.3742,-81.1742 2816.7744,-85.6569 2821.6651,-90.665"/>
+</g>
+<!-- Node32->Node16 -->
+<g id="edge59" class="edge">
<title>Node32->Node16</title>
-<path fill="none" stroke="#191970" d="M315.0839,-291.4268C260.8302,-269.5739 177.2453,-226.4794 213,-179 254.8486,-123.4283 471.4027,-87.6977 554.4924,-75.8491"/>
-<polygon fill="#191970" stroke="#191970" points="554.98,-79.315 564.397,-74.4602 554.0079,-72.3829 554.98,-79.315"/>
+<path fill="none" stroke="#191970" d="M1674.0589,-371.1008C1774.0542,-366.4652 1974.8022,-353.7697 2142,-322 2155.2847,-319.4757 2365.5968,-263.6388 2376,-255 2424.5028,-214.7231 2418.3031,-129.512 2412.5632,-91.0566"/>
+<polygon fill="#191970" stroke="#191970" points="2416.0061,-90.4231 2410.9299,-81.1235 2409.0989,-91.5589 2416.0061,-90.4231"/>
</g>
<!-- Node32->Node18 -->
-<g id="edge59" class="edge">
+<g id="edge60" class="edge">
<title>Node32->Node18</title>
-<path fill="none" stroke="#191970" d="M420.24,-300.7562C456.2088,-297.6082 502.1046,-293.7907 543,-291 866.3621,-268.9335 1258.3405,-251.4186 1374.2139,-246.4314"/>
-<polygon fill="#191970" stroke="#191970" points="1374.4075,-249.9264 1384.2483,-246.0012 1374.1076,-242.9329 1374.4075,-249.9264"/>
+<path fill="none" stroke="#191970" d="M1674.0837,-370.6672C1861.2686,-362.1169 2402.7517,-336.2814 2438,-322 2442.2379,-320.283 2484.044,-283.7716 2509.1608,-261.6511"/>
+<polygon fill="#191970" stroke="#191970" points="2511.7757,-264.0116 2516.9604,-254.7721 2507.1455,-258.7617 2511.7757,-264.0116"/>
+</g>
+<!-- Node32->Node20 -->
+<g id="edge62" class="edge">
+<title>Node32->Node20</title>
+<path fill="none" stroke="#191970" d="M1630.0609,-358.2967C1641.2304,-349.3876 1655.5223,-337.9882 1667.8533,-328.1527"/>
+<polygon fill="#191970" stroke="#191970" points="1670.3792,-330.615 1676.0145,-321.6432 1666.0143,-325.1426 1670.3792,-330.615"/>
</g>
<!-- Node32->Node21 -->
-<g id="edge60" class="edge">
+<g id="edge61" class="edge">
<title>Node32->Node21</title>
-<path fill="none" stroke="#191970" d="M420.1472,-296.9581C505.7349,-284.0253 655.9027,-261.3341 725.9781,-250.7453"/>
-<polygon fill="#191970" stroke="#191970" points="726.7525,-254.1681 736.1173,-249.2132 725.7066,-247.2467 726.7525,-254.1681"/>
+<path fill="none" stroke="#191970" d="M1611.1957,-358.3894C1611.8655,-341.4469 1614.3404,-313.4114 1623,-291 1626.9255,-280.8407 1633.4327,-270.6801 1639.4744,-262.5021"/>
+<polygon fill="#191970" stroke="#191970" points="1642.3227,-264.541 1645.7035,-254.5001 1636.7989,-260.2412 1642.3227,-264.541"/>
</g>
<!-- Node32->Node24 -->
<g id="edge56" class="edge">
<title>Node32->Node24</title>
-<path fill="none" stroke="#191970" d="M343.2457,-291.2358C334.0692,-281.019 321.7593,-267.2499 311,-255 296.7393,-238.7635 280.6613,-220.1151 269.2258,-206.7919"/>
-<polygon fill="#191970" stroke="#191970" points="271.6128,-204.1987 262.4471,-198.8844 266.2983,-208.7546 271.6128,-204.1987"/>
+<path fill="none" stroke="#191970" d="M1593.8967,-358.4016C1555.1235,-324.1736 1460.799,-240.9063 1420.6205,-205.4377"/>
+<polygon fill="#191970" stroke="#191970" points="1422.7214,-202.6236 1412.9082,-198.6295 1418.0887,-207.8714 1422.7214,-202.6236"/>
</g>
<!-- Node32->Node29 -->
<g id="edge57" class="edge">
<title>Node32->Node29</title>
-<path fill="none" stroke="#191970" d="M355.0225,-291.2977C353.9814,-283.2945 352.6923,-273.3843 351.5742,-264.7889"/>
-<polygon fill="#191970" stroke="#191970" points="355.0417,-264.3119 350.2809,-254.8469 348.1002,-265.2149 355.0417,-264.3119"/>
+<path fill="none" stroke="#191970" d="M1547.8626,-359.4641C1544.8722,-358.9391 1541.9067,-358.4471 1539,-358 1369.1406,-331.8697 1323.3141,-351.4575 1154,-322 1147.3026,-320.8348 943.4644,-268.0754 893.9595,-255.2368"/>
+<polygon fill="#191970" stroke="#191970" points="894.6001,-251.7871 884.0414,-252.6618 892.8409,-258.5625 894.6001,-251.7871"/>
</g>
<!-- Node33->Node9 -->
-<g id="edge62" class="edge">
+<g id="edge64" class="edge">
<title>Node33->Node9</title>
-<path fill="none" stroke="#191970" d="M2397.49,-296.7781C2400.6591,-282.0869 2404.3552,-253.938 2392,-235 2319.4918,-123.8597 2156.973,-88.2669 2060.9921,-76.8689"/>
-<polygon fill="#191970" stroke="#191970" points="2061.1021,-73.3592 2050.7729,-75.7168 2060.3178,-80.3151 2061.1021,-73.3592"/>
+<path fill="none" stroke="#191970" d="M625.6395,-296.747C626.6158,-282.9475 628.718,-256.9572 632,-235 639.3845,-185.5966 651.8007,-128.5358 659.1147,-96.5684"/>
+<polygon fill="#191970" stroke="#191970" points="662.5453,-97.2677 661.3849,-86.7366 655.7247,-95.6928 662.5453,-97.2677"/>
</g>
<!-- Node33->Node13 -->
-<g id="edge63" class="edge">
+<g id="edge65" class="edge">
<title>Node33->Node13</title>
-<path fill="none" stroke="#191970" d="M2366.8816,-296.9974C2296.0704,-273.4247 2104.3158,-211.8037 1940,-179 1603.9344,-111.9085 1194.2852,-84.0849 1030.1404,-75.0834"/>
-<polygon fill="#191970" stroke="#191970" points="1029.8615,-71.5633 1019.6869,-74.5175 1029.4831,-78.553 1029.8615,-71.5633"/>
+<path fill="none" stroke="#191970" d="M628.0459,-296.6931C638.4625,-265.0034 675.9544,-164.9814 746,-123 786.6793,-98.6191 1089.1322,-81.5004 1230.9711,-74.7262"/>
+<polygon fill="#191970" stroke="#191970" points="1231.4878,-78.2058 1241.3112,-74.2371 1231.157,-71.2136 1231.4878,-78.2058"/>
</g>
<!-- Node33->Node14 -->
-<g id="edge64" class="edge">
+<g id="edge66" class="edge">
<title>Node33->Node14</title>
-<path fill="none" stroke="#191970" d="M2401.7118,-296.6285C2408.759,-286.2628 2420.1614,-269.4879 2430,-255 2446.8977,-230.1171 2456.5333,-226.8066 2468,-199 2482.893,-162.8849 2488.6221,-117.1834 2490.7684,-91.4615"/>
-<polygon fill="#191970" stroke="#191970" points="2494.2833,-91.3843 2491.5225,-81.1555 2487.302,-90.8734 2494.2833,-91.3843"/>
+<path fill="none" stroke="#191970" d="M608.9454,-296.9967C547.5964,-260.6818 327.6859,-130.5081 253.0146,-86.3071"/>
+<polygon fill="#191970" stroke="#191970" points="254.6426,-83.2036 244.2543,-81.1216 251.0769,-89.2274 254.6426,-83.2036"/>
</g>
<!-- Node33->Node15 -->
-<g id="edge65" class="edge">
+<g id="edge67" class="edge">
<title>Node33->Node15</title>
-<path fill="none" stroke="#191970" d="M2451.3073,-296.9768C2533.2122,-280.2326 2684.9127,-237.8573 2772,-143 2785.1113,-128.7189 2791.0845,-107.0397 2793.7907,-91.2974"/>
-<polygon fill="#191970" stroke="#191970" points="2797.3016,-91.4665 2795.2391,-81.0744 2790.3708,-90.4845 2797.3016,-91.4665"/>
+<path fill="none" stroke="#191970" d="M694.2519,-302.5394C756.2593,-299.101 849.677,-294.187 931,-291 1219.334,-279.7004 1943.3657,-295.9545 2229,-255 2422.7858,-227.2148 2468.9654,-203.9595 2655,-143 2708.5594,-125.4497 2769.173,-100.4118 2804.9052,-85.0966"/>
+<polygon fill="#191970" stroke="#191970" points="2806.4989,-88.2211 2814.2972,-81.0491 2803.7285,-81.7926 2806.4989,-88.2211"/>
</g>
<!-- Node34->Node6 -->
-<g id="edge68" class="edge">
+<g id="edge70" class="edge">
<title>Node34->Node6</title>
-<path fill="none" stroke="#191970" d="M2170.7894,-425.389C2172.0596,-401.0735 2174.5571,-353.2644 2175.9673,-326.2698"/>
-<polygon fill="#191970" stroke="#191970" points="2179.4718,-326.2732 2176.4983,-316.1042 2172.4813,-325.9079 2179.4718,-326.2732"/>
+<path fill="none" stroke="#191970" d="M2079.789,-434.2715C2157.4693,-425.5049 2301.13,-408.4108 2423,-389 2559.875,-367.1992 2719.7728,-335.0668 2801.9836,-318.0527"/>
+<polygon fill="#191970" stroke="#191970" points="2802.777,-321.4627 2811.8576,-316.0045 2801.3552,-314.6086 2802.777,-321.4627"/>
</g>
<!-- Node34->Node14 -->
-<g id="edge90" class="edge">
+<g id="edge92" class="edge">
<title>Node34->Node14</title>
-<path fill="none" stroke="#191970" d="M2226.6423,-435.2144C2296.2359,-425.5014 2412.4967,-398.1933 2473,-322 2527.6227,-253.2124 2508.0613,-137.6222 2497.2311,-91.4405"/>
-<polygon fill="#191970" stroke="#191970" points="2500.5714,-90.3676 2494.7767,-81.4979 2493.7754,-92.0453 2500.5714,-90.3676"/>
+<path fill="none" stroke="#191970" d="M1966.4185,-438.6878C1726.401,-430.9415 801.9138,-400.4841 741,-389 526.5289,-348.5657 486.9629,-278.6631 316,-143 292.2798,-124.1774 265.2919,-102.159 247.5281,-87.5817"/>
+<polygon fill="#191970" stroke="#191970" points="249.5516,-84.7144 239.6034,-81.0699 245.1075,-90.1228 249.5516,-84.7144"/>
</g>
<!-- Node34->Node33 -->
-<g id="edge69" class="edge">
+<g id="edge71" class="edge">
<title>Node34->Node33</title>
-<path fill="none" stroke="#191970" d="M2196.8961,-425.3773C2214.844,-415.2201 2238.9273,-401.4645 2260,-389 2299.1237,-365.8583 2344.0794,-338.1607 2370.978,-321.4678"/>
-<polygon fill="#191970" stroke="#191970" points="2373.0537,-324.2987 2379.7001,-316.0479 2369.3591,-318.3531 2373.0537,-324.2987"/>
+<path fill="none" stroke="#191970" d="M1966.2629,-438.5341C1738.0351,-430.5608 896.5952,-400.5126 841,-389 770.7323,-374.449 693.2765,-339.9924 652.9338,-320.5248"/>
+<polygon fill="#191970" stroke="#191970" points="654.2583,-317.2766 643.7365,-316.0346 651.1873,-323.567 654.2583,-317.2766"/>
</g>
<!-- Node34->Node35 -->
-<g id="edge70" class="edge">
+<g id="edge72" class="edge">
<title>Node34->Node35</title>
-<path fill="none" stroke="#191970" d="M2113.4852,-433.488C2014.9701,-421.2648 1811.9226,-396.0719 1702.9379,-382.5497"/>
-<polygon fill="#191970" stroke="#191970" points="1703.0462,-379.0364 1692.6913,-381.2784 1702.1843,-385.9831 1703.0462,-379.0364"/>
+<path fill="none" stroke="#191970" d="M2025.2692,-425.2967C2026.6911,-415.7699 2028.538,-403.3954 2030.0701,-393.1306"/>
+<polygon fill="#191970" stroke="#191970" points="2033.5593,-393.4621 2031.5739,-383.055 2026.636,-392.4287 2033.5593,-393.4621"/>
</g>
<!-- Node40 -->
<g id="node41" class="node">
<title>Node40</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1924.5,-297 1924.5,-316 1989.5,-316 1989.5,-297 1924.5,-297"/>
-<text text-anchor="middle" x="1957" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">functional</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2604.5,-297 2604.5,-316 2669.5,-316 2669.5,-297 2604.5,-297"/>
+<text text-anchor="middle" x="2637" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">functional</text>
</g>
<!-- Node34->Node40 -->
-<g id="edge89" class="edge">
+<g id="edge91" class="edge">
<title>Node34->Node40</title>
-<path fill="none" stroke="#191970" d="M2145.9802,-425.389C2104.9138,-399.5538 2021.6892,-347.1965 1980.7845,-321.463"/>
-<polygon fill="#191970" stroke="#191970" points="1982.5944,-318.4667 1972.2663,-316.1042 1978.8669,-324.3917 1982.5944,-318.4667"/>
+<path fill="none" stroke="#191970" d="M2079.6104,-433.5582C2144.9223,-425.0877 2255.3517,-409.3446 2349,-389 2457.6392,-365.3987 2483.6246,-354.3074 2590,-322 2592.9468,-321.105 2595.9898,-320.1568 2599.0443,-319.1878"/>
+<polygon fill="#191970" stroke="#191970" points="2600.2931,-322.4626 2608.7358,-316.0617 2598.1442,-315.8006 2600.2931,-322.4626"/>
</g>
<!-- Node35->Node8 -->
-<g id="edge79" class="edge">
+<g id="edge81" class="edge">
<title>Node35->Node8</title>
-<path fill="none" stroke="#191970" d="M1624.5645,-363.8518C1614.6702,-347.0397 1592.2073,-312.1531 1565,-291 1533.8986,-266.8193 1513.5684,-281.0329 1484,-255 1450.9903,-225.9373 1428.1749,-178.5083 1417.2903,-152.1654"/>
-<polygon fill="#191970" stroke="#191970" points="1420.4255,-150.5777 1413.4754,-142.581 1413.9218,-153.1665 1420.4255,-150.5777"/>
+<path fill="none" stroke="#191970" d="M2070.9087,-363.9347C2141.5746,-344.377 2281.7001,-296.4278 2229,-235 2194.7428,-195.0695 2045.3543,-214.2461 1995,-199 1974.8272,-192.8922 1971.7051,-186.4803 1952,-179 1916.9609,-165.6987 1876.2965,-153.6921 1845.5554,-145.2832"/>
+<polygon fill="#191970" stroke="#191970" points="1846.1,-141.8048 1835.533,-142.5728 1844.2725,-148.5621 1846.1,-141.8048"/>
</g>
<!-- Node35->Node9 -->
-<g id="edge71" class="edge">
+<g id="edge73" class="edge">
<title>Node35->Node9</title>
-<path fill="none" stroke="#191970" d="M1672.3272,-363.9989C1715.8816,-353.6578 1780.2141,-336.6395 1801,-322 1888.0217,-260.7107 1950.6105,-145.5371 1974.8794,-95.6061"/>
-<polygon fill="#191970" stroke="#191970" points="1978.0651,-97.057 1979.2181,-86.5251 1971.749,-94.0392 1978.0651,-97.057"/>
+<path fill="none" stroke="#191970" d="M1970.3066,-370.4875C1797.1181,-362.0081 1320.3261,-337.5459 1251,-322 1171.7561,-304.2301 1155.9448,-286.2831 1081,-255 943.2066,-197.4829 781.9178,-124.7777 707.0878,-90.7328"/>
+<polygon fill="#191970" stroke="#191970" points="708.5208,-87.5396 697.9694,-86.5808 705.62,-93.9102 708.5208,-87.5396"/>
</g>
<!-- Node35->Node16 -->
-<g id="edge87" class="edge">
+<g id="edge89" class="edge">
<title>Node35->Node16</title>
-<path fill="none" stroke="#191970" d="M1567.3408,-368.8153C1524.0061,-365.6267 1465.0119,-361.3872 1413,-358 1277.0529,-349.1467 932.6661,-356.9863 801,-322 768.6317,-313.3991 691.7449,-276.4214 666,-255 642.0848,-235.1011 639.7731,-225.8163 624,-199 610.1812,-175.5063 606.8745,-169.0942 599,-143 593.8926,-126.0753 590.7774,-106.1996 589.0057,-91.7055"/>
-<polygon fill="#191970" stroke="#191970" points="592.4431,-90.947 587.8605,-81.3945 585.4858,-91.7198 592.4431,-90.947"/>
+<path fill="none" stroke="#191970" d="M2095.5561,-367.8945C2219.5616,-356.6125 2486.1934,-331.4382 2504,-322 2544.8438,-300.3512 2577.7929,-278.0684 2561,-235 2535.041,-168.4237 2467.6303,-112.9369 2431.9413,-87.0997"/>
+<polygon fill="#191970" stroke="#191970" points="2433.7775,-84.1111 2423.5958,-81.181 2429.728,-89.8209 2433.7775,-84.1111"/>
</g>
<!-- Node35->Node17 -->
-<g id="edge85" class="edge">
+<g id="edge87" class="edge">
<title>Node35->Node17</title>
-<path fill="none" stroke="#191970" d="M1628.1288,-363.8518C1624.4116,-347.0397 1614.6126,-312.1532 1594,-291 1567.635,-263.9436 1541.5353,-283.7259 1517,-255 1477.3507,-208.5786 1500.855,-180.0032 1479,-123 1474.7221,-111.8422 1468.9479,-99.8051 1464.0266,-90.1759"/>
-<polygon fill="#191970" stroke="#191970" points="1467.0419,-88.3891 1459.3053,-81.1507 1460.8394,-91.6339 1467.0419,-88.3891"/>
+<path fill="none" stroke="#191970" d="M2087.7458,-363.9193C2135.5747,-354.5846 2200.3484,-339.2999 2220,-322 2262.8485,-284.2791 2252.4564,-255.9054 2257,-199 2260.038,-160.9506 2248.1853,-145.0105 2217,-123 2173.1862,-92.0763 2111.082,-79.6212 2072.4618,-74.6738"/>
+<polygon fill="#191970" stroke="#191970" points="2072.5672,-71.1621 2062.2264,-73.4683 2071.7484,-78.114 2072.5672,-71.1621"/>
</g>
<!-- Node35->Node18 -->
-<g id="edge88" class="edge">
+<g id="edge90" class="edge">
<title>Node35->Node18</title>
-<path fill="none" stroke="#191970" d="M1620.5994,-363.8875C1603.7711,-347.1309 1566.805,-312.3272 1530,-291 1501.9415,-274.7411 1467.0802,-262.2738 1441.8171,-254.4463"/>
-<polygon fill="#191970" stroke="#191970" points="1442.5491,-251.0115 1431.9645,-251.4774 1440.5294,-257.7138 1442.5491,-251.0115"/>
+<path fill="none" stroke="#191970" d="M2095.8171,-369.1063C2210.3264,-360.6729 2444.2023,-341.3329 2476,-322 2497.994,-308.6277 2512.7729,-282.0091 2520.786,-263.8531"/>
+<polygon fill="#191970" stroke="#191970" points="2524.05,-265.1187 2524.6352,-254.54 2517.5808,-262.4448 2524.05,-265.1187"/>
</g>
<!-- Node35->Node20 -->
-<g id="edge72" class="edge">
+<g id="edge74" class="edge">
<title>Node35->Node20</title>
-<path fill="none" stroke="#191970" d="M1567.344,-368.7656C1524.0107,-365.5548 1465.0169,-361.3092 1413,-358 1126.4664,-339.7715 1054.2318,-344.475 768,-322 742.0333,-319.9611 713.6379,-317.2251 688.4359,-314.6195"/>
-<polygon fill="#191970" stroke="#191970" points="688.5252,-311.1099 678.2157,-313.5522 687.798,-318.0721 688.5252,-311.1099"/>
+<path fill="none" stroke="#191970" d="M1984.9318,-363.9717C1928.3304,-352.7519 1833.7235,-333.9984 1768.0406,-320.9785"/>
+<polygon fill="#191970" stroke="#191970" points="1768.6526,-317.5317 1758.1629,-319.0204 1767.2914,-324.3981 1768.6526,-317.5317"/>
</g>
<!-- Node35->Node26 -->
-<g id="edge77" class="edge">
+<g id="edge79" class="edge">
<title>Node35->Node26</title>
-<path fill="none" stroke="#191970" d="M1567.279,-366.6447C1453.0114,-354.1554 1211.5296,-327.7618 1090.0461,-314.4838"/>
-<polygon fill="#191970" stroke="#191970" points="1090.4116,-311.003 1080.0905,-313.3957 1089.651,-317.9616 1090.4116,-311.003"/>
+<path fill="none" stroke="#191970" d="M1970.3037,-367.5836C1839.1338,-355.2056 1536.347,-326.6327 1396.3876,-313.4253"/>
+<polygon fill="#191970" stroke="#191970" points="1396.4759,-309.9182 1386.1913,-312.4631 1395.8182,-316.8872 1396.4759,-309.9182"/>
</g>
<!-- Node35->Node33 -->
-<g id="edge78" class="edge">
+<g id="edge80" class="edge">
<title>Node35->Node33</title>
-<path fill="none" stroke="#191970" d="M1692.9194,-368.5352C1803.7314,-359.7158 2040.9243,-340.4995 2241,-322 2265.2319,-319.7595 2291.5917,-317.1644 2315.5414,-314.746"/>
-<polygon fill="#191970" stroke="#191970" points="2316.0487,-318.2126 2325.6447,-313.7219 2315.3427,-311.2483 2316.0487,-318.2126"/>
+<path fill="none" stroke="#191970" d="M1970.2498,-370.6536C1801.9555,-363.0008 1326.516,-341.2589 931,-322 853.9976,-318.2505 766.1062,-313.7785 704.3355,-310.6039"/>
+<polygon fill="#191970" stroke="#191970" points="704.435,-307.1045 694.2684,-310.0861 704.0754,-314.0952 704.435,-307.1045"/>
</g>
<!-- Node36 -->
<g id="node37" class="node">
<title>Node36</title>
<g id="a_node37"><a xlink:href="shape__tuple_8h.html" target="_top" xlink:title="Runtime ShapeTuple container types. ">
-<polygon fill="#ffffff" stroke="#000000" points="810,-291.5 810,-321.5 936,-321.5 936,-291.5 810,-291.5"/>
-<text text-anchor="start" x="818" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="873" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/shape_tuple.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2270,-291.5 2270,-321.5 2396,-321.5 2396,-291.5 2270,-291.5"/>
+<text text-anchor="start" x="2278" y="-309.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
+<text text-anchor="middle" x="2333" y="-298.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/shape_tuple.h</text>
</a>
</g>
</g>
<!-- Node35->Node36 -->
-<g id="edge73" class="edge">
+<g id="edge75" class="edge">
<title>Node35->Node36</title>
-<path fill="none" stroke="#191970" d="M1567.3803,-369.0259C1413.2943,-357.9872 1023.2151,-329.8518 946.4479,-321.8224"/>
-<polygon fill="#191970" stroke="#191970" points="946.4695,-318.3007 936.1276,-320.6017 945.6471,-325.2522 946.4695,-318.3007"/>
+<path fill="none" stroke="#191970" d="M2075.6641,-363.9717C2123.6046,-353.265 2202.263,-335.6979 2260.0175,-322.7994"/>
+<polygon fill="#191970" stroke="#191970" points="2260.8245,-326.2055 2269.8212,-320.6099 2259.2987,-319.3738 2260.8245,-326.2055"/>
</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...">
-<polygon fill="#ffffff" stroke="#000000" points="1658,-297 1658,-316 1792,-316 1792,-297 1658,-297"/>
-<text text-anchor="middle" x="1725" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/serializer.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1966,-297 1966,-316 2100,-316 2100,-297 1966,-297"/>
+<text text-anchor="middle" x="2033" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/serializer.h</text>
</a>
</g>
</g>
<!-- Node35->Node37 -->
-<g id="edge80" class="edge">
+<g id="edge82" class="edge">
<title>Node35->Node37</title>
-<path fill="none" stroke="#191970" d="M1638.7106,-363.9005C1652.2322,-352.8554 1677.8652,-334.5776 1697.8148,-321.6172"/>
-<polygon fill="#191970" stroke="#191970" points="1699.9579,-324.4021 1706.5169,-316.0817 1696.2008,-318.4958 1699.9579,-324.4021"/>
+<path fill="none" stroke="#191970" d="M2028.0994,-363.9005C2026.2106,-354.149 2025.7686,-338.7597 2026.7734,-326.3695"/>
+<polygon fill="#191970" stroke="#191970" points="2030.2923,-326.4478 2028.1029,-316.0817 2023.35,-325.5506 2030.2923,-326.4478"/>
</g>
<!-- Node35->Node40 -->
-<g id="edge86" class="edge">
+<g id="edge88" class="edge">
<title>Node35->Node40</title>
-<path fill="none" stroke="#191970" d="M1692.5303,-364.5073C1749.6274,-355.7045 1836.134,-340.8788 1910,-322 1913.348,-321.1443 1916.804,-320.1749 1920.2516,-319.1487"/>
-<polygon fill="#191970" stroke="#191970" points="1921.3453,-322.4739 1929.8482,-316.1531 1919.2595,-315.7919 1921.3453,-322.4739"/>
+<path fill="none" stroke="#191970" d="M2095.5723,-369.506C2220.529,-361.2356 2496.9474,-341.3841 2590,-322 2593.864,-321.1951 2597.8547,-320.1799 2601.8021,-319.0559"/>
+<polygon fill="#191970" stroke="#191970" points="2602.8993,-322.3804 2611.4287,-316.0954 2600.8416,-315.6896 2602.8993,-322.3804"/>
</g>
<!-- Node36->Node16 -->
-<g id="edge74" class="edge">
+<g id="edge76" class="edge">
<title>Node36->Node16</title>
-<path fill="none" stroke="#191970" d="M809.7255,-292.1681C766.4412,-281.6399 713.4996,-267.1201 694,-255 632.8642,-217.0008 602.7514,-130.2692 591.8952,-91.2098"/>
-<polygon fill="#191970" stroke="#191970" points="595.2235,-90.1054 589.2828,-81.3328 588.4562,-91.8953 595.2235,-90.1054"/>
+<path fill="none" stroke="#191970" d="M2381.7109,-291.4836C2410.7789,-281.4285 2443.6268,-267.7474 2452,-255 2464.5597,-235.8791 2498.4901,-288.2153 2442,-123 2437.8344,-110.8169 2430.2119,-98.6215 2423.3415,-89.1585"/>
+<polygon fill="#191970" stroke="#191970" points="2425.9726,-86.837 2417.1224,-81.0126 2420.4088,-91.0848 2425.9726,-86.837"/>
</g>
<!-- Node36->Node18 -->
-<g id="edge75" class="edge">
+<g id="edge77" class="edge">
<title>Node36->Node18</title>
-<path fill="none" stroke="#191970" d="M936.1406,-292.485C939.1301,-291.9544 942.0946,-291.4556 945,-291 1104.6761,-265.9627 1297.5914,-251.9624 1374.2414,-247.0413"/>
-<polygon fill="#191970" stroke="#191970" points="1374.5108,-250.5314 1384.2698,-246.407 1374.0688,-243.5453 1374.5108,-250.5314"/>
+<path fill="none" stroke="#191970" d="M2380.7024,-291.4554C2416.1085,-280.2888 2463.4909,-265.3452 2494.841,-255.4578"/>
+<polygon fill="#191970" stroke="#191970" points="2495.9366,-258.7823 2504.4208,-252.4365 2493.8311,-252.1065 2495.9366,-258.7823"/>
</g>
<!-- Node36->Node21 -->
-<g id="edge76" class="edge">
+<g id="edge78" class="edge">
<title>Node36->Node21</title>
-<path fill="none" stroke="#191970" d="M846.3356,-291.4554C829.1059,-281.734 806.8025,-269.15 789.7554,-259.5317"/>
-<polygon fill="#191970" stroke="#191970" points="791.3204,-256.3961 780.8911,-254.5303 787.8806,-262.4926 791.3204,-256.3961"/>
+<path fill="none" stroke="#191970" d="M2269.7311,-300.7695C2130.8352,-288.1891 1802.7693,-258.4747 1691.5957,-248.4052"/>
+<polygon fill="#191970" stroke="#191970" points="1691.8027,-244.9097 1681.5277,-247.4933 1691.1712,-251.8812 1691.8027,-244.9097"/>
</g>
<!-- Node37->Node9 -->
-<g id="edge83" class="edge">
+<g id="edge85" class="edge">
<title>Node37->Node9</title>
-<path fill="none" stroke="#191970" d="M1733.5159,-296.6462C1758.5649,-267.966 1834.8395,-182.7887 1908,-123 1921.6055,-111.8812 1937.6562,-100.9547 1951.6494,-92.0469"/>
-<polygon fill="#191970" stroke="#191970" points="1953.8841,-94.7772 1960.5015,-86.5031 1950.1686,-88.8446 1953.8841,-94.7772"/>
+<path fill="none" stroke="#191970" d="M2045.7058,-296.9865C2062.4284,-283.2576 2088.2364,-257.1103 2075,-235 2020.8691,-144.5786 1967.6406,-150.8454 1866,-123 1811.6525,-108.111 993.9637,-81.6823 739.7999,-73.7879"/>
+<polygon fill="#191970" stroke="#191970" points="739.6627,-70.2821 729.559,-73.4704 739.4457,-77.2787 739.6627,-70.2821"/>
</g>
<!-- Node37->Node35 -->
-<g id="edge84" class="edge">
+<g id="edge86" class="edge">
<title>Node37->Node35</title>
-<path fill="none" stroke="#191970" d="M1716.3112,-316.0817C1702.8026,-327.119 1677.1731,-345.3957 1657.2174,-358.3619"/>
-<polygon fill="#191970" stroke="#191970" points="1655.0702,-355.5796 1648.5118,-363.9005 1658.8277,-361.4857 1655.0702,-355.5796"/>
+<path fill="none" stroke="#191970" d="M2037.8971,-316.0817C2039.7877,-325.8263 2040.2315,-341.214 2039.2285,-353.6079"/>
+<polygon fill="#191970" stroke="#191970" points="2035.709,-353.5349 2037.9006,-363.9005 2042.6514,-354.4306 2035.709,-353.5349"/>
</g>
<!-- Node38 -->
<g id="node39" class="node">
<title>Node38</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1526,-235.5 1526,-254.5 1588,-254.5 1588,-235.5 1526,-235.5"/>
-<text text-anchor="middle" x="1557" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/io.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2004,-235.5 2004,-254.5 2066,-254.5 2066,-235.5 2004,-235.5"/>
+<text text-anchor="middle" x="2035" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/io.h</text>
</g>
<!-- Node37->Node38 -->
-<g id="edge81" class="edge">
+<g id="edge83" class="edge">
<title>Node37->Node38</title>
-<path fill="none" stroke="#191970" d="M1698.75,-296.8906C1670.2707,-286.4652 1624.6257,-269.7559 1592.8543,-258.1252"/>
-<polygon fill="#191970" stroke="#191970" points="1593.6594,-254.6929 1583.0657,-254.5419 1591.2531,-261.2663 1593.6594,-254.6929"/>
+<path fill="none" stroke="#191970" d="M2033.3125,-296.8906C2033.5897,-288.3657 2034.0036,-275.6392 2034.3521,-264.9235"/>
+<polygon fill="#191970" stroke="#191970" points="2037.8561,-264.8534 2034.6831,-254.7449 2030.8598,-264.6258 2037.8561,-264.8534"/>
</g>
<!-- Node39 -->
<g id="node40" class="node">
<title>Node39</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1644.5,-235.5 1644.5,-254.5 1741.5,-254.5 1741.5,-235.5 1644.5,-235.5"/>
-<text text-anchor="middle" x="1693" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/serializer.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2122.5,-235.5 2122.5,-254.5 2219.5,-254.5 2219.5,-235.5 2122.5,-235.5"/>
+<text text-anchor="middle" x="2171" y="-242.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/serializer.h</text>
</g>
<!-- Node37->Node39 -->
-<g id="edge82" class="edge">
+<g id="edge84" class="edge">
<title>Node37->Node39</title>
-<path fill="none" stroke="#191970" d="M1720,-296.8906C1715.4242,-288.0965 1708.5219,-274.8312 1702.8417,-263.9145"/>
-<polygon fill="#191970" stroke="#191970" points="1705.7912,-262.0003 1698.0705,-254.7449 1699.5815,-265.2314 1705.7912,-262.0003"/>
+<path fill="none" stroke="#191970" d="M2054.5625,-296.8906C2077.5512,-286.6457 2114.156,-270.3326 2140.1824,-258.7339"/>
+<polygon fill="#191970" stroke="#191970" points="2141.8796,-261.8095 2149.5889,-254.5419 2139.0302,-255.4156 2141.8796,-261.8095"/>
</g>
<!-- Node41->Node8 -->
-<g id="edge120" class="edge">
+<g id="edge122" class="edge">
<title>Node41->Node8</title>
-<path fill="none" stroke="#191970" d="M1356.7211,-425.2756C1354.5571,-408.706 1351.9202,-381.4406 1354,-358 1358.9081,-302.6831 1362.2945,-288.8164 1376,-235 1383.4584,-205.7137 1395.0819,-172.6737 1402.6191,-152.3415"/>
-<polygon fill="#191970" stroke="#191970" points="1405.9985,-153.2981 1406.239,-142.706 1399.4457,-150.8362 1405.9985,-153.2981"/>
+<path fill="none" stroke="#191970" d="M1525.2724,-433.6673C1570.1322,-426.7671 1632.7554,-413.4989 1683,-389 1725.9239,-368.0707 1742.8568,-363.2019 1767,-322 1799.0065,-267.3787 1800.5518,-189.4623 1799.2418,-153.114"/>
+<polygon fill="#191970" stroke="#191970" points="1802.725,-152.6768 1798.7512,-142.8555 1795.733,-153.0113 1802.725,-152.6768"/>
</g>
<!-- Node41->Node9 -->
-<g id="edge97" class="edge">
+<g id="edge99" class="edge">
<title>Node41->Node9</title>
-<path fill="none" stroke="#191970" d="M1417.0433,-436.0806C1496.7523,-429.3572 1644.2967,-414.5628 1768,-389 1872.6861,-367.3671 1934.7973,-407.4703 1999,-322 2049.9117,-254.2235 2015.884,-144.6881 1996.6523,-96.0838"/>
-<polygon fill="#191970" stroke="#191970" points="1999.8303,-94.6081 1992.8089,-86.6739 1993.35,-97.255 1999.8303,-94.6081"/>
+<path fill="none" stroke="#191970" d="M1408.8838,-426.5557C1405.5502,-425.9735 1402.2385,-425.4477 1399,-425 1086.0163,-381.7277 995.4671,-465.8713 689,-389 621.3134,-372.0221 590.9195,-376.2292 547,-322 506.1859,-271.6052 488.6487,-236.8267 518,-179 538.9522,-137.7208 584.0233,-108.4953 618.7705,-91.0384"/>
+<polygon fill="#191970" stroke="#191970" points="620.3517,-94.1614 627.8138,-86.6402 617.2901,-87.8664 620.3517,-94.1614"/>
</g>
<!-- Node41->Node13 -->
-<g id="edge106" class="edge">
+<g id="edge108" class="edge">
<title>Node41->Node13</title>
-<path fill="none" stroke="#191970" d="M1300.9027,-439.1231C1087.9758,-433.8834 360.4784,-414.2058 318,-389 292.651,-373.9584 220.082,-227.9535 213,-199 210.8881,-190.3656 207.1814,-185.7198 213,-179 256.6825,-128.5522 705.4685,-90.0959 884.1654,-76.6714"/>
-<polygon fill="#191970" stroke="#191970" points="884.5502,-80.1525 894.262,-75.918 884.0292,-73.1719 884.5502,-80.1525"/>
+<path fill="none" stroke="#191970" d="M1408.8762,-426.6092C1405.5446,-426.0133 1402.2353,-425.47 1399,-425 1337.5165,-416.0678 893.5922,-422.0766 841,-389 778.8412,-349.9068 760,-318.4302 760,-245 760,-245 760,-245 760,-189 760,-141.3788 1082.6147,-97.4206 1230.9581,-79.7251"/>
+<polygon fill="#191970" stroke="#191970" points="1231.8118,-83.1485 1241.3308,-78.4967 1230.9885,-76.1971 1231.8118,-83.1485"/>
</g>
<!-- Node41->Node14 -->
-<g id="edge124" class="edge">
+<g id="edge126" class="edge">
<title>Node41->Node14</title>
-<path fill="none" stroke="#191970" d="M1417.322,-435.8411C1543.597,-425.5909 1834.4745,-401.0405 1878,-389 2005.4538,-353.7423 2368.5237,-143.8761 2467.0631,-86.1739"/>
-<polygon fill="#191970" stroke="#191970" points="2468.9009,-89.1536 2475.7566,-81.0758 2465.3599,-83.1153 2468.9009,-89.1536"/>
+<path fill="none" stroke="#191970" d="M1408.8873,-426.5302C1405.5528,-425.9544 1402.2399,-425.4371 1399,-425 1219.1199,-400.7323 759.4566,-431.5305 583,-389 443.7919,-355.4473 399.358,-347.442 290,-255 240.4277,-213.0958 230.3626,-130.055 228.4104,-91.7389"/>
+<polygon fill="#191970" stroke="#191970" points="231.8946,-91.2435 228.0302,-81.3786 224.8993,-91.5003 231.8946,-91.2435"/>
</g>
<!-- Node41->Node15 -->
-<g id="edge126" class="edge">
+<g id="edge128" class="edge">
<title>Node41->Node15</title>
-<path fill="none" stroke="#191970" d="M1417.13,-436.5217C1463.6281,-433.3542 1530.4944,-428.832 1589,-425 1983.7011,-399.148 2095.3672,-439.6891 2473,-322 2594.6611,-284.0844 2629.246,-274.37 2732,-199 2761.4129,-177.4257 2776.5331,-176.0355 2792,-143 2799.5615,-126.8495 2799.9053,-106.3506 2798.7457,-91.4466"/>
-<polygon fill="#191970" stroke="#191970" points="2802.2047,-90.8754 2797.6606,-81.3045 2795.2444,-91.6202 2802.2047,-90.8754"/>
+<path fill="none" stroke="#191970" d="M1525.3273,-438.8648C1724.2266,-433.1231 2375.0209,-412.9315 2583,-389 2784.478,-365.8165 3016,-447.8075 3016,-245 3016,-245 3016,-245 3016,-189 3016,-124.4909 2933.9311,-93.5263 2880.6244,-80.2177"/>
+<polygon fill="#191970" stroke="#191970" points="2881.2441,-76.7675 2870.7058,-77.8603 2879.6254,-83.5778 2881.2441,-76.7675"/>
</g>
<!-- Node41->Node16 -->
-<g id="edge127" class="edge">
+<g id="edge129" class="edge">
<title>Node41->Node16</title>
-<path fill="none" stroke="#191970" d="M1300.9139,-439.0578C1084.4797,-433.5089 334.9927,-412.6618 290,-389 182.3792,-332.402 123.6664,-188.6765 226,-123 280.0663,-88.3009 475.3002,-76.2514 553.8878,-72.7477"/>
-<polygon fill="#191970" stroke="#191970" points="554.3343,-76.2319 564.1759,-72.3086 554.0357,-69.2383 554.3343,-76.2319"/>
+<path fill="none" stroke="#191970" d="M1525.0205,-437.9762C1789.8869,-426.1783 2873.3758,-374.8106 2921,-322 3067.9543,-159.042 2576.1809,-90.4427 2441.9421,-75.002"/>
+<polygon fill="#191970" stroke="#191970" points="2442.1716,-71.5058 2431.8425,-73.8637 2441.3876,-78.4617 2442.1716,-71.5058"/>
</g>
<!-- Node41->Node18 -->
-<g id="edge128" class="edge">
+<g id="edge130" class="edge">
<title>Node41->Node18</title>
-<path fill="none" stroke="#191970" d="M1382.222,-425.322C1393.8101,-416.3119 1406.6771,-403.7675 1413,-389 1430.7319,-347.5858 1421.0409,-293.1751 1413.7349,-264.5496"/>
-<polygon fill="#191970" stroke="#191970" points="1417.0301,-263.3299 1411.0213,-254.6038 1410.277,-265.1725 1417.0301,-263.3299"/>
+<path fill="none" stroke="#191970" d="M1525.3829,-439.4945C1757.3268,-434.883 2604.6798,-411.9451 2679,-322 2687.7761,-311.3789 2687.2493,-302.0352 2679,-291 2647.0682,-248.2845 2614.732,-269.4475 2561.5244,-256.0766"/>
+<polygon fill="#191970" stroke="#191970" points="2562.287,-252.6546 2551.7103,-253.2734 2560.3645,-259.3854 2562.287,-252.6546"/>
</g>
<!-- Node41->Node29 -->
-<g id="edge123" class="edge">
+<g id="edge125" class="edge">
<title>Node41->Node29</title>
-<path fill="none" stroke="#191970" d="M1300.8191,-439.5213C1101.9852,-435.8947 456.9504,-421.7061 370,-389 347.4885,-380.5324 295.4969,-343.6399 285,-322 278.9869,-309.6037 278.7136,-303.26 285,-291 291.933,-277.4789 304.8936,-267.0326 317.3481,-259.5377"/>
-<polygon fill="#191970" stroke="#191970" points="319.1928,-262.5168 326.234,-254.6001 315.7928,-256.3979 319.1928,-262.5168"/>
+<path fill="none" stroke="#191970" d="M1408.7679,-427.1964C1405.4639,-426.4509 1402.1902,-425.7146 1399,-425 1347.0432,-413.3611 977.6167,-347.7267 931,-322 904.5328,-307.3934 881.3057,-280.7183 867.706,-262.9228"/>
+<polygon fill="#191970" stroke="#191970" points="870.2652,-260.4971 861.5056,-254.5373 864.6368,-264.6589 870.2652,-260.4971"/>
</g>
<!-- Node41->Node32 -->
-<g id="edge98" class="edge">
+<g id="edge100" class="edge">
<title>Node41->Node32</title>
-<path fill="none" stroke="#191970" d="M1300.8926,-438.9784C1098.8154,-433.4801 436.9062,-413.6592 399,-389 379.2209,-376.1331 368.3147,-350.891 362.6094,-331.6995"/>
-<polygon fill="#191970" stroke="#191970" points="365.9199,-330.531 359.9516,-321.7771 359.1583,-332.3422 365.9199,-330.531"/>
+<path fill="none" stroke="#191970" d="M1499.3164,-425.4639C1519.9225,-415.8763 1546.8351,-403.3545 1569.0783,-393.0052"/>
+<polygon fill="#191970" stroke="#191970" points="1570.7031,-396.1096 1578.2932,-388.7177 1567.7501,-389.763 1570.7031,-396.1096"/>
</g>
<!-- Node41->Node33 -->
-<g id="edge105" class="edge">
+<g id="edge107" class="edge">
<title>Node41->Node33</title>
-<path fill="none" stroke="#191970" d="M1417.0983,-436.5722C1581.1236,-425.4457 2037.9682,-394.1841 2070,-389 2174.5152,-372.085 2294.5648,-337.5636 2355.3232,-318.9904"/>
-<polygon fill="#191970" stroke="#191970" points="2356.3774,-322.328 2364.9063,-316.0425 2354.3192,-315.6374 2356.3774,-322.328"/>
+<path fill="none" stroke="#191970" d="M1408.882,-426.5684C1405.5489,-425.9829 1402.2377,-425.453 1399,-425 1251.9941,-404.4311 871.1631,-440.6474 732,-389 729.4856,-388.0668 675.9227,-346.3367 645.5554,-322.5944"/>
+<polygon fill="#191970" stroke="#191970" points="647.3448,-319.5506 637.3119,-316.1458 643.0318,-325.064 647.3448,-319.5506"/>
</g>
<!-- Node41->Node35 -->
-<g id="edge119" class="edge">
+<g id="edge121" class="edge">
<title>Node41->Node35</title>
-<path fill="none" stroke="#191970" d="M1417.1363,-426.1268C1466.2551,-413.9831 1535.7712,-396.7964 1581.8307,-385.409"/>
-<polygon fill="#191970" stroke="#191970" points="1582.7058,-388.7981 1591.5734,-383.0003 1581.0257,-382.0027 1582.7058,-388.7981"/>
+<path fill="none" stroke="#191970" d="M1525.1758,-433.6135C1629.0952,-421.3121 1846.3629,-395.5931 1960.0753,-382.1324"/>
+<polygon fill="#191970" stroke="#191970" points="1960.5631,-385.5992 1970.0823,-380.9479 1959.7401,-378.6478 1960.5631,-385.5992"/>
</g>
<!-- Node41->Node40 -->
-<g id="edge121" class="edge">
+<g id="edge123" class="edge">
<title>Node41->Node40</title>
-<path fill="none" stroke="#191970" d="M1417.1992,-434.6817C1485.8731,-427.146 1603.2154,-412.1638 1702,-389 1796.5572,-366.8275 1817.6594,-352.0982 1910,-322 1912.757,-321.1014 1915.604,-320.1703 1918.4693,-319.2309"/>
-<polygon fill="#191970" stroke="#191970" points="1919.7702,-322.4876 1928.1775,-316.0403 1917.5846,-315.8375 1919.7702,-322.4876"/>
+<path fill="none" stroke="#191970" d="M1525.0926,-439.3767C1692.5651,-435.0738 2189.9015,-414.8011 2590,-322 2593.6461,-321.1543 2597.4123,-320.1476 2601.1521,-319.0592"/>
+<polygon fill="#191970" stroke="#191970" points="2602.3735,-322.3449 2610.8893,-316.0416 2600.3014,-315.6586 2602.3735,-322.3449"/>
</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. ">
-<polygon fill="#ffffff" stroke="#000000" points="408,-358.5 408,-388.5 534,-388.5 534,-358.5 408,-358.5"/>
-<text text-anchor="start" x="416" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="471" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/map.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1404,-358.5 1404,-388.5 1530,-388.5 1530,-358.5 1404,-358.5"/>
+<text text-anchor="start" x="1412" y="-376.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
+<text text-anchor="middle" x="1467" y="-365.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/map.h</text>
</a>
</g>
</g>
<!-- Node41->Node42 -->
-<g id="edge99" class="edge">
+<g id="edge101" class="edge">
<title>Node41->Node42</title>
-<path fill="none" stroke="#191970" d="M1300.7358,-436.1039C1144.4915,-424.3152 716.3898,-392.0148 544.178,-379.0213"/>
-<polygon fill="#191970" stroke="#191970" points="544.4303,-375.5305 534.1953,-378.2681 543.9036,-382.5107 544.4303,-375.5305"/>
+<path fill="none" stroke="#191970" d="M1467,-425.2967C1467,-417.5013 1467,-407.7991 1467,-398.9064"/>
+<polygon fill="#191970" stroke="#191970" points="1470.5001,-398.6431 1467,-388.6432 1463.5001,-398.6432 1470.5001,-398.6431"/>
</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...">
-<polygon fill="#ffffff" stroke="#000000" points="1180,-364 1180,-383 1306,-383 1306,-364 1180,-364"/>
-<text text-anchor="middle" x="1243" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/module.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1260,-364 1260,-383 1386,-383 1386,-364 1260,-364"/>
+<text text-anchor="middle" x="1323" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/module.h</text>
</a>
</g>
</g>
<!-- Node41->Node43 -->
-<g id="edge107" class="edge">
+<g id="edge109" class="edge">
<title>Node41->Node43</title>
-<path fill="none" stroke="#191970" d="M1327.2268,-425.4639C1306.7656,-414.4247 1280.9205,-399.4954 1263.1881,-388.4715"/>
-<polygon fill="#191970" stroke="#191970" points="1265.0455,-385.5049 1254.7262,-383.1039 1261.2959,-391.416 1265.0455,-385.5049"/>
+<path fill="none" stroke="#191970" d="M1428.9431,-425.4639C1403.2933,-414.1715 1370.3089,-398.8086 1347.9233,-387.7192"/>
+<polygon fill="#191970" stroke="#191970" points="1349.2466,-384.4671 1338.7398,-383.1039 1346.1033,-390.7217 1349.2466,-384.4671"/>
</g>
<!-- Node41->Node45 -->
-<g id="edge122" class="edge">
+<g id="edge124" class="edge">
<title>Node41->Node45</title>
-<path fill="none" stroke="#191970" d="M1300.9656,-434.3647C1201.4445,-423.6958 1002.2523,-401.6572 933,-389 922.8028,-387.1363 911.7954,-384.6094 901.8781,-382.137"/>
-<polygon fill="#191970" stroke="#191970" points="902.5733,-378.7021 892.0175,-379.6103 900.8357,-385.483 902.5733,-378.7021"/>
+<path fill="none" stroke="#191970" d="M1408.8893,-426.515C1405.5543,-425.9431 1402.2408,-425.4308 1399,-425 1004.372,-372.5414 900.5287,-424.3288 504,-389 476.9934,-386.5938 446.5247,-382.2719 424.192,-378.8019"/>
+<polygon fill="#191970" stroke="#191970" points="424.6636,-375.3331 414.2397,-377.2284 423.5704,-382.2472 424.6636,-375.3331"/>
</g>
<!-- Node46 -->
<g id="node47" class="node">
<title>Node46</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1362.5,-364 1362.5,-383 1403.5,-383 1403.5,-364 1362.5,-364"/>
-<text text-anchor="middle" x="1383" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tuple</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1000.5,-364 1000.5,-383 1041.5,-383 1041.5,-364 1000.5,-364"/>
+<text text-anchor="middle" x="1021" y="-371" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tuple</text>
</g>
<!-- Node41->Node46 -->
-<g id="edge125" class="edge">
+<g id="edge127" class="edge">
<title>Node41->Node46</title>
-<path fill="none" stroke="#191970" d="M1364.446,-425.2967C1367.9312,-415.5672 1372.4803,-402.8675 1376.2018,-392.4784"/>
-<polygon fill="#191970" stroke="#191970" points="1379.5,-393.6495 1379.5773,-383.055 1372.91,-391.2889 1379.5,-393.6495"/>
+<path fill="none" stroke="#191970" d="M1408.852,-426.7649C1405.5265,-426.1293 1402.2252,-425.5349 1399,-425 1247.7835,-399.9225 1204.101,-428.5147 1056,-389 1053.8088,-388.4154 1051.5799,-387.7112 1049.3623,-386.9297"/>
+<polygon fill="#191970" stroke="#191970" points="1050.4488,-383.595 1039.8637,-383.1398 1047.8546,-390.0966 1050.4488,-383.595"/>
</g>
<!-- Node42->Node16 -->
-<g id="edge102" class="edge">
+<g id="edge104" class="edge">
<title>Node42->Node16</title>
-<path fill="none" stroke="#191970" d="M407.8023,-363.1689C355.2051,-353.6308 285.301,-338.4825 262,-322 205.59,-282.0972 171.7382,-241.5945 201,-179 217.4355,-143.8425 232.6934,-139.1126 268,-123 318.7282,-99.8495 483.0341,-81.5983 553.9318,-74.5975"/>
-<polygon fill="#191970" stroke="#191970" points="554.4701,-78.0617 564.0838,-73.6091 553.7917,-71.0947 554.4701,-78.0617"/>
+<path fill="none" stroke="#191970" d="M1530.1214,-359.355C1533.1161,-358.8592 1536.0869,-358.4033 1539,-358 1664.7201,-340.5963 1992.53,-372.43 2109,-322 2127.4664,-314.0043 2125.0937,-301.914 2142,-291 2227.6081,-235.7346 2280.1945,-271.2979 2352,-199 2382.4851,-168.3058 2398.1641,-118.4651 2404.8674,-91.0751"/>
+<polygon fill="#191970" stroke="#191970" points="2408.3217,-91.6702 2407.1423,-81.1413 2401.4983,-90.1077 2408.3217,-91.6702"/>
</g>
<!-- Node42->Node20 -->
-<g id="edge104" class="edge">
+<g id="edge106" class="edge">
<title>Node42->Node20</title>
-<path fill="none" stroke="#191970" d="M503.3164,-358.4639C523.9225,-348.8763 550.8351,-336.3545 573.0783,-326.0052"/>
-<polygon fill="#191970" stroke="#191970" points="574.7031,-329.1096 582.2932,-321.7177 571.7501,-322.763 574.7031,-329.1096"/>
+<path fill="none" stroke="#191970" d="M1518.1677,-358.4639C1552.3705,-348.4131 1597.5469,-335.1375 1633.6756,-324.5208"/>
+<polygon fill="#191970" stroke="#191970" points="1635.0017,-327.7792 1643.6092,-321.6017 1633.028,-321.0631 1635.0017,-327.7792"/>
</g>
<!-- Node42->Node21 -->
-<g id="edge103" class="edge">
+<g id="edge105" class="edge">
<title>Node42->Node21</title>
-<path fill="none" stroke="#191970" d="M479.8453,-358.343C491.7107,-339.443 514.6113,-307.6043 543,-291 573.6683,-273.0624 671.755,-257.4587 726.3764,-249.881"/>
-<polygon fill="#191970" stroke="#191970" points="726.9313,-253.3378 736.366,-248.5174 725.9845,-246.4021 726.9313,-253.3378"/>
+<path fill="none" stroke="#191970" d="M1489.1857,-358.2548C1525.0394,-333.6173 1595.3379,-285.3106 1631.4338,-260.5068"/>
+<polygon fill="#191970" stroke="#191970" points="1633.5724,-263.2839 1639.832,-254.7358 1629.608,-257.5146 1633.5724,-263.2839"/>
</g>
<!-- Node42->Node24 -->
-<g id="edge100" class="edge">
+<g id="edge102" class="edge">
<title>Node42->Node24</title>
-<path fill="none" stroke="#191970" d="M407.8293,-363.105C360.8505,-354.1119 302.0525,-339.7685 285,-322 255.2896,-291.042 252.0934,-237.3863 252.7718,-208.5964"/>
-<polygon fill="#191970" stroke="#191970" points="256.2706,-208.6953 253.1938,-198.5571 249.2768,-208.4013 256.2706,-208.6953"/>
+<path fill="none" stroke="#191970" d="M1461.6808,-358.4016C1449.9249,-325.0329 1421.7488,-245.0561 1408.7635,-208.1979"/>
+<polygon fill="#191970" stroke="#191970" points="1412.0166,-206.8983 1405.3925,-198.6295 1405.4143,-209.2243 1412.0166,-206.8983"/>
</g>
<!-- Node42->Node31 -->
-<g id="edge101" class="edge">
+<g id="edge103" class="edge">
<title>Node42->Node31</title>
-<path fill="none" stroke="#191970" d="M470.3618,-358.2473C469.8592,-341.6535 469.5925,-314.3676 472,-291 472.8944,-282.3185 474.6191,-272.8374 476.3213,-264.7505"/>
-<polygon fill="#191970" stroke="#191970" points="479.7813,-265.3159 478.5486,-254.793 472.9501,-263.7878 479.7813,-265.3159"/>
+<path fill="none" stroke="#191970" d="M1403.8293,-359.6625C1400.8478,-359.0844 1397.8933,-358.527 1395,-358 1288.4533,-338.594 1257.1626,-354.9585 1154,-322 1125.0313,-312.745 1121.1311,-302.5533 1093,-291 1060.5339,-277.6664 1022.6912,-265.7104 993.949,-257.3314"/>
+<polygon fill="#191970" stroke="#191970" points="994.7639,-253.9239 984.1859,-254.5204 992.8271,-260.6506 994.7639,-253.9239"/>
</g>
<!-- Node43->Node8 -->
-<g id="edge112" class="edge">
+<g id="edge114" class="edge">
<title>Node43->Node8</title>
-<path fill="none" stroke="#191970" d="M1231.9299,-363.8603C1215.013,-347.9414 1186.3551,-315.4849 1203,-291 1227.4524,-255.0301 1260.5133,-282.756 1294,-255 1324.7584,-229.5054 1314.1983,-207.6899 1342,-179 1353.9273,-166.6917 1369.6532,-155.7969 1382.9774,-147.7034"/>
-<polygon fill="#191970" stroke="#191970" points="1384.8855,-150.6419 1391.7419,-142.5648 1381.345,-144.6032 1384.8855,-150.6419"/>
+<path fill="none" stroke="#191970" d="M1336.6211,-363.9362C1351.1638,-353.6837 1374.8412,-336.8714 1395,-322 1479.4798,-259.6782 1487.5996,-222.8133 1583,-179 1628.9222,-157.9099 1684.8457,-146.331 1728.1455,-140.0661"/>
+<polygon fill="#191970" stroke="#191970" points="1728.8788,-143.4981 1738.3043,-138.6597 1727.9187,-136.5642 1728.8788,-143.4981"/>
</g>
<!-- Node43->Node9 -->
-<g id="edge109" class="edge">
+<g id="edge111" class="edge">
<title>Node43->Node9</title>
-<path fill="none" stroke="#191970" d="M1273.638,-363.9455C1306.0828,-353.7702 1358.7607,-337.0854 1404,-322 1490.1367,-293.2771 1513.51,-290.6926 1597,-255 1614.9115,-247.3427 1618.5059,-243.5682 1636,-235 1745.9354,-181.1563 1877.0271,-120.9776 1943.222,-90.8685"/>
-<polygon fill="#191970" stroke="#191970" points="1944.8515,-93.9725 1952.5073,-86.6487 1941.9553,-87.5998 1944.8515,-93.9725"/>
+<path fill="none" stroke="#191970" d="M1259.6537,-368.2699C1164.7101,-359.9089 990.3159,-342.3825 931,-322 875.4206,-302.9015 861.7526,-293.0924 817,-255 759.0646,-205.6867 705.9223,-132.5203 680.5879,-95.2252"/>
+<polygon fill="#191970" stroke="#191970" points="683.2426,-92.9005 674.76,-86.5526 677.4325,-96.8048 683.2426,-92.9005"/>
</g>
<!-- Node43->Node14 -->
-<g id="edge115" class="edge">
+<g id="edge117" class="edge">
<title>Node43->Node14</title>
-<path fill="none" stroke="#191970" d="M1305.719,-363.9685C1321.456,-361.7901 1338.3244,-359.6372 1354,-358 1468.7645,-346.0138 1763.2776,-360.6523 1872,-322 1894.5397,-313.9868 1895.022,-302.4966 1916,-291 2088.6936,-196.3588 2140.6678,-189.5734 2326,-123 2372.0668,-106.4523 2426.4062,-90.1971 2460.0367,-80.5081"/>
-<polygon fill="#191970" stroke="#191970" points="2461.1267,-83.8367 2469.7768,-77.7189 2459.1996,-77.1072 2461.1267,-83.8367"/>
+<path fill="none" stroke="#191970" d="M1259.6668,-368.6838C1151.1022,-360.0559 934.0104,-341.0766 860,-322 748.3832,-293.2302 488.9078,-163.5208 381,-123 339.6291,-107.4646 291.1787,-91.5294 260.0641,-81.5804"/>
+<polygon fill="#191970" stroke="#191970" points="260.8025,-78.1425 250.212,-78.4451 258.6797,-84.8129 260.8025,-78.1425"/>
</g>
<!-- Node43->Node18 -->
-<g id="edge117" class="edge">
+<g id="edge119" class="edge">
<title>Node43->Node18</title>
-<path fill="none" stroke="#191970" d="M1248.9785,-363.9544C1259.8043,-347.3011 1284.1331,-312.6526 1312,-291 1330.9764,-276.2554 1355.443,-264.6215 1374.9266,-256.7423"/>
-<polygon fill="#191970" stroke="#191970" points="1376.2822,-259.9704 1384.3247,-253.0734 1373.7365,-253.4497 1376.2822,-259.9704"/>
+<path fill="none" stroke="#191970" d="M1361.2728,-363.9502C1372.1096,-361.6043 1383.961,-359.374 1395,-358 1506.4336,-344.1296 2301.3489,-365.2004 2405,-322 2423.5744,-314.2585 2421.5446,-302.5827 2438,-291 2456.0398,-278.3021 2478.0665,-266.982 2495.8135,-258.7489"/>
+<polygon fill="#191970" stroke="#191970" points="2497.5171,-261.8195 2505.1802,-254.5033 2494.6272,-255.4439 2497.5171,-261.8195"/>
</g>
<!-- Node43->Node22 -->
-<g id="edge111" class="edge">
+<g id="edge113" class="edge">
<title>Node43->Node22</title>
-<path fill="none" stroke="#191970" d="M1202.7873,-363.9885C1192.4702,-361.7838 1181.3599,-359.6136 1171,-358 1007.7475,-332.572 952.1035,-388.826 801,-322 760.4244,-304.0553 744.9897,-295.5557 727,-255 717.783,-234.2212 738.8124,-215.7174 758.6473,-203.6287"/>
-<polygon fill="#191970" stroke="#191970" points="760.613,-206.5369 767.5637,-198.5407 757.1436,-200.4571 760.613,-206.5369"/>
+<path fill="none" stroke="#191970" d="M1355.2456,-363.9639C1410.0219,-347.9154 1524.8617,-314.9205 1623,-291 1739.11,-262.6991 1768.4587,-256.5988 1886,-235 1961.3143,-221.1606 2047.8853,-208.071 2108.0545,-199.4198"/>
+<polygon fill="#191970" stroke="#191970" points="2108.8218,-202.8458 2118.2247,-197.9637 2107.8296,-195.9164 2108.8218,-202.8458"/>
</g>
<!-- Node43->Node26 -->
-<g id="edge110" class="edge">
+<g id="edge112" class="edge">
<title>Node43->Node26</title>
-<path fill="none" stroke="#191970" d="M1210.6197,-363.9005C1175.9706,-353.6284 1120.2199,-337.1006 1077.4202,-324.4122"/>
-<polygon fill="#191970" stroke="#191970" points="1078.2528,-321.0085 1067.6704,-321.5218 1076.2631,-327.7198 1078.2528,-321.0085"/>
+<path fill="none" stroke="#191970" d="M1323,-363.9005C1323,-355.5099 1323,-342.9451 1323,-331.7085"/>
+<polygon fill="#191970" stroke="#191970" points="1326.5001,-331.5217 1323,-321.5218 1319.5001,-331.5218 1326.5001,-331.5217"/>
</g>
<!-- Node43->Node29 -->
-<g id="edge113" class="edge">
+<g id="edge115" class="edge">
<title>Node43->Node29</title>
-<path fill="none" stroke="#191970" d="M1203.958,-363.9446C1193.3257,-361.658 1181.77,-359.459 1171,-358 893.9613,-320.4691 815.9797,-382.3416 543,-322 481.1257,-308.3228 413.259,-277.4014 376.4587,-259.1768"/>
-<polygon fill="#191970" stroke="#191970" points="377.8718,-255.9701 367.3637,-254.6166 374.7342,-262.2276 377.8718,-255.9701"/>
+<path fill="none" stroke="#191970" d="M1259.7906,-365.5659C1188.3233,-355.9864 1076.8558,-339.0305 1038,-322 1017.0769,-312.8294 1017.2595,-301.5559 997,-291 978.9592,-281.6001 929.6373,-266.4282 894.1886,-256.0997"/>
+<polygon fill="#191970" stroke="#191970" points="894.7187,-252.6096 884.1399,-253.1928 892.7735,-259.3339 894.7187,-252.6096"/>
</g>
<!-- Node43->Node31 -->
-<g id="edge116" class="edge">
+<g id="edge118" class="edge">
<title>Node43->Node31</title>
-<path fill="none" stroke="#191970" d="M1203.1813,-363.9575C1192.758,-361.7267 1181.4986,-359.5518 1171,-358 976.4616,-329.2454 914.5234,-389.9935 730,-322 707.8933,-313.8541 708.4913,-300.6533 687,-291 660.7543,-279.2111 589.1967,-264.5768 537.7914,-255.0236"/>
-<polygon fill="#191970" stroke="#191970" points="538.267,-251.5524 527.7984,-253.183 536.999,-258.4366 538.267,-251.5524"/>
+<path fill="none" stroke="#191970" d="M1273.7219,-363.9724C1231.2959,-355.126 1168.7571,-340.5396 1116,-322 1065.0088,-304.0809 1008.4282,-276.1891 975.949,-259.3502"/>
+<polygon fill="#191970" stroke="#191970" points="977.5476,-256.2366 967.0638,-254.7069 974.3055,-262.4405 977.5476,-256.2366"/>
</g>
<!-- Node43->Node38 -->
-<g id="edge108" class="edge">
+<g id="edge110" class="edge">
<title>Node43->Node38</title>
-<path fill="none" stroke="#191970" d="M1262.7971,-363.8987C1296.1105,-347.8958 1365.6558,-315.1453 1426,-291 1456.5602,-278.7721 1491.7916,-266.5008 1518.0155,-257.7116"/>
-<polygon fill="#191970" stroke="#191970" points="1519.2146,-261.0013 1527.5965,-254.5209 1517.0028,-254.3599 1519.2146,-261.0013"/>
+<path fill="none" stroke="#191970" d="M1363.2122,-363.9848C1373.5292,-361.7802 1384.6397,-359.611 1395,-358 1559.1334,-332.4784 1603.825,-353.0647 1767,-322 1851.2633,-305.9582 1947.2698,-275.2354 1998.3968,-257.8391"/>
+<polygon fill="#191970" stroke="#191970" points="1999.6818,-261.0986 2008.0071,-254.5458 1997.4125,-254.4767 1999.6818,-261.0986"/>
</g>
<!-- Node43->Node41 -->
-<g id="edge118" class="edge">
+<g id="edge120" class="edge">
<title>Node43->Node41</title>
-<path fill="none" stroke="#191970" d="M1264.5291,-383.1039C1283.0892,-392.7457 1309.7723,-407.8959 1330.1316,-420.2103"/>
-<polygon fill="#191970" stroke="#191970" points="1328.3523,-423.2248 1338.7079,-425.4639 1332.0089,-417.2557 1328.3523,-423.2248"/>
+<path fill="none" stroke="#191970" d="M1348.5427,-383.1039C1371.6387,-392.9619 1405.6324,-408.5781 1431.3565,-421.0343"/>
+<polygon fill="#191970" stroke="#191970" points="1429.9026,-424.2193 1440.4241,-425.4639 1432.9752,-417.9297 1429.9026,-424.2193"/>
</g>
<!-- Node44 -->
<g id="node45" class="node">
<title>Node44</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1212,-297 1212,-316 1260,-316 1260,-297 1212,-297"/>
-<text text-anchor="middle" x="1236" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">mutex</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="940,-297 940,-316 988,-316 988,-297 940,-297"/>
+<text text-anchor="middle" x="964" y="-304" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">mutex</text>
</g>
<!-- Node43->Node44 -->
-<g id="edge114" class="edge">
+<g id="edge116" class="edge">
<title>Node43->Node44</title>
-<path fill="none" stroke="#191970" d="M1241.9971,-363.9005C1240.9783,-354.149 1239.3704,-338.7597 1238.0759,-326.3695"/>
-<polygon fill="#191970" stroke="#191970" points="1241.5213,-325.6638 1237.0011,-316.0817 1234.5592,-326.3912 1241.5213,-325.6638"/>
+<path fill="none" stroke="#191970" d="M1259.9514,-367.6817C1194.554,-360.653 1089.7732,-346.5934 1002,-322 999.5387,-321.3104 997.021,-320.5096 994.511,-319.6425"/>
+<polygon fill="#191970" stroke="#191970" points="995.5963,-316.3113 985.0043,-316.0632 993.1298,-322.8623 995.5963,-316.3113"/>
</g>
<!-- Node47->Node6 -->
-<g id="edge133" class="edge">
+<g id="edge135" class="edge">
<title>Node47->Node6</title>
-<path fill="none" stroke="#191970" d="M2311.9582,-492.3051C2289.4824,-461.3613 2219.2251,-364.6339 2189.9781,-324.3677"/>
-<polygon fill="#191970" stroke="#191970" points="2192.793,-322.2875 2184.0843,-316.2534 2187.1293,-326.4012 2192.793,-322.2875"/>
+<path fill="none" stroke="#191970" d="M3007.74,-492.4549C3017.131,-484.0029 3029.5909,-470.7442 3035,-456 3039.7453,-443.0652 3041.5051,-437.1454 3035,-425 3005.7295,-370.3501 2938.9048,-336.5747 2895.9751,-319.6945"/>
+<polygon fill="#191970" stroke="#191970" points="2896.9982,-316.3391 2886.4071,-316.0556 2894.5097,-322.8819 2896.9982,-316.3391"/>
</g>
<!-- Node48 -->
<g id="node49" class="node">
<title>Node48</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2499,-431 2499,-450 2559,-450 2559,-431 2499,-431"/>
-<text text-anchor="middle" x="2529" y="-438" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">iostream</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2966,-431 2966,-450 3026,-450 3026,-431 2966,-431"/>
+<text text-anchor="middle" x="2996" y="-438" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">iostream</text>
</g>
<!-- Node47->Node48 -->
-<g id="edge134" class="edge">
+<g id="edge136" class="edge">
<title>Node47->Node48</title>
-<path fill="none" stroke="#191970" d="M2356.1341,-492.4878C2389.9171,-483.6499 2441.0644,-469.8084 2485,-456 2487.5908,-455.1858 2490.2577,-454.3189 2492.9364,-453.4272"/>
-<polygon fill="#191970" stroke="#191970" points="2494.2397,-456.6807 2502.5735,-450.1386 2491.979,-450.0558 2494.2397,-456.6807"/>
+<path fill="none" stroke="#191970" d="M2996,-492.3906C2996,-483.8657 2996,-471.1392 2996,-460.4235"/>
+<polygon fill="#191970" stroke="#191970" points="2999.5001,-460.2448 2996,-450.2449 2992.5001,-460.2449 2999.5001,-460.2448"/>
</g>
<!-- Node49->Node2 -->
-<g id="edge147" class="edge">
+<g id="edge149" class="edge">
<title>Node49->Node2</title>
-<path fill="none" stroke="#191970" d="M2477.0972,-662.9667C2528.1754,-653.9932 2619.1683,-637.9606 2697,-624 2698.4628,-623.7376 2699.9473,-623.4709 2701.4458,-623.2012"/>
-<polygon fill="#191970" stroke="#191970" points="2702.2358,-626.6152 2711.4549,-621.3943 2700.9922,-619.7266 2702.2358,-626.6152"/>
+<path fill="none" stroke="#191970" d="M383.2288,-660.3733C353.335,-651.021 307.7907,-636.7725 275.0783,-626.5385"/>
+<polygon fill="#191970" stroke="#191970" points="275.9586,-623.1466 265.3697,-623.5011 273.8685,-629.8273 275.9586,-623.1466"/>
</g>
<!-- Node49->Node3 -->
-<g id="edge148" class="edge">
+<g id="edge150" class="edge">
<title>Node49->Node3</title>
-<path fill="none" stroke="#191970" d="M2422.9322,-660.4509C2394.3162,-641.0268 2329.117,-596.7703 2294.5247,-573.2895"/>
-<polygon fill="#191970" stroke="#191970" points="2296.2718,-570.2453 2286.0322,-567.5249 2292.3404,-576.037 2296.2718,-570.2453"/>
+<path fill="none" stroke="#191970" d="M454.0773,-669.4687C661.485,-666.6033 1608.2902,-652.0716 1736,-624 1784.2072,-613.4037 1836.3027,-588.5626 1866.7171,-572.5601"/>
+<polygon fill="#191970" stroke="#191970" points="1868.6298,-575.5063 1875.7976,-567.7043 1865.3288,-569.3335 1868.6298,-575.5063"/>
</g>
<!-- Node49->Node8 -->
-<g id="edge151" class="edge">
+<g id="edge153" class="edge">
<title>Node49->Node8</title>
-<path fill="none" stroke="#191970" d="M2475.984,-660.4794C2576.0765,-633.9856 2829.5275,-552.8322 2760,-425 2681.8822,-281.3739 2598.5713,-291.0957 2445,-235 2338.8618,-196.2304 2308.8881,-194.794 2197,-179 2059.5287,-159.5947 1645.6793,-142.0566 1479.8428,-135.6135"/>
-<polygon fill="#191970" stroke="#191970" points="1479.8851,-132.1126 1469.7573,-135.2235 1479.6145,-139.1074 1479.8851,-132.1126"/>
+<path fill="none" stroke="#191970" d="M454.1323,-668.7063C605.3538,-663.7124 1140.7307,-644.924 1310,-624 1437.6808,-608.2169 1484.2439,-635.12 1594,-568 1712.7892,-495.3559 1723.2447,-445.2672 1788,-322 1818.5829,-263.7829 1843.7556,-242.3189 1826,-179 1823.2122,-169.0584 1817.6188,-159.1336 1812.1593,-151.0613"/>
+<polygon fill="#191970" stroke="#191970" points="1814.9048,-148.8849 1806.2089,-142.8323 1809.2324,-152.9866 1814.9048,-148.8849"/>
</g>
<!-- Node49->Node14 -->
-<g id="edge152" class="edge">
+<g id="edge154" class="edge">
<title>Node49->Node14</title>
-<path fill="none" stroke="#191970" d="M2477.3277,-669.0338C2591.7123,-665.2098 2910,-646.6082 2910,-558 2910,-558 2910,-558 2910,-245 2910,-202.8832 2876.751,-202.7949 2842,-179 2739.4054,-108.751 2589.2478,-83.0927 2524.1135,-74.8964"/>
-<polygon fill="#191970" stroke="#191970" points="2524.3953,-71.4052 2514.0487,-73.6853 2523.5589,-78.355 2524.3953,-71.4052"/>
+<path fill="none" stroke="#191970" d="M393.1117,-660.4041C319.42,-624.9605 76,-493.7133 76,-306.5 76,-306.5 76,-306.5 76,-189 76,-128.8142 151.5179,-95.0962 196.0946,-80.4359"/>
+<polygon fill="#191970" stroke="#191970" points="197.4082,-83.6918 205.8979,-77.3534 195.3085,-77.0142 197.4082,-83.6918"/>
</g>
<!-- Node49->Node32 -->
-<g id="edge149" class="edge">
+<g id="edge151" class="edge">
<title>Node49->Node32</title>
-<path fill="none" stroke="#191970" d="M2396.708,-665.123C2212.7969,-642.7816 1434.6334,-547.3766 798,-456 607.4143,-428.645 506.2142,-525.0768 370,-389 355.251,-374.2658 352.9686,-350.0582 353.8017,-331.6663"/>
-<polygon fill="#191970" stroke="#191970" points="357.2966,-331.8661 354.5816,-321.6251 350.3177,-331.324 357.2966,-331.8661"/>
+<path fill="none" stroke="#191970" d="M454.2052,-668.2986C659.5864,-659.349 1582,-615.9294 1582,-558 1582,-558 1582,-558 1582,-502 1582,-465.2158 1593.4474,-423.9295 1601.9578,-398.3184"/>
+<polygon fill="#191970" stroke="#191970" points="1605.3307,-399.2726 1605.2705,-388.6779 1598.7107,-396.9977 1605.3307,-399.2726"/>
</g>
<!-- Node49->Node33 -->
-<g id="edge150" class="edge">
+<g id="edge152" class="edge">
<title>Node49->Node33</title>
-<path fill="none" stroke="#191970" d="M2472.9287,-660.4171C2517.3049,-646.149 2587,-614.7126 2587,-558 2587,-558 2587,-558 2587,-502 2587,-466.7513 2588.0792,-453.9706 2568,-425 2532.632,-373.9707 2467.6073,-338.3399 2428.1374,-320.2386"/>
-<polygon fill="#191970" stroke="#191970" points="2429.2447,-316.8999 2418.6869,-316.0156 2426.3888,-323.2909 2429.2447,-316.8999"/>
+<path fill="none" stroke="#191970" d="M411.522,-660.4867C398.6845,-610.77 341.0878,-382.1719 361,-358 372.8392,-343.6281 472.4048,-327.1736 545.7871,-316.8104"/>
+<polygon fill="#191970" stroke="#191970" points="546.5367,-320.2397 555.956,-315.3894 545.5678,-313.3071 546.5367,-320.2397"/>
</g>
</g>
</svg>
diff --git a/docs/reference/api/doxygen/affine__type_8h_source.html b/docs/reference/api/doxygen/affine__type_8h_source.html
index cfcdaca9c8..aee56b0c01 100644
--- a/docs/reference/api/doxygen/affine__type_8h_source.html
+++ b/docs/reference/api/doxygen/affine__type_8h_source.html
@@ -90,7 +90,7 @@ $(function() {
<div class="ttc" id="classtvm_1_1runtime_1_1DataType_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1DataType.html">tvm::runtime::DataType</a></div><div class="ttdoc">Runtime primitive data type. </div><div class="ttdef"><b>Definition:</b> data_type.h:41</div></div>
<div class="ttc" id="classtvm_1_1SHashReducer_1_1Handler_html_a8f9a489881fc55552f13a58313a863cf"><div class="ttname"><a href="classtvm_1_1SHashReducer_1_1Handler.html#a8f9a489881fc55552f13a58313a863cf">tvm::SHashReducer::Handler::MarkGraphNode</a></div><div class="ttdeci">virtual void MarkGraphNode()=0</div><div class="ttdoc">Mark current comparison as graph node in hashing. Graph node hash will depends on the graph structure...</div></div>
<div class="ttc" id="classtvm_1_1TensorAffineTypeNode_html_a367710244d5e50bf400d64fd5fc1897e"><div class="ttname"><a href="classtvm_1_1TensorAffineTypeNode.html#a367710244d5e50bf400d64fd5fc1897e">tvm::TensorAffineTypeNode::VisitAttrs</a></div><div class="ttdeci">void VisitAttrs(tvm::AttrVisitor *v)</div><div class="ttdef"><b>Definition:</b> affine_type.h:77</div></div>
-<div class="ttc" id="classtvm_1_1runtime_1_1Array_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html">tvm::runtime::Array</a></div><div class="ttdoc">Array, container representing a contiguous sequence of ObjectRefs. </div><div class="ttdef"><b>Definition:</b> array.h:270</div></div>
+<div class="ttc" id="classtvm_1_1runtime_1_1Array_html"><div class="ttname"><a href="classtvm_1_1runtime_1_1Array.html">tvm::runtime::Array</a></div><div class="ttdoc">Array, container representing a contiguous sequence of ObjectRefs. </div><div class="ttdef"><b>Definition:</b> array.h:289</div></div>
<div class="ttc" id="classtvm_1_1TensorAffineTypeNode_html_a45fefe29872d61434bd7c7f01cd98536"><div class="ttname"><a href="classtvm_1_1TensorAffineTypeNode.html#a45fefe29872d61434bd7c7f01cd98536">tvm::TensorAffineTypeNode::zero_point</a></div><div class="ttdeci">RelayExpr zero_point</div><div class="ttdoc">The zero point of this type. </div><div class="ttdef"><b>Definition:</b> affine_type.h:71</div></div>
<div class="ttc" id="classtvm_1_1AffineTypeNode_html_a8064dec5da4e223b235c14e4ca72d06b"><div class="ttname"><a href="classtvm_1_1AffineTypeNode.html#a8064dec5da4e223b235c14e4ca72d06b">tvm::AffineTypeNode::_type_has_method_shash_reduce</a></div><div class="ttdeci">static constexpr const bool _type_has_method_shash_reduce</div><div class="ttdef"><b>Definition:</b> affine_type.h:46</div></div>
<div class="ttc" id="classtvm_1_1RelayExpr_html"><div class="ttname"><a href="classtvm_1_1RelayExpr.html">tvm::RelayExpr</a></div><div class="ttdoc">Managed reference to RelayExprNode. </div><div class="ttdef"><b>Definition:</b> expr.h:431</div></div>
diff --git a/docs/reference/api/doxygen/algorithm_8h__incl.svg b/docs/reference/api/doxygen/algorithm_8h__incl.svg
index ca1de438ca..e547012e56 100644
--- a/docs/reference/api/doxygen/algorithm_8h__incl.svg
+++ b/docs/reference/api/doxygen/algorithm_8h__incl.svg
@@ -12,936 +12,942 @@
<!-- Node0 -->
<g id="node1" class="node">
<title>Node0</title>
-<polygon fill="#bfbfbf" stroke="#000000" points="2775,-1007.5 2775,-1037.5 2903,-1037.5 2903,-1007.5 2775,-1007.5"/>
-<text text-anchor="start" x="2783" y="-1025.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">include/tvm/relay/attrs</text>
-<text text-anchor="middle" x="2839" y="-1014.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/algorithm.h</text>
+<polygon fill="#bfbfbf" stroke="#000000" points="2919,-1007.5 2919,-1037.5 3047,-1037.5 3047,-1007.5 2919,-1007.5"/>
+<text text-anchor="start" x="2927" y="-1025.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">include/tvm/relay/attrs</text>
+<text text-anchor="middle" x="2983" y="-1014.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/algorithm.h</text>
</g>
<!-- Node1 -->
<g id="node2" class="node">
<title>Node1</title>
<g id="a_node2"><a xlink:href="ir_2attrs_8h.html" target="_top" xlink:title="Helpers for attribute objects. ">
-<polygon fill="#ffffff" stroke="#000000" points="1500,-660.5 1500,-679.5 1580,-679.5 1580,-660.5 1500,-660.5"/>
-<text text-anchor="middle" x="1540" y="-667.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/attrs.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1377,-660.5 1377,-679.5 1457,-679.5 1457,-660.5 1377,-660.5"/>
+<text text-anchor="middle" x="1417" y="-667.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/attrs.h</text>
</a>
</g>
</g>
<!-- Node0->Node1 -->
<g id="edge1" class="edge">
<title>Node0->Node1</title>
-<path fill="none" stroke="#191970" d="M2774.68,-1020.3684C2493.1413,-1010.7288 1381.1898,-969.3157 1332,-915 1254.5754,-829.5072 1437.8168,-722.3326 1511.2358,-684.2375"/>
-<polygon fill="#191970" stroke="#191970" points="1513.1206,-687.2047 1520.4272,-679.5323 1509.9308,-680.9736 1513.1206,-687.2047"/>
+<path fill="none" stroke="#191970" d="M2918.9892,-1022.1318C2578.9835,-1019.766 1000,-1003.4779 1000,-905 1000,-905 1000,-905 1000,-787.5 1000,-746.8601 1021.58,-735.9255 1057,-716 1108.9126,-686.7967 1280.502,-675.6388 1366.683,-671.8004"/>
+<polygon fill="#191970" stroke="#191970" points="1366.9151,-675.2937 1376.7557,-671.3679 1366.6147,-668.3002 1366.9151,-675.2937"/>
</g>
<!-- Node16 -->
<g id="node13" class="node">
<title>Node16</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="3524,-6 3524,-25 3568,-25 3568,-6 3524,-6"/>
-<text text-anchor="middle" x="3546" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="3562,-6 3562,-25 3606,-25 3606,-6 3562,-6"/>
+<text text-anchor="middle" x="3584" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">string</text>
</g>
<!-- Node0->Node16 -->
-<g id="edge223" class="edge">
+<g id="edge224" class="edge">
<title>Node0->Node16</title>
-<path fill="none" stroke="#191970" d="M2903.005,-1019.8214C3156.7175,-1008.7886 4078,-964.6216 4078,-905 4078,-905 4078,-905 4078,-133 4078,-91.201 4048.7446,-84.9584 4011,-67 3934.0113,-30.3697 3670.5759,-19.1837 3578.2682,-16.3442"/>
-<polygon fill="#191970" stroke="#191970" points="3578.2296,-12.8417 3568.1307,-16.0451 3578.0231,-19.8386 3578.2296,-12.8417"/>
+<path fill="none" stroke="#191970" d="M3047.0065,-1019.0138C3281.2171,-1005.8407 4078,-957.1904 4078,-905 4078,-905 4078,-905 4078,-133 4078,-91.201 4048.6568,-85.1418 4011,-67 3941.0963,-33.3228 3703.2638,-20.3837 3616.2384,-16.7068"/>
+<polygon fill="#191970" stroke="#191970" points="3616.2942,-13.2063 3606.1594,-16.2943 3616.0078,-20.2004 3616.2942,-13.2063"/>
</g>
<!-- Node52 -->
<g id="node36" class="node">
<title>Node52</title>
<g id="a_node36"><a xlink:href="relay_2base_8h.html" target="_top" xlink:title="Base classes for the Relay IR. ">
-<polygon fill="#ffffff" stroke="#000000" points="3650.5,-839.5 3650.5,-858.5 3749.5,-858.5 3749.5,-839.5 3650.5,-839.5"/>
-<text text-anchor="middle" x="3700" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/relay/base.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2144.5,-839.5 2144.5,-858.5 2243.5,-858.5 2243.5,-839.5 2144.5,-839.5"/>
+<text text-anchor="middle" x="2194" y="-846.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/relay/base.h</text>
</a>
</g>
</g>
<!-- Node0->Node52 -->
-<g id="edge133" class="edge">
+<g id="edge134" class="edge">
<title>Node0->Node52</title>
-<path fill="none" stroke="#191970" d="M2903.0115,-1009.601C3064.7336,-977.0125 3487.8339,-891.7536 3642.3978,-860.6074"/>
-<polygon fill="#191970" stroke="#191970" points="3643.5088,-863.954 3652.6203,-858.5475 3642.1259,-857.0919 3643.5088,-863.954"/>
+<path fill="none" stroke="#191970" d="M2918.7674,-1008.3753C2766.9505,-974.991 2388.2016,-891.7047 2247.6276,-860.7926"/>
+<polygon fill="#191970" stroke="#191970" points="2248.0471,-857.3013 2237.5288,-858.5719 2246.5437,-864.138 2248.0471,-857.3013"/>
</g>
<!-- Node56 -->
<g id="node40" class="node">
<title>Node56</title>
<g id="a_node40"><a xlink:href="relay_2expr_8h.html" target="_top" xlink:title="Relay expression language. ">
-<polygon fill="#ffffff" stroke="#000000" points="1905.5,-951.5 1905.5,-970.5 2002.5,-970.5 2002.5,-951.5 1905.5,-951.5"/>
-<text text-anchor="middle" x="1954" y="-958.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/relay/expr.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1644.5,-951.5 1644.5,-970.5 1741.5,-970.5 1741.5,-951.5 1644.5,-951.5"/>
+<text text-anchor="middle" x="1693" y="-958.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/relay/expr.h</text>
</a>
</g>
</g>
<!-- Node0->Node56 -->
-<g id="edge164" class="edge">
+<g id="edge165" class="edge">
<title>Node0->Node56</title>
-<path fill="none" stroke="#191970" d="M2774.997,-1018.0523C2609.974,-1006.5846 2171.863,-976.1396 2012.7758,-965.0844"/>
-<polygon fill="#191970" stroke="#191970" points="2012.925,-961.5864 2002.7064,-964.3847 2012.4397,-968.5696 2012.925,-961.5864"/>
+<path fill="none" stroke="#191970" d="M2918.5873,-1019.4292C2696.5733,-1008.8448 1965.1639,-973.9753 1751.8705,-963.8066"/>
+<polygon fill="#191970" stroke="#191970" points="1751.9679,-960.3074 1741.8125,-963.3271 1751.6345,-967.2994 1751.9679,-960.3074"/>
</g>
<!-- Node2 -->
<g id="node3" class="node">
<title>Node2</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1450.5,-604.5 1450.5,-623.5 1545.5,-623.5 1545.5,-604.5 1450.5,-604.5"/>
-<text text-anchor="middle" x="1498" y="-611.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/common.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1369.5,-604.5 1369.5,-623.5 1464.5,-623.5 1464.5,-604.5 1369.5,-604.5"/>
+<text text-anchor="middle" x="1417" y="-611.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">dmlc/common.h</text>
</g>
<!-- Node1->Node2 -->
<g id="edge2" class="edge">
<title>Node1->Node2</title>
-<path fill="none" stroke="#191970" d="M1532.6842,-660.2455C1526.7821,-652.3761 1518.3258,-641.101 1511.2406,-631.6542"/>
-<polygon fill="#191970" stroke="#191970" points="1514.0321,-629.5427 1505.232,-623.6427 1508.432,-633.7427 1514.0321,-629.5427"/>
+<path fill="none" stroke="#191970" d="M1417,-660.2455C1417,-652.9382 1417,-642.6944 1417,-633.7046"/>
+<polygon fill="#191970" stroke="#191970" points="1420.5001,-633.6426 1417,-623.6427 1413.5001,-633.6427 1420.5001,-633.6426"/>
</g>
<!-- Node3 -->
<g id="node4" class="node">
<title>Node3</title>
<g id="a_node4"><a xlink:href="ir_2expr_8h.html" target="_top" xlink:title="Base expr nodes in TVM. ">
-<polygon fill="#ffffff" stroke="#000000" points="2132.5,-604.5 2132.5,-623.5 2211.5,-623.5 2211.5,-604.5 2132.5,-604.5"/>
-<text text-anchor="middle" x="2172" y="-611.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/expr.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2268.5,-604.5 2268.5,-623.5 2347.5,-623.5 2347.5,-604.5 2268.5,-604.5"/>
+<text text-anchor="middle" x="2308" y="-611.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/expr.h</text>
</a>
</g>
</g>
<!-- Node1->Node3 -->
<g id="edge3" class="edge">
<title>Node1->Node3</title>
-<path fill="none" stroke="#191970" d="M1580.0139,-666.4545C1690.6177,-656.6541 2000.3274,-629.2115 2122.0102,-618.4295"/>
-<polygon fill="#191970" stroke="#191970" points="2122.509,-621.8991 2132.161,-617.53 2121.8911,-614.9264 2122.509,-621.8991"/>
+<path fill="none" stroke="#191970" d="M1457.0154,-667.485C1601.8958,-658.3792 2098.8964,-627.1423 2258.3413,-617.1211"/>
+<polygon fill="#191970" stroke="#191970" points="2258.5875,-620.6126 2268.3482,-616.4921 2258.1483,-613.6264 2258.5875,-620.6126"/>
</g>
<!-- Node7 -->
<g id="node8" class="node">
<title>Node7</title>
<g id="a_node8"><a xlink:href="structural__equal_8h.html" target="_top" xlink:title="Structural equality comparison. ">
-<polygon fill="#ffffff" stroke="#000000" points="2474.5,-313.5 2474.5,-343.5 2587.5,-343.5 2587.5,-313.5 2474.5,-313.5"/>
-<text text-anchor="start" x="2482.5" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
-<text text-anchor="middle" x="2531" y="-320.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_equal.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2615.5,-313.5 2615.5,-343.5 2728.5,-343.5 2728.5,-313.5 2615.5,-313.5"/>
+<text text-anchor="start" x="2623.5" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
+<text text-anchor="middle" x="2672" y="-320.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_equal.h</text>
</a>
</g>
</g>
<!-- Node1->Node7 -->
-<g id="edge124" class="edge">
+<g id="edge125" class="edge">
<title>Node1->Node7</title>
-<path fill="none" stroke="#191970" d="M1578.0173,-660.4976C1611.0868,-651.9259 1660.1373,-638.4727 1702,-624 1848.1799,-573.4628 1878.7345,-545.1084 2024,-492 2183.1865,-433.8022 2374.2842,-375.1389 2470.0244,-346.5018"/>
-<polygon fill="#191970" stroke="#191970" points="2471.2443,-349.7903 2479.825,-343.5756 2469.2416,-343.0829 2471.2443,-349.7903"/>
+<path fill="none" stroke="#191970" d="M1448.1269,-660.4576C1559.8976,-626.4313 1951.8096,-509.1812 2281,-436 2371.7282,-415.8305 2399.9801,-432.649 2487,-400 2503.0571,-393.9756 2504.4527,-387.2392 2520,-380 2548.2432,-366.8493 2580.8125,-355.3615 2608.5407,-346.5928"/>
+<polygon fill="#191970" stroke="#191970" points="2609.6243,-349.9212 2618.1315,-343.6063 2607.5431,-343.2377 2609.6243,-349.9212"/>
</g>
<!-- Node1->Node16 -->
-<g id="edge128" class="edge">
+<g id="edge129" class="edge">
<title>Node1->Node16</title>
-<path fill="none" stroke="#191970" d="M1580.2587,-667.5607C1685.466,-661.1244 1977.2785,-642.8768 2220,-624 2532.6886,-599.6817 3584,-575.1328 3584,-261.5 3584,-261.5 3584,-261.5 3584,-133 3584,-96.7454 3567.4195,-57.1181 3556.163,-34.3929"/>
-<polygon fill="#191970" stroke="#191970" points="3559.1292,-32.5073 3551.4445,-25.2137 3552.9035,-35.7076 3559.1292,-32.5073"/>
+<path fill="none" stroke="#191970" d="M1457.2686,-668.151C1625.3646,-660.4158 2266.5184,-630.7417 2357,-624 2605.9978,-605.4475 2667.9228,-596.3157 2916,-568 3110.5473,-545.7943 3164.0613,-563.4189 3353,-512 3531.4577,-463.4334 3622,-379.4483 3622,-194.5 3622,-194.5 3622,-194.5 3622,-133 3622,-96.7454 3605.4195,-57.1181 3594.163,-34.3929"/>
+<polygon fill="#191970" stroke="#191970" points="3597.1292,-32.5073 3589.4445,-25.2137 3590.9035,-35.7076 3597.1292,-32.5073"/>
</g>
<!-- Node17 -->
<g id="node14" class="node">
<title>Node17</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="2191.5,-6 2191.5,-25 2260.5,-25 2260.5,-6 2191.5,-6"/>
-<text text-anchor="middle" x="2226" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="2060.5,-6 2060.5,-25 2129.5,-25 2129.5,-6 2060.5,-6"/>
+<text text-anchor="middle" x="2095" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">type_traits</text>
</g>
<!-- Node1->Node17 -->
-<g id="edge129" class="edge">
+<g id="edge130" class="edge">
<title>Node1->Node17</title>
-<path fill="none" stroke="#191970" d="M1499.9824,-667.783C1319.0165,-655.8487 591.2171,-586.4102 714,-246 761.3526,-114.7169 835.0252,-93.7475 972,-67 1092.4639,-43.4766 1972.3251,-21.4658 2181.4788,-16.5281"/>
-<polygon fill="#191970" stroke="#191970" points="2181.5762,-20.0269 2191.4912,-16.2927 2181.4116,-13.0288 2181.5762,-20.0269"/>
+<path fill="none" stroke="#191970" d="M1379.5807,-660.4976C1212.4802,-616.4041 551.6136,-423.4808 758,-179 811.8994,-115.1519 830.2485,-94.9272 909,-67 936.3994,-57.2835 1837.2721,-24.6723 2050.2123,-17.0862"/>
+<polygon fill="#191970" stroke="#191970" points="2050.5319,-20.5772 2060.401,-16.7237 2050.2829,-13.5816 2050.5319,-20.5772"/>
</g>
<!-- Node18 -->
<g id="node15" class="node">
<title>Node18</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="493.5,-6 493.5,-25 538.5,-25 538.5,-6 493.5,-6"/>
-<text text-anchor="middle" x="516" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="417.5,-6 417.5,-25 462.5,-25 462.5,-6 417.5,-6"/>
+<text text-anchor="middle" x="440" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">utility</text>
</g>
<!-- Node1->Node18 -->
-<g id="edge131" class="edge">
+<g id="edge132" class="edge">
<title>Node1->Node18</title>
-<path fill="none" stroke="#191970" d="M1499.9683,-669.0554C1371.0599,-665.7488 967.7952,-653.2705 841,-624 776.9064,-609.2041 762.1731,-598.705 704,-568 652.0819,-540.5965 307.3347,-315.4822 263,-277 226.0257,-244.9066 190,-243.46 190,-194.5 190,-194.5 190,-194.5 190,-133 190,-71.1145 400.5436,-32.8103 483.2278,-20.1538"/>
-<polygon fill="#191970" stroke="#191970" points="483.9651,-23.5824 493.3351,-18.6374 482.9265,-16.6599 483.9651,-23.5824"/>
+<path fill="none" stroke="#191970" d="M1376.8448,-666.9148C1311.5285,-661.4045 1178.4576,-648.1249 1068,-624 690.1197,-541.4676 539.7219,-593.3064 244,-344 184.3509,-293.7133 152,-272.5177 152,-194.5 152,-194.5 152,-194.5 152,-133 152,-92.7538 179.0805,-87.0095 214,-67 276.1938,-31.3618 361.254,-20.3846 407.2492,-17.004"/>
+<polygon fill="#191970" stroke="#191970" points="407.5091,-20.4945 417.2572,-16.3442 407.0486,-13.5097 407.5091,-20.4945"/>
</g>
<!-- Node20 -->
<g id="node16" class="node">
<title>Node20</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1738.5,-123.5 1738.5,-142.5 1785.5,-142.5 1785.5,-123.5 1738.5,-123.5"/>
-<text text-anchor="middle" x="1762" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1646.5,-123.5 1646.5,-142.5 1693.5,-142.5 1693.5,-123.5 1646.5,-123.5"/>
+<text text-anchor="middle" x="1670" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">vector</text>
</g>
<!-- Node1->Node20 -->
-<g id="edge132" class="edge">
+<g id="edge133" class="edge">
<title>Node1->Node20</title>
-<path fill="none" stroke="#191970" d="M1545.01,-660.2017C1569.4262,-612.1526 1677.3878,-396.2315 1742,-210 1748.6302,-190.8899 1754.2199,-168.4445 1757.7897,-152.7342"/>
-<polygon fill="#191970" stroke="#191970" points="1761.2824,-153.1482 1760.0181,-142.629 1754.4467,-151.6407 1761.2824,-153.1482"/>
+<path fill="none" stroke="#191970" d="M1398.2444,-660.4399C1376.0146,-647.8453 1343.8939,-624.8467 1360,-604 1398.7531,-553.8405 1450.5439,-610.0202 1498,-568 1564.9331,-508.7338 1643.96,-229.6502 1664.7692,-152.6936"/>
+<polygon fill="#191970" stroke="#191970" points="1668.1976,-153.4216 1667.4086,-142.8562 1661.4367,-151.6076 1668.1976,-153.4216"/>
</g>
<!-- Node33 -->
<g id="node22" class="node">
<title>Node33</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="401.5,-123.5 401.5,-142.5 494.5,-142.5 494.5,-123.5 401.5,-123.5"/>
-<text text-anchor="middle" x="448" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">unordered_map</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="454.5,-123.5 454.5,-142.5 547.5,-142.5 547.5,-123.5 454.5,-123.5"/>
+<text text-anchor="middle" x="501" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">unordered_map</text>
</g>
<!-- Node1->Node33 -->
-<g id="edge130" class="edge">
+<g id="edge131" class="edge">
<title>Node1->Node33</title>
-<path fill="none" stroke="#191970" d="M1499.5252,-668.9139C1355.6115,-664.4375 870.2248,-643.9061 738,-568 566.5201,-469.5588 476.8868,-223.659 453.9779,-152.5477"/>
-<polygon fill="#191970" stroke="#191970" points="457.2141,-151.1716 450.8717,-142.6849 450.5374,-153.2744 457.2141,-151.1716"/>
+<path fill="none" stroke="#191970" d="M1376.9801,-665.2733C1296.4582,-654.8452 1110.089,-626.0474 964,-568 873.2594,-531.9449 649.6013,-418.0857 586,-344 535.5532,-285.2372 512.5158,-192.6875 504.4972,-152.5964"/>
+<polygon fill="#191970" stroke="#191970" points="507.8925,-151.7157 502.5864,-142.5454 501.0157,-153.0231 507.8925,-151.7157"/>
</g>
<!-- Node36 -->
<g id="node25" class="node">
<title>Node36</title>
<g id="a_node25"><a xlink:href="structural__hash_8h.html" target="_top" xlink:title="tvm/node/structural\l_hash.h">
-<polygon fill="#ffffff" stroke="#000000" points="2699.5,-313.5 2699.5,-343.5 2812.5,-343.5 2812.5,-313.5 2699.5,-313.5"/>
-<text text-anchor="start" x="2707.5" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
-<text text-anchor="middle" x="2756" y="-320.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_hash.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2275.5,-313.5 2275.5,-343.5 2388.5,-343.5 2388.5,-313.5 2275.5,-313.5"/>
+<text text-anchor="start" x="2283.5" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/structural</text>
+<text text-anchor="middle" x="2332" y="-320.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_hash.h</text>
</a>
</g>
</g>
<!-- Node1->Node36 -->
-<g id="edge125" class="edge">
+<g id="edge126" class="edge">
<title>Node1->Node36</title>
-<path fill="none" stroke="#191970" d="M1580.3516,-661.3331C1621.5749,-652.3777 1687.3868,-637.8006 1744,-624 2064.0664,-545.9776 2142.1362,-518.7999 2461,-436 2523.1841,-419.8525 2541.7821,-424.9167 2601,-400 2616.8077,-393.3487 2618.7263,-387.7999 2634,-380 2657.4695,-368.0147 2684.3809,-356.4374 2706.9467,-347.3057"/>
-<polygon fill="#191970" stroke="#191970" points="2708.3473,-350.515 2716.3289,-343.5477 2705.7445,-344.0169 2708.3473,-350.515"/>
+<path fill="none" stroke="#191970" d="M1435.7291,-660.493C1517.909,-618.8189 1844.6412,-453.5716 1893,-436 2021.4025,-389.3438 2178.2064,-356.4266 2265.3885,-340.1409"/>
+<polygon fill="#191970" stroke="#191970" points="2266.2152,-343.5474 2275.4106,-338.2848 2264.9404,-336.6644 2266.2152,-343.5474"/>
</g>
<!-- Node42 -->
<g id="node27" class="node">
<title>Node42</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1668.5,-185 1668.5,-204 1733.5,-204 1733.5,-185 1668.5,-185"/>
-<text text-anchor="middle" x="1701" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">functional</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1698.5,-185 1698.5,-204 1763.5,-204 1763.5,-185 1698.5,-185"/>
+<text text-anchor="middle" x="1731" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">functional</text>
</g>
<!-- Node1->Node42 -->
-<g id="edge127" class="edge">
+<g id="edge128" class="edge">
<title>Node1->Node42</title>
-<path fill="none" stroke="#191970" d="M1499.782,-662.1875C1478.4123,-655.7706 1453.9463,-644.2911 1441,-624 1436.2189,-616.5064 1438.0106,-612.3711 1441,-604 1480.4065,-493.6528 1537.922,-493.9067 1608,-400 1625.989,-375.8942 1630.1234,-369.5465 1646,-344 1664.0702,-314.9239 1671.9541,-309.0444 1684,-277 1691.6929,-256.5355 1696.1983,-231.8373 1698.6156,-214.8294"/>
-<polygon fill="#191970" stroke="#191970" points="1702.1492,-214.7979 1699.9691,-204.4298 1695.2077,-213.8945 1702.1492,-214.7979"/>
+<path fill="none" stroke="#191970" d="M1429.2051,-660.3598C1446.224,-647.0789 1478.237,-622.6785 1507,-604 1534.0465,-586.4362 1546.3351,-589.9083 1570,-568 1680.2426,-465.9407 1718.4463,-275.5439 1728.2072,-214.1342"/>
+<polygon fill="#191970" stroke="#191970" points="1731.6719,-214.6307 1729.7098,-204.2191 1724.7509,-213.5818 1731.6719,-214.6307"/>
</g>
<!-- Node43 -->
<g id="node28" class="node">
<title>Node43</title>
<g id="a_node28"><a xlink:href="packed__func_8h.html" target="_top" xlink:title="Type-erased function used across TVM API. ">
-<polygon fill="#ffffff" stroke="#000000" points="1164,-313.5 1164,-343.5 1280,-343.5 1280,-313.5 1164,-313.5"/>
-<text text-anchor="start" x="1172" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/packed</text>
-<text text-anchor="middle" x="1222" y="-320.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_func.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="1331,-313.5 1331,-343.5 1447,-343.5 1447,-313.5 1331,-313.5"/>
+<text text-anchor="start" x="1339" y="-331.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/packed</text>
+<text text-anchor="middle" x="1389" y="-320.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_func.h</text>
</a>
</g>
</g>
<!-- Node1->Node43 -->
-<g id="edge126" class="edge">
+<g id="edge127" class="edge">
<title>Node1->Node43</title>
-<path fill="none" stroke="#191970" d="M1499.9299,-663.3239C1471.8346,-657.014 1434.68,-645.2316 1408,-624 1312.6648,-548.1335 1252.376,-409.246 1230.8681,-353.0603"/>
-<polygon fill="#191970" stroke="#191970" points="1234.0951,-351.6974 1227.3037,-343.5655 1227.5416,-354.1576 1234.0951,-351.6974"/>
+<path fill="none" stroke="#191970" d="M1386.5976,-660.3822C1349.5762,-646.7249 1289.3439,-617.6861 1268,-568 1232.7016,-485.8296 1318.3874,-392.2161 1363.4558,-350.53"/>
+<polygon fill="#191970" stroke="#191970" points="1366.0442,-352.9076 1371.0995,-343.5966 1361.3411,-347.7228 1366.0442,-352.9076"/>
</g>
<!-- Node4 -->
<g id="node5" class="node">
<title>Node4</title>
<g id="a_node5"><a xlink:href="ir_2span_8h.html" target="_top" xlink:title="Span information for debugging purposes. ">
-<polygon fill="#ffffff" stroke="#000000" points="2397.5,-492.5 2397.5,-511.5 2478.5,-511.5 2478.5,-492.5 2397.5,-492.5"/>
-<text text-anchor="middle" x="2438" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/span.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2932.5,-492.5 2932.5,-511.5 3013.5,-511.5 3013.5,-492.5 2932.5,-492.5"/>
+<text text-anchor="middle" x="2973" y="-499.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/span.h</text>
</a>
</g>
</g>
<!-- Node3->Node4 -->
<g id="edge4" class="edge">
<title>Node3->Node4</title>
-<path fill="none" stroke="#191970" d="M2211.5197,-605.308C2255.1429,-595.4056 2322.3281,-579.1552 2346,-568 2373.8117,-554.8939 2401.9809,-533.1664 2419.6893,-518.2443"/>
-<polygon fill="#191970" stroke="#191970" points="2422.194,-520.7066 2427.4948,-511.5332 2417.6303,-515.3987 2422.194,-520.7066"/>
+<path fill="none" stroke="#191970" d="M2347.6415,-607.5711C2444.8889,-591.7566 2703.0131,-549.5166 2918,-512 2919.464,-511.7445 2920.9496,-511.4838 2922.449,-511.2195"/>
+<polygon fill="#191970" stroke="#191970" points="2923.2304,-514.6355 2932.4626,-509.4376 2922.0041,-507.7437 2923.2304,-514.6355"/>
</g>
<!-- Node5 -->
<g id="node6" class="node">
<title>Node5</title>
<g id="a_node6"><a xlink:href="node_8h.html" target="_top" xlink:title="Definitions and helper macros for IR/AST nodes. ">
-<polygon fill="#ffffff" stroke="#000000" points="2469.5,-436.5 2469.5,-455.5 2568.5,-455.5 2568.5,-436.5 2469.5,-436.5"/>
-<text text-anchor="middle" x="2519" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/node.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2289.5,-436.5 2289.5,-455.5 2388.5,-455.5 2388.5,-436.5 2289.5,-436.5"/>
+<text text-anchor="middle" x="2339" y="-443.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/node.h</text>
</a>
</g>
</g>
<!-- Node3->Node5 -->
-<g id="edge117" class="edge">
+<g id="edge118" class="edge">
<title>Node3->Node5</title>
-<path fill="none" stroke="#191970" d="M2186.71,-604.4785C2220.9689,-582.5776 2309.167,-527.7659 2388,-492 2417.0991,-478.7979 2451.1888,-466.9339 2477.2977,-458.5623"/>
-<polygon fill="#191970" stroke="#191970" points="2478.4075,-461.8822 2486.8864,-455.5293 2476.2964,-455.2082 2478.4075,-461.8822"/>
+<path fill="none" stroke="#191970" d="M2310.1371,-604.4699C2314.1364,-586.4613 2322.8978,-546.1725 2329,-512 2331.7417,-496.6465 2334.3644,-479.196 2336.2354,-466.0964"/>
+<polygon fill="#191970" stroke="#191970" points="2339.7473,-466.2559 2337.6704,-455.8666 2332.8151,-465.2834 2339.7473,-466.2559"/>
</g>
<!-- Node10 -->
<g id="node10" class="node">
<title>Node10</title>
<g id="a_node10"><a xlink:href="object_8h.html" target="_top" xlink:title="A managed object in the TVM runtime. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="2166.5,-67.5 2166.5,-86.5 2285.5,-86.5 2285.5,-67.5 2166.5,-67.5"/>
-<text text-anchor="middle" x="2226" y="-74.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/object.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="2080.5,-67.5 2080.5,-86.5 2199.5,-86.5 2199.5,-67.5 2080.5,-67.5"/>
+<text text-anchor="middle" x="2140" y="-74.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/object.h</text>
</a>
</g>
</g>
<!-- Node3->Node10 -->
-<g id="edge119" class="edge">
+<g id="edge120" class="edge">
<title>Node3->Node10</title>
-<path fill="none" stroke="#191970" d="M2175.1293,-604.4373C2188.5381,-562.5062 2240.2728,-390.5476 2236,-246 2234.399,-191.8375 2229.9177,-128.0386 2227.5349,-96.5726"/>
-<polygon fill="#191970" stroke="#191970" points="2231.0184,-96.2223 2226.7613,-86.5204 2224.0391,-96.7596 2231.0184,-96.2223"/>
+<path fill="none" stroke="#191970" d="M2273.6875,-604.418C2254.9166,-597.3906 2232.9166,-585.8906 2220,-568 2163.9838,-490.4126 2145.0009,-179.2346 2140.9109,-97.0101"/>
+<polygon fill="#191970" stroke="#191970" points="2144.3866,-96.4164 2140.4129,-86.595 2137.3946,-96.7508 2144.3866,-96.4164"/>
</g>
<!-- Node3->Node16 -->
-<g id="edge122" class="edge">
+<g id="edge123" class="edge">
<title>Node3->Node16</title>
-<path fill="none" stroke="#191970" d="M2211.6394,-612.5943C2436.921,-603.5608 3546,-544.7068 3546,-261.5 3546,-261.5 3546,-261.5 3546,-133 3546,-98.6399 3546,-58.628 3546,-35.2764"/>
-<polygon fill="#191970" stroke="#191970" points="3549.5001,-35.2489 3546,-25.2489 3542.5001,-35.249 3549.5001,-35.2489"/>
+<path fill="none" stroke="#191970" d="M2347.7116,-612.0021C2433.5445,-607.3339 2642.6106,-594.0777 2816,-568 3191.3369,-511.5494 3584,-574.0582 3584,-194.5 3584,-194.5 3584,-194.5 3584,-133 3584,-98.6399 3584,-58.628 3584,-35.2764"/>
+<polygon fill="#191970" stroke="#191970" points="3587.5001,-35.2489 3584,-25.2489 3580.5001,-35.249 3587.5001,-35.2489"/>
</g>
<!-- Node3->Node17 -->
-<g id="edge123" class="edge">
+<g id="edge124" class="edge">
<title>Node3->Node17</title>
-<path fill="none" stroke="#191970" d="M2211.6478,-611.3928C2379.1936,-599.8581 3028.7602,-549.7899 3208,-456 3269.8189,-423.6523 3296.3184,-410.3164 3318,-344 3382.8208,-145.7359 3309.2615,-275.2781 2942,-123 2868.5168,-92.5316 2849.8653,-83.2817 2772,-67 2588.7095,-28.6739 2364.7942,-18.8401 2270.8514,-16.3406"/>
-<polygon fill="#191970" stroke="#191970" points="2270.6633,-12.8349 2260.5787,-16.0831 2270.4878,-19.8327 2270.6633,-12.8349"/>
+<path fill="none" stroke="#191970" d="M2347.6686,-609.2438C2413.6132,-601.2575 2549.93,-584.4103 2665,-568 2950.2296,-527.323 3304.5873,-597.7994 3261,-313 3251.558,-251.3059 3264.4761,-221.7462 3219,-179 3067.5692,-36.6596 2970.196,-99.9651 2765,-67 2531.7415,-29.5266 2248.2671,-19.1133 2139.9832,-16.3976"/>
+<polygon fill="#191970" stroke="#191970" points="2139.8383,-12.8932 2129.7569,-16.1512 2139.6696,-19.8912 2139.8383,-12.8932"/>
</g>
<!-- Node28 -->
<g id="node18" class="node">
<title>Node28</title>
<g id="a_node18"><a xlink:href="string_8h.html" target="_top" xlink:title="Runtime String container types. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="1030,-179.5 1030,-209.5 1156,-209.5 1156,-179.5 1030,-179.5"/>
-<text text-anchor="start" x="1038" y="-197.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="1093" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/string.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="984,-179.5 984,-209.5 1110,-209.5 1110,-179.5 984,-179.5"/>
+<text text-anchor="start" x="992" y="-197.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
+<text text-anchor="middle" x="1047" y="-186.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/string.h</text>
</a>
</g>
</g>
<!-- Node3->Node28 -->
-<g id="edge118" class="edge">
+<g id="edge119" class="edge">
<title>Node3->Node28</title>
-<path fill="none" stroke="#191970" d="M2132.3052,-613.0687C1935.5208,-608.3657 1073.2237,-586.7159 1053,-568 953.0927,-475.5414 1042.701,-285.546 1079.1957,-218.5395"/>
-<polygon fill="#191970" stroke="#191970" points="1082.3795,-220.0151 1084.1659,-209.572 1076.257,-216.6217 1082.3795,-220.0151"/>
+<path fill="none" stroke="#191970" d="M2268.2944,-611.4734C2234.2579,-609.3466 2183.9137,-606.2937 2140,-604 2095.5157,-601.6765 1374.4508,-594.972 1339,-568 1289.8956,-530.6399 1326.7805,-489.475 1296,-436 1259.4138,-372.4388 1238.4694,-364.24 1186,-313 1149.2447,-277.1059 1103.3419,-239.1844 1074.5896,-216.1767"/>
+<polygon fill="#191970" stroke="#191970" points="1076.6046,-213.3072 1066.6017,-209.8156 1072.244,-218.783 1076.6046,-213.3072"/>
</g>
<!-- Node26 -->
<g id="node20" class="node">
<title>Node26</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1494,-123.5 1494,-142.5 1558,-142.5 1558,-123.5 1494,-123.5"/>
-<text text-anchor="middle" x="1526" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1246,-123.5 1246,-142.5 1310,-142.5 1310,-123.5 1246,-123.5"/>
+<text text-anchor="middle" x="1278" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">algorithm</text>
</g>
<!-- Node3->Node26 -->
-<g id="edge120" class="edge">
+<g id="edge121" class="edge">
<title>Node3->Node26</title>
-<path fill="none" stroke="#191970" d="M2161.0746,-604.4083C2141.8526,-586.4089 2104,-545.4481 2104,-502 2104,-502 2104,-502 2104,-390 2104,-280.1071 1999.0152,-299.4558 1903,-246 1835.2832,-208.2992 1815.9947,-202.0587 1742,-179 1682.8655,-160.5721 1612.3482,-147.0837 1568.3632,-139.6392"/>
-<polygon fill="#191970" stroke="#191970" points="1568.7547,-136.1562 1558.3156,-137.9658 1567.6047,-143.0611 1568.7547,-136.1562"/>
+<path fill="none" stroke="#191970" d="M2300.5238,-604.4439C2257.0826,-549.7769 2030.1109,-275.7348 1772,-179 1689.6353,-148.1313 1424.2705,-137.2385 1320.3226,-134.0954"/>
+<polygon fill="#191970" stroke="#191970" points="1320.2963,-130.5932 1310.198,-133.7985 1320.091,-137.5902 1320.2963,-130.5932"/>
</g>
<!-- Node47 -->
<g id="node31" class="node">
<title>Node47</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1850,-252 1850,-271 1894,-271 1894,-252 1850,-252"/>
-<text text-anchor="middle" x="1872" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1812,-252 1812,-271 1856,-271 1856,-252 1812,-252"/>
+<text text-anchor="middle" x="1834" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">limits</text>
</g>
<!-- Node3->Node47 -->
-<g id="edge121" class="edge">
+<g id="edge122" class="edge">
<title>Node3->Node47</title>
-<path fill="none" stroke="#191970" d="M2138.9635,-604.4779C2096.4498,-589.958 2028,-557.9003 2028,-502 2028,-502 2028,-502 2028,-390 2028,-356.7503 1942.5027,-302.3051 1898.0972,-276.2698"/>
-<polygon fill="#191970" stroke="#191970" points="1899.6973,-273.1518 1889.2908,-271.1632 1896.1858,-279.2074 1899.6973,-273.1518"/>
+<path fill="none" stroke="#191970" d="M2294.8278,-604.2042C2229.1767,-555.3814 1937.2186,-338.2607 1855.3858,-277.404"/>
+<polygon fill="#191970" stroke="#191970" points="1856.9986,-274.2416 1846.8856,-271.0827 1852.8213,-279.8587 1856.9986,-274.2416"/>
</g>
<!-- Node51 -->
<g id="node35" class="node">
<title>Node51</title>
<g id="a_node35"><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="2398,-548.5 2398,-567.5 2478,-567.5 2478,-548.5 2398,-548.5"/>
-<text text-anchor="middle" x="2438" y="-555.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2078,-548.5 2078,-567.5 2158,-567.5 2158,-548.5 2078,-548.5"/>
+<text text-anchor="middle" x="2118" y="-555.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/ir/type.h</text>
</a>
</g>
</g>
<!-- Node3->Node51 -->
-<g id="edge110" class="edge">
+<g id="edge111" class="edge">
<title>Node3->Node51</title>
-<path fill="none" stroke="#191970" d="M2211.5357,-605.6767C2258.4809,-595.7935 2336.8404,-579.2968 2387.9666,-568.5333"/>
-<polygon fill="#191970" stroke="#191970" points="2388.7469,-571.9459 2397.8113,-566.4608 2387.3048,-565.096 2388.7469,-571.9459"/>
+<path fill="none" stroke="#191970" d="M2275.3378,-604.3733C2243.4702,-594.9807 2194.8469,-580.6496 2160.093,-570.4064"/>
+<polygon fill="#191970" stroke="#191970" points="2160.8175,-566.9711 2150.236,-567.5011 2158.8385,-573.6855 2160.8175,-566.9711"/>
</g>
<!-- Node4->Node5 -->
<g id="edge5" class="edge">
<title>Node4->Node5</title>
-<path fill="none" stroke="#191970" d="M2452.1091,-492.2455C2464.4209,-483.7337 2482.4958,-471.2375 2496.7538,-461.3801"/>
-<polygon fill="#191970" stroke="#191970" points="2498.8174,-464.2085 2505.0526,-455.6427 2494.8366,-458.4506 2498.8174,-464.2085"/>
+<path fill="none" stroke="#191970" d="M2932.376,-498.4118C2823.809,-488.8222 2526.5172,-462.563 2398.7034,-451.2735"/>
+<polygon fill="#191970" stroke="#191970" points="2398.9686,-447.7834 2388.6994,-450.3898 2398.3526,-454.7562 2398.9686,-447.7834"/>
</g>
<!-- Node4->Node10 -->
-<g id="edge108" class="edge">
+<g id="edge109" class="edge">
<title>Node4->Node10</title>
-<path fill="none" stroke="#191970" d="M2418.0622,-492.3628C2387.3566,-475.9628 2332,-439.5285 2332,-390 2332,-390 2332,-390 2332,-328.5 2332,-234.066 2327.5007,-203.4205 2278,-123 2270.6947,-111.1316 2259.3806,-100.6003 2249.1692,-92.6194"/>
-<polygon fill="#191970" stroke="#191970" points="2251.0823,-89.681 2240.9632,-86.5419 2246.9162,-95.3062 2251.0823,-89.681"/>
+<path fill="none" stroke="#191970" d="M2966.025,-492.4841C2944.2715,-463.2287 2874.8167,-372.9686 2803,-313 2667.1391,-199.553 2623.4805,-170.5898 2453,-123 2371.1495,-100.1513 2273.9043,-88.1185 2209.7716,-82.1739"/>
+<polygon fill="#191970" stroke="#191970" points="2210.0445,-78.6843 2199.7704,-81.2713 2209.4153,-85.656 2210.0445,-78.6843"/>
</g>
<!-- Node4->Node16 -->
-<g id="edge109" class="edge">
+<g id="edge110" class="edge">
<title>Node4->Node16</title>
-<path fill="none" stroke="#191970" d="M2478.6083,-500.2483C2624.2172,-493.4661 3118.9976,-465.9576 3261,-400 3334.0218,-366.0827 3351.5214,-345.3969 3394,-277 3445.343,-194.3301 3395.6393,-143.3348 3456,-67 3470.9969,-48.0343 3495.0303,-34.7937 3514.4904,-26.4912"/>
-<polygon fill="#191970" stroke="#191970" points="3515.9286,-29.6852 3523.8965,-22.7021 3513.313,-23.1922 3515.9286,-29.6852"/>
+<path fill="none" stroke="#191970" d="M3013.759,-493.4669C3116.8819,-471.1126 3384.3441,-408.1272 3446,-344 3472.4664,-316.4727 3470,-299.6867 3470,-261.5 3470,-261.5 3470,-261.5 3470,-133 3470,-84.5473 3520.5515,-48.5212 3554.3292,-29.8573"/>
+<polygon fill="#191970" stroke="#191970" points="3556.1693,-32.843 3563.3555,-25.0577 3552.8828,-26.6624 3556.1693,-32.843"/>
</g>
<!-- Node6 -->
<g id="node7" class="node">
<title>Node6</title>
<g id="a_node7"><a xlink:href="reflection_8h.html" target="_top" xlink:title="Reflection and serialization of compiler IR/AST nodes. ">
-<polygon fill="#ffffff" stroke="#000000" points="2470.5,-380.5 2470.5,-399.5 2591.5,-399.5 2591.5,-380.5 2470.5,-380.5"/>
-<text text-anchor="middle" x="2531" y="-387.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/reflection.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2356.5,-380.5 2356.5,-399.5 2477.5,-399.5 2477.5,-380.5 2356.5,-380.5"/>
+<text text-anchor="middle" x="2417" y="-387.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/reflection.h</text>
</a>
</g>
</g>
<!-- Node5->Node6 -->
<g id="edge6" class="edge">
<title>Node5->Node6</title>
-<path fill="none" stroke="#191970" d="M2521.0902,-436.2455C2522.6561,-428.9382 2524.8512,-418.6944 2526.7776,-409.7046"/>
-<polygon fill="#191970" stroke="#191970" points="2530.2607,-410.1541 2528.9337,-399.6427 2523.416,-408.6873 2530.2607,-410.1541"/>
+<path fill="none" stroke="#191970" d="M2352.5866,-436.2455C2364.3305,-427.814 2381.5199,-415.4729 2395.1882,-405.6598"/>
+<polygon fill="#191970" stroke="#191970" points="2397.4871,-408.3179 2403.5691,-399.6427 2393.4046,-402.6316 2397.4871,-408.3179"/>
</g>
<!-- Node5->Node7 -->
-<g id="edge99" class="edge">
+<g id="edge100" class="edge">
<title>Node5->Node7</title>
-<path fill="none" stroke="#191970" d="M2500.1179,-436.4487C2479.3843,-424.5627 2450.4629,-403.0074 2462,-380 2468.3005,-367.4355 2479.2818,-357.1529 2490.6552,-349.1711"/>
-<polygon fill="#191970" stroke="#191970" points="2492.8797,-351.8996 2499.3579,-343.5161 2489.0656,-346.03 2492.8797,-351.8996"/>
+<path fill="none" stroke="#191970" d="M2386.9997,-436.4218C2423.9157,-428.4305 2475.8987,-415.8129 2520,-400 2560.5508,-385.4601 2604.9137,-363.8934 2635.1875,-348.2504"/>
+<polygon fill="#191970" stroke="#191970" points="2636.8602,-351.3256 2644.1107,-343.6002 2633.6251,-345.1179 2636.8602,-351.3256"/>
</g>
<!-- Node5->Node10 -->
-<g id="edge103" class="edge">
+<g id="edge104" class="edge">
<title>Node5->Node10</title>
-<path fill="none" stroke="#191970" d="M2501.5692,-436.4757C2487.6581,-428.2202 2468.3541,-415.2246 2455,-400 2414.0902,-353.3601 2416.9623,-332.3805 2389,-277 2354.8604,-209.3851 2368.7546,-177.3529 2316,-123 2301.9758,-108.5509 2282.4412,-97.8128 2265.1539,-90.362"/>
-<polygon fill="#191970" stroke="#191970" points="2266.4116,-87.0951 2255.8304,-86.5582 2263.7673,-93.5765 2266.4116,-87.0951"/>
+<path fill="none" stroke="#191970" d="M2329.0679,-436.4544C2312.3791,-419.6435 2279.3276,-382.8409 2267,-344 2253.8176,-302.4658 2281.1167,-288.1341 2270,-246 2261.2666,-212.899 2250.0702,-208.0761 2232,-179 2216.1234,-153.4535 2214.0742,-145.3991 2194,-123 2184.2396,-112.1091 2171.9149,-101.4242 2161.5051,-93.0975"/>
+<polygon fill="#191970" stroke="#191970" points="2163.3472,-90.0956 2153.3096,-86.705 2159.042,-95.6151 2163.3472,-90.0956"/>
</g>
<!-- Node11 -->
<g id="node11" class="node">
<title>Node11</title>
<g id="a_node11"><a xlink:href="c__runtime__api_8h.html" target="_top" xlink:title="tvm/runtime/c_runtime\l_api.h">
-<polygon fill="#ffffff" stroke="#ff0000" points="2826.5,-.5 2826.5,-30.5 2955.5,-30.5 2955.5,-.5 2826.5,-.5"/>
-<text text-anchor="start" x="2834.5" y="-18.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/c_runtime</text>
-<text text-anchor="middle" x="2891" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="2722.5,-.5 2722.5,-30.5 2851.5,-30.5 2851.5,-.5 2722.5,-.5"/>
+<text text-anchor="start" x="2730.5" y="-18.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/c_runtime</text>
+<text text-anchor="middle" x="2787" y="-7.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">_api.h</text>
</a>
</g>
</g>
<!-- Node5->Node11 -->
-<g id="edge101" class="edge">
+<g id="edge102" class="edge">
<title>Node5->Node11</title>
-<path fill="none" stroke="#191970" d="M2568.5692,-441.1376C2625.9445,-434.7907 2722.9395,-421.8366 2804,-400 2871.1061,-381.9226 2892.2977,-382.7358 2950,-344 2983.9367,-321.2181 2997.9407,-314.9991 3013,-277 3048.3104,-187.9014 3028.9386,-137.4866 2964,-67 2953.1559,-55.2294 2939.1934,-44.5805 2926.3884,-36.0614"/>
-<polygon fill="#191970" stroke="#191970" points="2928.0655,-32.9787 2917.7608,-30.5162 2924.2807,-38.8673 2928.0655,-32.9787"/>
+<path fill="none" stroke="#191970" d="M2388.8396,-442.7285C2492.5472,-435.6552 2728.4392,-418.1094 2807,-400 2876.7485,-383.922 2900.0933,-386.0734 2958,-344 3037.8915,-285.953 3030.8844,-208.227 2981,-123 2954.271,-77.3338 2900.3387,-49.7186 2856.1093,-33.9336"/>
+<polygon fill="#191970" stroke="#191970" points="2857.1392,-30.5867 2846.5446,-30.6475 2854.8648,-37.2069 2857.1392,-30.5867"/>
</g>
<!-- Node5->Node16 -->
-<g id="edge104" class="edge">
+<g id="edge105" class="edge">
<title>Node5->Node16</title>
-<path fill="none" stroke="#191970" d="M2568.7756,-440.7359C2634.585,-433.4226 2754.0167,-418.9242 2855,-400 2869.4563,-397.2909 3359.6507,-287.4505 3370,-277 3437.3685,-208.9727 3355.8483,-139.8244 3418,-67 3442.2359,-38.6023 3484.4065,-25.6937 3513.6901,-19.94"/>
-<polygon fill="#191970" stroke="#191970" points="3514.6093,-23.3313 3523.8387,-18.1284 3513.3792,-16.4402 3514.6093,-23.3313"/>
+<path fill="none" stroke="#191970" d="M2388.5324,-445.324C2548.2887,-442.8287 3043.2765,-432.5263 3107,-400 3128.3072,-389.1242 3359.8043,-79.8222 3380,-67 3434.0139,-32.7067 3509.6143,-21.2316 3551.9763,-17.4041"/>
+<polygon fill="#191970" stroke="#191970" points="3552.3145,-20.8882 3561.9968,-16.5865 3551.7452,-13.9114 3552.3145,-20.8882"/>
</g>
<!-- Node5->Node17 -->
-<g id="edge105" class="edge">
+<g id="edge106" class="edge">
<title>Node5->Node17</title>
-<path fill="none" stroke="#191970" d="M2568.6503,-439.0845C2710.162,-417.249 3098.3642,-341.3905 2987,-179 2904.6214,-58.876 2421.3413,-25.1219 2270.7701,-17.4572"/>
-<polygon fill="#191970" stroke="#191970" points="2270.6749,-13.9482 2260.514,-16.9495 2270.3288,-20.9397 2270.6749,-13.9482"/>
+<path fill="none" stroke="#191970" d="M2388.6522,-444.0841C2562.5406,-435.5225 3125.5398,-391.1831 2975,-179 2876.7873,-40.5711 2305.3931,-19.3371 2139.898,-16.0862"/>
+<polygon fill="#191970" stroke="#191970" points="2139.656,-12.5813 2129.5934,-15.8971 2139.5275,-19.5801 2139.656,-12.5813"/>
</g>
<!-- Node5->Node18 -->
-<g id="edge106" class="edge">
+<g id="edge107" class="edge">
<title>Node5->Node18</title>
-<path fill="none" stroke="#191970" d="M2469.2158,-445.1811C2275.2668,-441.8145 1570.0571,-427.8839 1348,-400 1229.8532,-385.1642 1202.2256,-369.8916 1086,-344 957.3037,-315.3303 913.4208,-338.8941 797,-277 675.1085,-212.1974 566.4909,-81.0466 529.4759,-33.3531"/>
-<polygon fill="#191970" stroke="#191970" points="532.0923,-31.0134 523.2271,-25.2118 526.5394,-35.2756 532.0923,-31.0134"/>
+<path fill="none" stroke="#191970" d="M2289.1254,-444.8953C2050.1394,-438.981 1021.8279,-406.3147 733,-277 716.2101,-269.4828 519.4365,-88.7389 458.1236,-32.2263"/>
+<polygon fill="#191970" stroke="#191970" points="460.1281,-29.3137 450.4039,-25.1075 455.3826,-34.4597 460.1281,-29.3137"/>
</g>
<!-- Node5->Node20 -->
-<g id="edge107" class="edge">
+<g id="edge108" class="edge">
<title>Node5->Node20</title>
-<path fill="none" stroke="#191970" d="M2478.9669,-436.4257C2445.6546,-428.0501 2397.1952,-414.914 2356,-400 2266.7667,-367.6947 2248.4768,-349.7972 2161,-313 2120.9084,-296.1354 2106.7066,-300.3319 2070,-277 2016.4009,-242.9307 2022.3706,-208.2564 1966,-179 1898.0407,-143.729 1869.286,-162.7537 1795.4797,-144.3199"/>
-<polygon fill="#191970" stroke="#191970" points="1796.3189,-140.9216 1785.7559,-141.742 1794.525,-147.6878 1796.3189,-140.9216"/>
+<path fill="none" stroke="#191970" d="M2301.3858,-436.4735C2244.0337,-421.1881 2132.0973,-388.34 2044,-344 1996.871,-320.2796 1986.9416,-310.2804 1946,-277 1897.2174,-237.3458 1898.8071,-209.7957 1844,-179 1799.2485,-153.8545 1740.2774,-142.0386 1703.7665,-136.7997"/>
+<polygon fill="#191970" stroke="#191970" points="1704.0244,-133.3029 1693.6461,-135.434 1703.0882,-140.24 1704.0244,-133.3029"/>
</g>
<!-- Node24 -->
<g id="node19" class="node">
<title>Node24</title>
<g id="a_node19"><a xlink:href="runtime_2memory_8h.html" target="_top" xlink:title="Runtime memory management. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="1158.5,-123.5 1158.5,-142.5 1287.5,-142.5 1287.5,-123.5 1158.5,-123.5"/>
-<text text-anchor="middle" x="1223" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="1984.5,-123.5 1984.5,-142.5 2113.5,-142.5 2113.5,-123.5 1984.5,-123.5"/>
+<text text-anchor="middle" x="2049" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/memory.h</text>
</a>
</g>
</g>
<!-- Node5->Node24 -->
-<g id="edge102" class="edge">
+<g id="edge103" class="edge">
<title>Node5->Node24</title>
-<path fill="none" stroke="#191970" d="M2469.4182,-444.05C2225.9348,-434.267 1166.4428,-389.2139 1114,-344 1080.6278,-315.228 1076.5075,-283.8896 1099,-246 1116.0563,-217.268 1139.7105,-231.8377 1165,-210 1184.6954,-192.9929 1201.8618,-168.2404 1212.3059,-151.4247"/>
-<polygon fill="#191970" stroke="#191970" points="1215.4843,-152.9313 1217.6467,-142.5595 1209.4883,-149.319 1215.4843,-152.9313"/>
+<path fill="none" stroke="#191970" d="M2304.3603,-436.4163C2282.2483,-429.057 2253.9143,-417.2058 2233,-400 2209.8318,-380.94 2097.6676,-208.4927 2060.7831,-151.3211"/>
+<polygon fill="#191970" stroke="#191970" points="2063.5278,-149.1187 2055.1694,-142.6081 2057.6434,-152.9101 2063.5278,-149.1187"/>
</g>
<!-- Node5->Node36 -->
-<g id="edge100" class="edge">
+<g id="edge101" class="edge">
<title>Node5->Node36</title>
-<path fill="none" stroke="#191970" d="M2546.4511,-436.3385C2570.0677,-427.7168 2604.8379,-414.2783 2634,-400 2665.4775,-384.588 2699.7879,-364.1298 2723.9824,-349.0464"/>
-<polygon fill="#191970" stroke="#191970" points="2726.1262,-351.8331 2732.7329,-343.5505 2722.4031,-345.9053 2726.1262,-351.8331"/>
+<path fill="none" stroke="#191970" d="M2338.4272,-436.3845C2337.3643,-418.544 2335.0552,-379.7839 2333.514,-353.9138"/>
+<polygon fill="#191970" stroke="#191970" points="2336.995,-353.4884 2332.9064,-343.7143 2330.0074,-353.9048 2336.995,-353.4884"/>
</g>
<!-- Node49 -->
<g id="node33" class="node">
<title>Node49</title>
<g id="a_node33"><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="3120.5,-380.5 3120.5,-399.5 3251.5,-399.5 3251.5,-380.5 3120.5,-380.5"/>
-<text text-anchor="middle" x="3186" y="-387.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/repr_printer.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2966.5,-380.5 2966.5,-399.5 3097.5,-399.5 3097.5,-380.5 2966.5,-380.5"/>
+<text text-anchor="middle" x="3032" y="-387.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/repr_printer.h</text>
</a>
</g>
</g>
<!-- Node5->Node49 -->
-<g id="edge96" class="edge">
+<g id="edge97" class="edge">
<title>Node5->Node49</title>
-<path fill="none" stroke="#191970" d="M2568.5449,-442.8924C2669.6911,-436.3773 2907.1798,-420.2056 3106,-400 3107.2924,-399.8687 3108.5963,-399.7337 3109.9091,-399.5955"/>
-<polygon fill="#191970" stroke="#191970" points="3110.5456,-403.0469 3120.1062,-398.4812 3109.7851,-396.0883 3110.5456,-403.0469"/>
+<path fill="none" stroke="#191970" d="M2388.6153,-442.11C2488.2197,-434.2831 2719.6253,-416.0111 2914,-400 2927.6852,-398.8727 2942.2614,-397.653 2956.3322,-396.4658"/>
+<polygon fill="#191970" stroke="#191970" points="2956.799,-399.9389 2966.4686,-395.6088 2956.2093,-392.9637 2956.799,-399.9389"/>
</g>
<!-- Node6->Node7 -->
<g id="edge7" class="edge">
<title>Node6->Node7</title>
-<path fill="none" stroke="#191970" d="M2531,-380.3906C2531,-373.2234 2531,-363.0862 2531,-353.671"/>
-<polygon fill="#191970" stroke="#191970" points="2534.5001,-353.5446 2531,-343.5446 2527.5001,-353.5446 2534.5001,-353.5446"/>
+<path fill="none" stroke="#191970" d="M2456.564,-380.4581C2496.2671,-370.8826 2558.1511,-355.9577 2605.4257,-344.5562"/>
+<polygon fill="#191970" stroke="#191970" points="2606.3779,-347.9269 2615.2786,-342.1799 2604.7367,-341.122 2606.3779,-347.9269"/>
</g>
<!-- Node6->Node10 -->
-<g id="edge63" class="edge">
+<g id="edge64" class="edge">
<title>Node6->Node10</title>
-<path fill="none" stroke="#191970" d="M2554.7859,-380.4178C2569.8669,-372.9442 2588.1462,-360.9629 2597,-344 2603.3752,-331.7859 2601.8118,-325.9102 2597,-313 2581.0271,-270.1438 2442.7744,-138.5376 2415,-123 2393.4459,-110.9422 2334.823,-97.6464 2288.5268,-88.4635"/>
-<polygon fill="#191970" stroke="#191970" points="2289.0406,-84.9977 2278.5544,-86.5106 2287.6953,-91.8672 2289.0406,-84.9977"/>
+<path fill="none" stroke="#191970" d="M2419.4732,-380.1574C2422.7596,-365.1553 2427.2516,-335.8966 2419,-313 2392.7804,-240.2458 2364.2137,-232.1099 2308,-179 2277.5018,-150.1858 2267.784,-144.9067 2232,-123 2212.3769,-110.9869 2189.2021,-99.414 2171.0213,-90.8811"/>
+<polygon fill="#191970" stroke="#191970" points="2172.2974,-87.6151 2161.7522,-86.5902 2169.3566,-93.9674 2172.2974,-87.6151"/>
</g>
<!-- Node6->Node11 -->
-<g id="edge59" class="edge">
+<g id="edge60" class="edge">
<title>Node6->Node11</title>
-<path fill="none" stroke="#191970" d="M2591.6939,-386.0287C2672.4042,-379.9604 2810.1028,-366.6356 2855,-344 2892.3238,-325.1826 2971.1495,-248.6771 2987,-210 3013.3295,-145.753 2951.9298,-73.1545 2915.5166,-37.564"/>
-<polygon fill="#191970" stroke="#191970" points="2917.8163,-34.9206 2908.1662,-30.5474 2912.9829,-39.984 2917.8163,-34.9206"/>
+<path fill="none" stroke="#191970" d="M2477.6475,-385.7955C2566.1806,-379.0697 2726.1296,-364.5271 2780,-344 2878.2648,-306.5567 2926.0342,-303.0609 2975,-210 3015.2573,-133.4898 2900.3985,-66.4388 2832.696,-34.8072"/>
+<polygon fill="#191970" stroke="#191970" points="2834.0259,-31.5667 2823.4769,-30.5817 2831.1092,-37.9301 2834.0259,-31.5667"/>
</g>
<!-- Node6->Node16 -->
-<g id="edge93" class="edge">
+<g id="edge94" class="edge">
<title>Node6->Node16</title>
-<path fill="none" stroke="#191970" d="M2591.565,-385.3078C2679.0902,-378.0065 2836.3285,-362.7928 2890,-344 2915.0352,-335.2341 2916.8381,-323.9456 2941,-313 3100.9871,-240.5239 3318,-370.1378 3318,-194.5 3318,-194.5 3318,-194.5 3318,-133 3318,-101.7875 3318.8384,-87.9227 3342,-67 3367.1161,-44.3117 3462.7714,-27.4839 3513.6777,-19.9212"/>
-<polygon fill="#191970" stroke="#191970" points="3514.4263,-23.3493 3523.8199,-18.4492 3513.4208,-16.4219 3514.4263,-23.3493"/>
+<path fill="none" stroke="#191970" d="M2477.503,-386.126C2576.7721,-379.3039 2769.7209,-363.9587 2835,-344 3013.2183,-289.5106 3047.914,-246.3402 3203,-143 3250.0459,-111.6514 3253.1845,-89.6097 3305,-67 3388.8372,-30.4176 3497.858,-19.8094 3551.408,-16.7418"/>
+<polygon fill="#191970" stroke="#191970" points="3551.8547,-20.2234 3561.6596,-16.2092 3551.4914,-13.2329 3551.8547,-20.2234"/>
</g>
<!-- Node6->Node17 -->
-<g id="edge94" class="edge">
+<g id="edge95" class="edge">
<title>Node6->Node17</title>
-<path fill="none" stroke="#191970" d="M2552.0319,-380.4794C2609.4499,-353.058 2761.6331,-270.0229 2716,-179 2676.278,-99.7677 2634.0474,-97.9635 2551,-67 2455.4204,-31.3639 2335.0602,-20.3851 2270.9882,-17.0038"/>
-<polygon fill="#191970" stroke="#191970" points="2270.9839,-13.4995 2260.8248,-16.5068 2270.6419,-20.4912 2270.9839,-13.4995"/>
+<path fill="none" stroke="#191970" d="M2441.3391,-380.4022C2477.5899,-364.7171 2544.9292,-329.9355 2577,-277 2612.9301,-217.6946 2599.2602,-173.7403 2552,-123 2522.6669,-91.5069 2513.0307,-82.8047 2473,-67 2413.1367,-43.3651 2226.3064,-25.9343 2139.7925,-18.9096"/>
+<polygon fill="#191970" stroke="#191970" points="2140.0024,-15.4152 2129.7546,-18.1048 2139.4429,-22.3929 2140.0024,-15.4152"/>
</g>
<!-- Node6->Node20 -->
-<g id="edge95" class="edge">
+<g id="edge96" class="edge">
<title>Node6->Node20</title>
-<path fill="none" stroke="#191970" d="M2507.6314,-380.4793C2467.4554,-364.4416 2382.4247,-331.9394 2308,-313 2214.1067,-289.1063 2169.4434,-336.5254 2093,-277 2055.4719,-247.7774 2087.6377,-208.0814 2050,-179 2010.8268,-148.7322 1863.4693,-137.8653 1796.0498,-134.4206"/>
-<polygon fill="#191970" stroke="#191970" points="1796.1027,-130.9191 1785.9446,-133.9296 1795.7629,-137.9109 1796.1027,-130.9191"/>
+<path fill="none" stroke="#191970" d="M2356.1154,-380.4262C2307.9508,-372.315 2239.1884,-359.5383 2180,-344 2088.9206,-320.0896 2056.6694,-330.2413 1979,-277 1932.3918,-245.0507 1946.479,-208.0323 1898,-179 1865.6166,-159.6067 1758.9372,-144.0063 1703.7455,-137.0017"/>
+<polygon fill="#191970" stroke="#191970" points="1704.1058,-133.5196 1693.7497,-135.7557 1703.2399,-140.4658 1704.1058,-133.5196"/>
</g>
<!-- Node6->Node24 -->
-<g id="edge61" class="edge">
+<g id="edge62" class="edge">
<title>Node6->Node24</title>
-<path fill="none" stroke="#191970" d="M2470.3117,-389.5534C2211.7596,-387.449 1212.685,-377.1921 1155,-344 1113.927,-320.3665 1086.8295,-285.5049 1113,-246 1135.6577,-211.7978 1170.2742,-240.2397 1198,-210 1212.5704,-194.1084 1218.6557,-169.6144 1221.193,-152.5333"/>
-<polygon fill="#191970" stroke="#191970" points="1224.6751,-152.892 1222.4016,-142.544 1217.7257,-152.0512 1224.6751,-152.892"/>
+<path fill="none" stroke="#191970" d="M2417.0746,-380.3858C2416.7083,-364.6407 2413.8689,-333.1014 2398,-313 2321.2884,-215.8282 2177.7945,-165.9414 2100.8648,-145.1515"/>
+<polygon fill="#191970" stroke="#191970" points="2101.4796,-141.694 2090.9175,-142.5255 2099.6929,-148.4621 2101.4796,-141.694"/>
</g>
<!-- Node35 -->
<g id="node24" class="node">
<title>Node35</title>
<g id="a_node24"><a xlink:href="data__type_8h.html" target="_top" xlink:title="tvm/runtime/data_type.h">
-<polygon fill="#ffffff" stroke="#000000" points="2840,-185 2840,-204 2978,-204 2978,-185 2840,-185"/>
-<text text-anchor="middle" x="2909" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/data_type.h</text>
+<polygon fill="#ffffff" stroke="#000000" points="2828,-185 2828,-204 2966,-204 2966,-185 2828,-185"/>
+<text text-anchor="middle" x="2897" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/data_type.h</text>
</a>
</g>
</g>
<!-- Node6->Node35 -->
-<g id="edge60" class="edge">
+<g id="edge61" class="edge">
<title>Node6->Node35</title>
-<path fill="none" stroke="#191970" d="M2591.5554,-384.3979C2664.4769,-376.9671 2782.1355,-362.5934 2822,-344 2832.278,-339.2062 2893.6859,-287.0189 2899,-277 2909.3705,-257.4481 2910.9033,-231.7056 2910.4661,-214.2117"/>
-<polygon fill="#191970" stroke="#191970" points="2913.9537,-213.8711 2909.9674,-204.0548 2906.9621,-214.2144 2913.9537,-213.8711"/>
+<path fill="none" stroke="#191970" d="M2477.5979,-384.658C2557.0648,-377.08 2691.8057,-362.0981 2738,-344 2786.785,-324.8869 2797.2173,-313.3009 2835,-277 2855.3851,-257.4144 2874.1939,-230.6324 2885.6258,-212.9646"/>
+<polygon fill="#191970" stroke="#191970" points="2888.7888,-214.5116 2891.1811,-204.1904 2882.8746,-210.767 2888.7888,-214.5116"/>
</g>
<!-- Node6->Node36 -->
-<g id="edge46" class="edge">
+<g id="edge47" class="edge">
<title>Node6->Node36</title>
-<path fill="none" stroke="#191970" d="M2566.1563,-380.3906C2599.4973,-371.2774 2650.4047,-357.3627 2691.1129,-346.2358"/>
-<polygon fill="#191970" stroke="#191970" points="2692.2355,-349.5574 2700.9588,-343.5446 2690.3898,-342.8051 2692.2355,-349.5574"/>
+<path fill="none" stroke="#191970" d="M2403.7188,-380.3906C2392.3216,-372.1444 2375.4908,-359.9669 2360.9977,-349.4807"/>
+<polygon fill="#191970" stroke="#191970" points="2362.9468,-346.5709 2352.7933,-343.5446 2358.8435,-352.2421 2362.9468,-346.5709"/>
</g>
<!-- Node37 -->
<g id="node26" class="node">
<title>Node37</title>
<g id="a_node26"><a xlink:href="ndarray_8h.html" target="_top" xlink:title="A device-independent managed NDArray abstraction. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="2102.5,-252 2102.5,-271 2227.5,-271 2227.5,-252 2102.5,-252"/>
-<text text-anchor="middle" x="2165" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/ndarray.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="1988.5,-252 1988.5,-271 2113.5,-271 2113.5,-252 1988.5,-252"/>
+<text text-anchor="middle" x="2051" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/ndarray.h</text>
</a>
</g>
</g>
<!-- Node6->Node37 -->
-<g id="edge62" class="edge">
+<g id="edge63" class="edge">
<title>Node6->Node37</title>
-<path fill="none" stroke="#191970" d="M2517.2935,-380.2903C2488.1594,-359.6551 2422.6617,-313.2848 2422,-313 2362.8035,-287.5246 2290.4501,-274.4745 2237.6137,-267.8987"/>
-<polygon fill="#191970" stroke="#191970" points="2237.9949,-264.4195 2227.6498,-266.7064 2237.1631,-271.3699 2237.9949,-264.4195"/>
+<path fill="none" stroke="#191970" d="M2383.7263,-380.4943C2353.1921,-371.6063 2306.7966,-357.6849 2267,-344 2201.458,-321.4621 2126.0716,-291.798 2083.9333,-274.8722"/>
+<polygon fill="#191970" stroke="#191970" points="2085.0456,-271.547 2074.4621,-271.0572 2082.4302,-278.04 2085.0456,-271.547"/>
</g>
<!-- Node6->Node43 -->
-<g id="edge64" class="edge">
+<g id="edge65" class="edge">
<title>Node6->Node43</title>
-<path fill="none" stroke="#191970" d="M2470.4895,-387.1571C2253.9622,-376.9841 1518.4724,-342.429 1290.1688,-331.7027"/>
-<polygon fill="#191970" stroke="#191970" points="1290.1585,-328.1985 1280.0053,-331.2252 1289.83,-335.1908 1290.1585,-328.1985"/>
+<path fill="none" stroke="#191970" d="M2356.1527,-386.3598C2176.0487,-375.5851 1646.4245,-343.9004 1457.1846,-332.5791"/>
+<polygon fill="#191970" stroke="#191970" points="1457.2804,-329.0787 1447.0892,-331.9752 1456.8623,-336.0662 1457.2804,-329.0787"/>
</g>
<!-- Node8 -->
<g id="node9" class="node">
<title>Node8</title>
<g id="a_node9"><a xlink:href="functor_8h.html" target="_top" xlink:title="Defines the Functor data structures. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="2597.5,-185 2597.5,-204 2706.5,-204 2706.5,-185 2597.5,-185"/>
-<text text-anchor="middle" x="2652" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/functor.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="2467.5,-185 2467.5,-204 2576.5,-204 2576.5,-185 2467.5,-185"/>
+<text text-anchor="middle" x="2522" y="-192" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/functor.h</text>
</a>
</g>
</g>
<!-- Node7->Node8 -->
<g id="edge8" class="edge">
<title>Node7->Node8</title>
-<path fill="none" stroke="#191970" d="M2544.645,-313.389C2567.2877,-288.3136 2612.4913,-238.2534 2636.3771,-211.8014"/>
-<polygon fill="#191970" stroke="#191970" points="2639.2234,-213.8718 2643.3276,-204.1042 2634.028,-209.1804 2639.2234,-213.8718"/>
+<path fill="none" stroke="#191970" d="M2655.0847,-313.389C2626.6507,-287.988 2569.5175,-236.9489 2540.227,-210.7828"/>
+<polygon fill="#191970" stroke="#191970" points="2542.5403,-208.1562 2532.7509,-204.1042 2537.8768,-213.3765 2542.5403,-208.1562"/>
</g>
<!-- Node7->Node16 -->
-<g id="edge45" class="edge">
+<g id="edge46" class="edge">
<title>Node7->Node16</title>
-<path fill="none" stroke="#191970" d="M2568.9404,-313.4664C2619.8746,-292.7218 2712.5661,-252.9791 2787,-210 2807.7162,-198.0382 2810.0855,-190.6116 2831,-179 2858.9478,-163.4836 3065.0829,-75.1224 3096,-67 3248.5896,-26.9123 3438.4026,-18.0275 3513.3698,-16.0595"/>
-<polygon fill="#191970" stroke="#191970" points="3513.6653,-19.5535 3523.5791,-15.8164 3513.4986,-12.5555 3513.6653,-19.5535"/>
+<path fill="none" stroke="#191970" d="M2722.5722,-313.4273C2754.8981,-303.57 2797.6191,-290.1289 2835,-277 2988.6367,-223.0398 3036.1367,-226.3919 3176,-143 3221.2599,-116.0143 3219.0684,-88.8918 3267,-67 3317.5318,-43.9206 3481.4683,-25.567 3551.6433,-18.5661"/>
+<polygon fill="#191970" stroke="#191970" points="3552.072,-22.0409 3561.6813,-17.5785 3551.3866,-15.0745 3552.072,-22.0409"/>
</g>
<!-- Node21 -->
<g id="node17" class="node">
<title>Node21</title>
<g id="a_node17"><a xlink:href="object__path_8h.html" target="_top" xlink:title="tvm/node/object_path.h">
-<polygon fill="#ffffff" stroke="#ff0000" points="2398,-252 2398,-271 2530,-271 2530,-252 2398,-252"/>
-<text text-anchor="middle" x="2464" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/object_path.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="2436,-252 2436,-271 2568,-271 2568,-252 2436,-252"/>
+<text text-anchor="middle" x="2502" y="-259" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/node/object_path.h</text>
</a>
</g>
</g>
<!-- Node7->Node21 -->
<g id="edge18" class="edge">
<title>Node7->Node21</title>
-<path fill="none" stroke="#191970" d="M2515.7967,-313.2967C2505.3578,-302.8578 2491.4997,-288.9997 2480.7446,-278.2446"/>
-<polygon fill="#191970" stroke="#191970" points="2483.101,-275.6512 2473.555,-271.055 2478.1512,-280.601 2483.101,-275.6512"/>
+<path fill="none" stroke="#191970" d="M2633.8487,-313.4639C2604.4255,-301.8677 2564.1107,-285.9789 2535.8314,-274.8335"/>
+<polygon fill="#191970" stroke="#191970" points="2536.955,-271.5144 2526.3682,-271.1039 2534.3883,-278.0269 2536.955,-271.5144"/>
</g>
<!-- Node34 -->
<g id="node23" class="node">
<title>Node34</title>
<g id="a_node23"><a xlink:href="array_8h.html" target="_top" xlink:title="Runtime Array container types. ">
-<polygon fill="#ffffff" stroke="#ff0000" points="1478,-246.5 1478,-276.5 1604,-276.5 1604,-246.5 1478,-246.5"/>
-<text text-anchor="start" x="1486" y="-264.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
-<text text-anchor="middle" x="1541" y="-253.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/array.h</text>
+<polygon fill="#ffffff" stroke="#ff0000" points="742,-246.5 742,-276.5 868,-276.5 868,-246.5 742,-246.5"/>
+<text text-anchor="start" x="750" y="-264.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/container</text>
+<text text-anchor="middle" x="805" y="-253.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">/array.h</text>
</a>
</g>
</g>
<!-- Node7->Node34 -->
<g id="edge35" class="edge">
<title>Node7->Node34</title>
-<path fill="none" stroke="#191970" d="M2474.2175,-324.6571C2304.8732,-313.1965 1803.4099,-279.2591 1614.2906,-266.4601"/>
-<polygon fill="#191970" stroke="#191970" points="1614.3867,-262.9587 1604.1732,-265.7754 1613.914,-269.9427 1614.3867,-262.9587"/>
+<path fill="none" stroke="#191970" d="M2615.3458,-324.8813C2559.9525,-321.4409 2473.2069,-316.3131 2398,-313 1891.1281,-290.6706 1764.0887,-293.6942 1257,-277 1123.4413,-272.603 967.694,-267.2018 878.2894,-264.0749"/>
+<polygon fill="#191970" stroke="#191970" points="878.3695,-260.5757 868.2532,-263.7237 878.1246,-267.5714 878.3695,-260.5757"/>
</g>
<!-- Node7->Node35 -->
-<g id="edge40" class="edge">
+<g id="edge41" class="edge">
<title>Node7->Node35</title>
-<path fill="none" stroke="#191970" d="M2587.5771,-325.1793C2643.609,-320.1904 2730.813,-307.7747 2800,-277 2836.4019,-260.8082 2871.9334,-230.3766 2892.083,-211.343"/>
-<polygon fill="#191970" stroke="#191970" points="2894.7294,-213.6533 2899.4976,-204.192 2889.87,-208.6148 2894.7294,-213.6533"/>
+<path fill="none" stroke="#191970" d="M2719.6348,-313.4744C2744.4995,-304.6228 2774.8876,-292.1783 2800,-277 2831.0097,-258.2573 2862.0724,-229.6977 2880.4096,-211.5738"/>
+<polygon fill="#191970" stroke="#191970" points="2883.3203,-213.6108 2887.8934,-204.0537 2878.3585,-208.673 2883.3203,-213.6108"/>
</g>
<!-- Node8->Node10 -->
<g id="edge9" class="edge">
<title>Node8->Node10</title>
-<path fill="none" stroke="#191970" d="M2641.255,-184.9818C2622.8078,-169.2261 2583.2916,-137.9793 2544,-123 2499.9774,-106.2171 2375.6008,-91.5547 2296.0762,-83.5213"/>
-<polygon fill="#191970" stroke="#191970" points="2295.9456,-79.9909 2285.6473,-82.4797 2295.2499,-86.9563 2295.9456,-79.9909"/>
+<path fill="none" stroke="#191970" d="M2511.8899,-184.781C2494.8583,-169.0222 2458.6976,-138.1366 2422,-123 2384.6296,-107.5859 2280.5396,-93.1535 2209.8325,-84.7006"/>
+<polygon fill="#191970" stroke="#191970" points="2209.9962,-81.1957 2199.6545,-83.4974 2209.1744,-88.1473 2209.9962,-81.1957"/>
</g>
<!-- Node8->Node17 -->
<g id="edge15" class="edge">
<title>Node8->Node17</title>
-<path fill="none" stroke="#191970" d="M2649.9532,-184.9875C2646.3046,-170.0665 2637.3748,-141.008 2620,-123 2580.8047,-82.3762 2559.5973,-84.717 2506,-67 2425.3728,-40.348 2326.8541,-26.2737 2270.7733,-19.8989"/>
-<polygon fill="#191970" stroke="#191970" points="2271.1337,-16.4175 2260.8095,-18.7969 2270.3641,-23.3751 2271.1337,-16.4175"/>
+<path fill="none" stroke="#191970" d="M2527.8842,-184.9291C2535.9607,-170.435 2548.1717,-142.5401 2536,-123 2506.1537,-75.0858 2475.8205,-84.0272 2422,-67 2324.1604,-36.0464 2203.7199,-23.218 2139.8061,-18.2893"/>
+<polygon fill="#191970" stroke="#191970" points="2139.9025,-14.7869 2129.671,-17.5378 2139.3848,-21.7678 2139.9025,-14.7869"/>
</g>
<!-- Node8->Node18 -->
<g id="edge16" class="edge">
<title>Node8->Node18</title>
-<path fill="none" stroke="#191970" d="M2644.9065,-184.8412C2632.8362,-169.169 2606.7038,-138.405 2577,-123 2463.1799,-63.9704 2421.4111,-81.3502 2294,-67 2114.4489,-46.7773 777.1274,-20.4866 549.0356,-16.1253"/>
-<polygon fill="#191970" stroke="#191970" points="548.7966,-12.6202 538.7316,-15.9288 548.663,-19.619 548.7966,-12.6202"/>
+<path fill="none" stroke="#191970" d="M2515.6116,-184.924C2504.7049,-169.3709 2480.9411,-138.7743 2453,-123 2355.7331,-68.0873 2318.7987,-81.1381 2208,-67 2029.7742,-44.2581 699.7129,-20.0545 472.8568,-16.0697"/>
+<polygon fill="#191970" stroke="#191970" points="472.6683,-12.566 462.6086,-15.8903 472.5458,-19.5649 472.6683,-12.566"/>
</g>
<!-- Node8->Node20 -->
<g id="edge17" class="edge">
<title>Node8->Node20</title>
-<path fill="none" stroke="#191970" d="M2597.3327,-190.7224C2430.3427,-179.1832 1930.1256,-144.6177 1795.6158,-135.3229"/>
-<polygon fill="#191970" stroke="#191970" points="1795.7237,-131.8221 1785.5062,-134.6243 1795.2411,-138.8054 1795.7237,-131.8221"/>
+<path fill="none" stroke="#191970" d="M2467.4075,-190.5593C2306.1166,-178.9169 1834.8676,-144.9007 1704.0369,-135.4569"/>
+<polygon fill="#191970" stroke="#191970" points="1704.0507,-131.9489 1693.8246,-134.7197 1703.5466,-138.9307 1704.0507,-131.9489"/>
</g>
<!-- Node10->Node11 -->
<g id="edge10" class="edge">
<title>Node10->Node11</title>
-<path fill="none" stroke="#191970" d="M2285.5995,-71.4882C2407.0849,-60.2531 2683.0964,-34.7272 2816.3146,-22.407"/>
-<polygon fill="#191970" stroke="#191970" points="2816.7953,-25.8776 2826.4304,-21.4715 2816.1506,-18.9073 2816.7953,-25.8776"/>
+<path fill="none" stroke="#191970" d="M2199.7085,-71.3245C2318.0395,-60.0766 2582.0719,-34.9793 2711.9376,-22.635"/>
+<polygon fill="#191970" stroke="#191970" points="2712.5637,-26.0913 2722.1876,-21.6607 2711.9013,-19.1228 2712.5637,-26.0913"/>
</g>
<!-- Node15 -->
<g id="node12" class="node">
<title>Node15</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="1490.5,-6 1490.5,-25 1615.5,-25 1615.5,-6 1490.5,-6"/>
-<text text-anchor="middle" x="1553" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/logging.h</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1467.5,-6 1467.5,-25 1592.5,-25 1592.5,-6 1467.5,-6"/>
+<text text-anchor="middle" x="1530" y="-13" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">tvm/runtime/logging.h</text>
</g>
<!-- Node10->Node15 -->
<g id="edge11" class="edge">
<title>Node10->Node15</title>
-<path fill="none" stroke="#191970" d="M2166.2757,-71.5423C2042.7457,-60.2539 1759.5483,-34.3748 1625.6663,-22.1404"/>
-<polygon fill="#191970" stroke="#191970" points="1625.7891,-18.6371 1615.5121,-21.2125 1625.152,-25.6081 1625.7891,-18.6371"/>
+<path fill="none" stroke="#191970" d="M2080.4,-70.9911C1967.8457,-59.6435 1724.4024,-35.0996 1602.5521,-22.8147"/>
+<polygon fill="#191970" stroke="#191970" points="1602.8701,-19.3291 1592.5694,-21.8082 1602.1679,-26.2938 1602.8701,-19.3291"/>
</g>
<!-- Node10->Node16 -->
<g id="edge12" class="edge">
<title>Node10->Node16</title>
-<path fill="none" stroke="#191970" d="M2285.7142,-74.2179C2515.8023,-63.4978 3340.6904,-25.0656 3513.9355,-16.9939"/>
-<polygon fill="#191970" stroke="#191970" points="3514.1226,-20.4891 3523.9488,-16.5274 3513.7968,-13.4967 3514.1226,-20.4891"/>
+<path fill="none" stroke="#191970" d="M2199.7527,-74.4551C2444.6282,-64.0259 3367.175,-24.7346 3551.7436,-16.8738"/>
+<polygon fill="#191970" stroke="#191970" points="3552.0419,-20.3644 3561.8838,-16.4419 3551.7439,-13.3707 3552.0419,-20.3644"/>
</g>
<!-- Node10->Node17 -->
<g id="edge13" class="edge">
<title>Node10->Node17</title>
-<path fill="none" stroke="#191970" d="M2226,-67.3906C2226,-58.8657 2226,-46.1392 2226,-35.4235"/>
-<polygon fill="#191970" stroke="#191970" points="2229.5001,-35.2448 2226,-25.2449 2222.5001,-35.2449 2229.5001,-35.2448"/>
+<path fill="none" stroke="#191970" d="M2132.9688,-67.3906C2126.337,-58.3273 2116.2304,-44.5149 2108.1113,-33.4188"/>
+<polygon fill="#191970" stroke="#191970" points="2110.8601,-31.2484 2102.1304,-25.2449 2105.2109,-35.382 2110.8601,-31.2484"/>
</g>
<!-- Node10->Node18 -->
<g id="edge14" class="edge">
<title>Node10->Node18</title>
-<path fill="none" stroke="#191970" d="M2166.1429,-74.8472C1891.5992,-64.9733 757.3223,-24.1791 548.9636,-16.6855"/>
-<polygon fill="#191970" stroke="#191970" points="548.7982,-13.1774 538.6788,-16.3156 548.5466,-20.1729 548.7982,-13.1774"/>
+<path fill="none" stroke="#191970" d="M2080.4929,-74.8472C1807.5547,-64.9733 679.911,-24.1791 472.7708,-16.6855"/>
+<polygon fill="#191970" stroke="#191970" points="472.6663,-13.1796 462.5462,-16.3156 472.4131,-20.175 472.6663,-13.1796"/>
</g>
<!-- Node21->Node10 -->
<g id="edge33" class="edge">
<title>Node21->Node10</title>
-<path fill="none" stroke="#191970" d="M2461.7975,-251.7352C2455.5665,-226.4085 2435.3789,-158.1404 2392,-123 2375.6847,-109.7833 2327.3224,-97.2255 2286.979,-88.5805"/>
-<polygon fill="#191970" stroke="#191970" points="2287.6917,-85.1539 2277.1858,-86.5237 2286.2529,-92.0044 2287.6917,-85.1539"/>
+<path fill="none" stroke="#191970" d="M2488.177,-251.8383C2453.0622,-227.6135 2357.0322,-163.346 2270,-123 2241.4293,-109.7553 2207.8972,-97.9768 2182.0454,-89.6545"/>
+<polygon fill="#191970" stroke="#191970" points="2182.7762,-86.2146 2172.186,-86.5277 2180.6601,-92.8871 2182.7762,-86.2146"/>
</g>
<!-- Node21->Node16 -->
<g id="edge34" class="edge">
<title>Node21->Node16</title>
-<path fill="none" stroke="#191970" d="M2471.9847,-251.8367C2486.6634,-234.6843 2519.6837,-198.859 2555,-179 2631.3644,-136.059 2659.0659,-144.4861 2744,-123 2861.1241,-93.3706 2890.5727,-85.2503 3010,-67 3199.8985,-37.9806 3430.2961,-22.366 3513.8166,-17.3342"/>
-<polygon fill="#191970" stroke="#191970" points="3514.1621,-20.8199 3523.937,-16.7333 3513.7471,-13.8322 3514.1621,-20.8199"/>
+<path fill="none" stroke="#191970" d="M2524.2237,-251.9103C2547.9742,-241.6361 2586.7099,-224.806 2620,-210 2705.6005,-171.9286 2723.4816,-153.682 2812,-123 2910.1828,-88.9681 2936.5502,-84.3797 3039,-67 3231.9693,-34.2645 3467.1667,-20.8693 3551.7691,-16.8823"/>
+<polygon fill="#191970" stroke="#191970" points="3551.9543,-20.3776 3561.7829,-16.4216 3551.6325,-13.385 3551.9543,-20.3776"/>
</g>
<!-- Node21->Node28 -->
<g id="edge19" class="edge">
<title>Node21->Node28</title>
-<path fill="none" stroke="#191970" d="M2397.6789,-256.5897C2352.1983,-253.3211 2290.4658,-249.078 2236,-246 1832.3176,-223.187 1347.9728,-204.0933 1166.4459,-197.2269"/>
-<polygon fill="#191970" stroke="#191970" points="1166.3443,-193.7206 1156.2193,-196.8409 1166.0802,-200.7156 1166.3443,-193.7206"/>
+<path fill="none" stroke="#191970" d="M2435.7008,-258.447C2195.5289,-247.3876 1371.9688,-209.4642 1120.5058,-197.8848"/>
+<polygon fill="#191970" stroke="#191970" points="1120.5322,-194.3824 1110.3818,-197.4186 1120.2102,-201.375 1120.5322,-194.3824"/>
</g>
<!-- Node28->Node10 -->
<g id="edge25" class="edge">
<title>Node28->Node10</title>
-<path fill="none" stroke="#191970" d="M1099.9496,-179.2529C1108.6837,-162.1136 1125.4377,-135.1034 1149,-123 1193.7792,-99.998 1924.2863,-83.1631 2156.0397,-78.3805"/>
-<polygon fill="#191970" stroke="#191970" points="2156.4416,-81.8731 2166.3676,-78.1685 2156.2979,-74.8745 2156.4416,-81.8731"/>
+<path fill="none" stroke="#191970" d="M1058.5242,-179.4572C1072.7251,-162.2381 1098.6121,-134.8848 1128,-123 1214.677,-87.9468 1855.5938,-79.3885 2070.2055,-77.4951"/>
+<polygon fill="#191970" stroke="#191970" points="2070.4329,-80.9933 2080.4023,-77.4073 2070.3726,-73.9936 2070.4329,-80.9933"/>
</g>
<!-- Node28->Node15 -->
<g id="edge20" class="edge">
<title>Node28->Node15</title>
-<path fill="none" stroke="#191970" d="M1070.4732,-179.4495C1059.0259,-170.397 1046.2434,-157.7722 1040,-143 1026.2101,-110.3725 1035.4915,-88.0234 1064,-67 1096.7355,-42.8593 1351.7622,-26.1191 1479.9503,-19.1404"/>
-<polygon fill="#191970" stroke="#191970" points="1480.2815,-22.6277 1490.0788,-18.595 1479.905,-15.6379 1480.2815,-22.6277"/>
+<path fill="none" stroke="#191970" d="M1041.9753,-179.1586C1034.315,-152.5103 1023.4203,-97.5964 1052,-67 1079.1124,-37.9745 1330.167,-23.6404 1457.1566,-18.1972"/>
+<polygon fill="#191970" stroke="#191970" points="1457.3501,-21.6923 1467.1939,-17.7744 1457.0554,-14.6985 1457.3501,-21.6923"/>
</g>
<!-- Node28->Node16 -->
<g id="edge28" class="edge">
<title>Node28->Node16</title>
-<path fill="none" stroke="#191970" d="M1094.2568,-179.3115C1096.5381,-162.4938 1102.9557,-136.0399 1121,-123 1171.8742,-86.2352 3229.4794,-24.7622 3513.5932,-16.4423"/>
-<polygon fill="#191970" stroke="#191970" points="3513.7719,-19.9386 3523.6653,-16.1479 3513.5673,-12.9416 3513.7719,-19.9386"/>
+<path fill="none" stroke="#191970" d="M1110.1739,-193.7431C1276.3028,-191.2615 1738.9956,-181.1802 2122,-143 2345.6661,-120.7036 2398.3974,-89.924 2622,-67 2985.7934,-29.7034 3429.9258,-18.5482 3551.7757,-16.0871"/>
+<polygon fill="#191970" stroke="#191970" points="3551.9875,-19.5837 3561.9166,-15.8876 3551.8497,-12.5851 3551.9875,-19.5837"/>
</g>
<!-- Node28->Node17 -->
<g id="edge29" class="edge">
<title>Node28->Node17</title>
-<path fill="none" stroke="#191970" d="M1087.3813,-179.4257C1083.9104,-169.2891 1079.7888,-155.538 1078,-143 1072.9971,-107.9331 1072.8824,-87.1713 1102,-67 1124.5651,-51.368 1974.5273,-23.4685 2181.101,-16.9084"/>
-<polygon fill="#191970" stroke="#191970" points="2181.4007,-20.4008 2191.2849,-16.5858 2181.179,-13.4043 2181.4007,-20.4008"/>
+<path fill="none" stroke="#191970" d="M1052.0004,-179.2852C1058.4017,-162.4435 1071.2203,-135.9676 1092,-123 1174.3451,-71.6126 1866.2761,-28.6078 2050.2335,-18.0053"/>
+<polygon fill="#191970" stroke="#191970" points="2050.581,-21.4912 2060.3643,-17.4247 2050.1804,-14.5027 2050.581,-21.4912"/>
</g>
<!-- Node28->Node18 -->
<g id="edge31" class="edge">
<title>Node28->Node18</title>
-<path fill="none" stroke="#191970" d="M1061.1017,-179.4809C1030.1809,-164.9487 981.9324,-142.343 940,-123 885.4575,-97.8402 874.4123,-84.6558 817,-67 722.0259,-37.7929 604.7825,-23.7482 548.7823,-18.3193"/>
-<polygon fill="#191970" stroke="#191970" points="549.024,-14.8267 538.7403,-17.3753 548.3688,-21.796 549.024,-14.8267"/>
+<path fill="none" stroke="#191970" d="M1017.0939,-179.2951C963.3882,-152.61 846.4487,-97.3097 742,-67 646.265,-39.219 528.8389,-24.5122 472.7953,-18.6263"/>
+<polygon fill="#191970" stroke="#191970" points="473.0512,-15.1343 462.7465,-17.5971 472.3379,-22.0979 473.0512,-15.1343"/>
</g>
<!-- Node28->Node20 -->
<g id="edge32" class="edge">
<title>Node28->Node20</title>
-<path fill="none" stroke="#191970" d="M1156.2392,-188.6865C1294.953,-175.9348 1622.0123,-145.8688 1728.1197,-136.1146"/>
-<polygon fill="#191970" stroke="#191970" points="1728.616,-139.5838 1738.2535,-135.183 1727.9751,-132.6132 1728.616,-139.5838"/>
+<path fill="none" stroke="#191970" d="M1110.165,-188.2646C1240.7455,-175.3742 1536.5426,-146.1744 1636.372,-136.3196"/>
+<polygon fill="#191970" stroke="#191970" points="1636.8787,-139.7867 1646.4865,-135.3212 1636.191,-132.8206 1636.8787,-139.7867"/>
</g>
<!-- Node28->Node24 -->
<g id="edge21" class="edge">
<title>Node28->Node24</title>
-<path fill="none" stroke="#191970" d="M1124.8016,-179.4554C1145.8265,-169.509 1173.1859,-156.5659 1193.6843,-146.8686"/>
-<polygon fill="#191970" stroke="#191970" points="1195.3119,-149.9706 1202.8546,-142.5303 1192.3184,-143.6429 1195.3119,-149.9706"/>
+<path fill="none" stroke="#191970" d="M1110.0592,-190.6296C1286.4836,-179.8012 1784.5641,-149.2303 1974.3171,-137.5838"/>
+<polygon fill="#191970" stroke="#191970" points="1974.7069,-141.0666 1984.4737,-136.9604 1974.278,-134.0797 1974.7069,-141.0666"/>
</g>
<!-- Node28->Node26 -->
<g id="edge26" class="edge">
<title>Node28->Node26</title>
-<path fill="none" stroke="#191970" d="M1156.4256,-185.4915C1246.0534,-172.7615 1407.3413,-149.8534 1483.7086,-139.0067"/>
-<polygon fill="#191970" stroke="#191970" points="1484.496,-142.4301 1493.9045,-137.5586 1483.5116,-135.4997 1484.496,-142.4301"/>
+<path fill="none" stroke="#191970" d="M1103.509,-179.4554C1144.4092,-168.5664 1198.8001,-154.0857 1235.9156,-144.2043"/>
+<polygon fill="#191970" stroke="#191970" points="1237.1374,-147.501 1245.9004,-141.546 1235.3365,-140.7366 1237.1374,-147.501"/>
</g>
<!-- Node31 -->
<g id="node21" class="node">
<title>Node31</title>
-<polygon fill="#ffffff" stroke="#bfbfbf" points="840,-123.5 840,-142.5 898,-142.5 898,-123.5 840,-123.5"/>
-<text text-anchor="middle" x="869" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">memory</text>
+<polygon fill="#ffffff" stroke="#bfbfbf" points="1170,-123.5 1170,-142.5 1228,-142.5 1228,-123.5 1170,-123.5"/>
+<text text-anchor="middle" x="1199" y="-130.5" font-family="Helvetica,sans-Serif" font-size="10.00" fill="#000000">memory</text>
</g>
<!-- Node28->Node31 -->
<g id="edge27" class="edge">
<title>Node28->Node31</title>
-<path fill="none" stroke="#191970" d="M1038.2034,-179.4554C997.8323,-168.3714 943.9064,-153.5658 907.8916,-143.6778"/>
-<polygon fill="#191970" stroke="#191970" points="908.8045,-140.299 898.2347,-141.0265 906.9512,-147.0492 908.8045,-140.299"/>
+<path fill="none" stroke="#191970" d="M1084.1834,-179.4554C1109.2113,-169.329 1141.9163,-156.0964 1166.0166,-146.3453"/>
+<polygon fill="#191970" stroke="#191970" points="1167.4882,-149.5256 1175.4454,-142.5303 1164.8627,-143.0366 1167.4882,-149.5256"/>
</g>
<!-- Node28->Node33 -->
<g id="edge30" class="edge">
<title>Node28->Node33</title>
-<path fill="none" stroke="#191970" d="M1029.9803,-188.4911C904.4652,-176.5234 625.0138,-149.8781 504.7904,-138.4149"/>
-<polygon fill="#191970" stroke="#191970" points="504.979,-134.9171 494.692,-137.452 504.3146,-141.8855 504.979,-134.9171"/>
+<path fill="none" stroke="#191970" d="M983.8335,-187.3851C876.6342,-175.3104 660.7381,-150.9925 557.9294,-139.4124"/>
+<polygon fill="#191970" stroke="#191970" points="558.0751,-135.9067 547.7462,-138.2654 557.2915,-142.8628 558.0751,-135.9067"/>
</g>
<!-- Node24->Node10 -->
<g id="edge22" class="edge">
<title>Node24->Node10</title>
-<path fill="none" stroke="#191970" d="M1287.6262,-124.4125C1292.4857,-123.8915 1297.3177,-123.4123 1302,-123 1620.2793,-94.9742 2001.8519,-82.7279 2156.1638,-78.6606"/>
-<polygon fill="#191970" stroke="#191970" points="2156.5317,-82.1523 2166.4372,-78.3931 2156.3495,-75.1547 2156.5317,-82.1523"/>
+<path fill="none" stroke="#191970" d="M2064.851,-123.2455C2078.8132,-114.6534 2099.3728,-102.0014 2115.4596,-92.1018"/>
+<polygon fill="#191970" stroke="#191970" points="2117.6484,-94.8645 2124.3307,-86.6427 2113.9797,-88.9029 2117.6484,-94.8645"/>
</g>
<!-- Node24->Node17 -->
<g id="edge23" class="edge">
<title>Node24->Node17</title>
-<path fill="none" stroke="#191970" d="M1287.6627,-124.7609C1292.5137,-124.1593 1297.3339,-123.5667 1302,-123 1639.5517,-82.0013 2046.8833,-35.7255 2180.8834,-20.5856"/>
-<polygon fill="#191970" stroke="#191970" points="2181.6178,-24.025 2191.1619,-19.4248 2180.8322,-17.0692 2181.6178,-24.025"/>
+<path fill="none" stroke="#191970" d="M2051.7988,-123.4784C2055.6868,-110.5388 2063.1364,-86.7496 2071,-67 2075.3873,-55.9812 2080.9475,-43.9525 2085.6111,-34.2932"/>
+<polygon fill="#191970" stroke="#191970" points="2088.797,-35.7458 2090.0598,-25.2265 2082.5127,-32.6623 2088.797,-35.7458"/>
</g>
<!-- Node24->Node18 -->
<g id="edge24" class="edge">
<title>Node24->Node18</title>
-<path fill="none" stroke="#191970" d="M1158.3656,-126.3188C1102.4567,-119.5473 1019.8483,-107.1503 950,-87 927.7944,-80.594 924.3121,-73.0243 902,-67 774.3243,-32.5273 616.396,-20.6675 548.9912,-16.9905"/>
-<polygon fill="#191970" stroke="#191970" points="548.7646,-13.4741 538.5963,-16.45 548.4011,-20.4646 548.7646,-13.4741"/>
+<path fill="none" stroke="#191970" d="M1984.4333,-128.2849C1713.1744,-108.4758 670.7071,-32.3478 472.6095,-17.8814"/>
+<polygon fill="#191970" stroke="#191970" points="472.7712,-14.3839 462.5428,-17.1462 472.2613,-21.3654 472.7712,-14.3839"/>
</g>
-<!-- Node34->Node18 -->
+<!-- Node34->Node17 -->
<g id="edge38" class="edge">
+<title>Node34->Node17</title>
+<path fill="none" stroke="#191970" d="M810.0074,-246.2724C823.5162,-207.9691 864.7201,-107.7738 938,-67 962.7431,-53.2326 1840.2387,-23.8041 2050.2,-16.9484"/>
+<polygon fill="#191970" stroke="#191970" points="2050.3749,-20.4447 2060.2556,-16.6207 2050.1469,-13.4484 2050.3749,-20.4447"/>
+</g>
+<!-- Node34->Node18 -->
+<g id="edge39" class="edge">
<title>Node34->Node18</title>
-<path fill="none" stroke="#191970" d="M1477.935,-254.842C1451.4052,-252.0665 1420.2062,-248.8356 1392,-246 1227.1675,-229.4294 1175.1934,-270.5701 1021,-210 1019.1462,-209.2718 908.7667,-123.9197 907,-123 834.1883,-85.0971 810.0778,-89.0199 731,-67 666.9168,-49.1555 591.0612,-31.8684 548.7444,-22.5627"/>
-<polygon fill="#191970" stroke="#191970" points="549.2531,-19.0913 538.7361,-20.3734 547.7571,-25.9296 549.2531,-19.0913"/>
+<path fill="none" stroke="#191970" d="M788.2885,-246.2185C738.7426,-200.9334 595.7304,-70.3811 590,-67 552.5026,-44.8757 504.0992,-30.4342 472.5686,-22.6255"/>
+<polygon fill="#191970" stroke="#191970" points="473.1027,-19.1546 462.5626,-20.2301 471.4729,-25.9622 473.1027,-19.1546"/>
</g>
<!-- Node34->Node20 -->
-<g id="edge39" class="edge">
+<g id="edge40" class="edge">
<title>Node34->Node20</title>
-<path fill="none" stroke="#191970" d="M1552.964,-246.3166C1568.1455,-227.9824 1595.9893,-197.3302 1626,-179 1658.1112,-159.3868 1699.6431,-146.9092 1728.3188,-139.9719"/>
-<polygon fill="#191970" stroke="#191970" points="1729.3113,-143.3346 1738.2607,-137.6637 1727.7282,-136.516 1729.3113,-143.3346"/>
+<path fill="none" stroke="#191970" d="M868.0502,-254.814C950.6588,-245.8232 1099.9557,-228.7809 1227,-210 1380.0958,-187.3679 1561.7472,-153.6108 1635.9123,-139.5323"/>
+<polygon fill="#191970" stroke="#191970" points="1636.8892,-142.9093 1646.0586,-137.6016 1635.5806,-136.0327 1636.8892,-142.9093"/>
</g>
<!-- Node34->Node26 -->
<g id="edge36" class="edge">
<title>Node34->Node26</title>
-<path fill="none" stroke="#191970" d="M1539.2204,-246.2548C1536.5029,-222.9744 1531.3184,-178.5607 1528.3127,-152.8117"/>
-<polygon fill="#191970" stroke="#191970" points="1531.7724,-152.2625 1527.1365,-142.7358 1524.8196,-153.0742 1531.7724,-152.2625"/>
+<path fill="none" stroke="#191970" d="M868.1273,-257.4889C962.16,-250.7712 1133.9622,-235.4228 1190,-210 1220.1859,-196.3055 1247.665,-168.7557 1263.6102,-150.596"/>
+<polygon fill="#191970" stroke="#191970" points="1266.654,-152.4202 1270.4727,-142.5374 1261.3245,-147.8817 1266.654,-152.4202"/>
</g>
<!-- Node34->Node31 -->
<g id="edge37" class="edge">
<title>Node34->Node31</title>
-<path fill="none" stroke="#191970" d="M1477.9458,-254.7323C1451.4177,-251.9392 1420.2171,-248.7247 1392,-246 1212.5676,-228.6735 1157.256,-272.037 988,-210 965.5395,-201.7676 963.9838,-192.149 944,-179 927.6708,-168.2556 908.9259,-156.7586 894.2608,-147.9462"/>
-<polygon fill="#191970" stroke="#191970" points="895.6586,-144.7041 885.2793,-142.5778 892.0671,-150.7126 895.6586,-144.7041"/>
+<path fill="none" stroke="#191970" d="M830.9241,-246.3781C863.5334,-227.9619 921.7227,-197.081 975,-179 1037.9542,-157.635 1114.3936,-144.5383 1159.7394,-138.0345"/>
+<polygon fill="#191970" stroke="#191970" points="1160.3304,-141.486 1169.7489,-136.6339 1159.3603,-134.5536 1160.3304,-141.486"/>
</g>
<!-- Node35->Node11 -->
-<g id="edge41" class="edge">
+<g id="edge42" class="edge">
<title>Node35->Node11</title>
-<path fill="none" stroke="#191970" d="M2908.0322,-184.8762C2905.3232,-157.9361 2897.5977,-81.1108 2893.5478,-40.8362"/>
-<polygon fill="#191970" stroke="#191970" points="2897.0163,-40.3465 2892.5332,-30.7469 2890.0514,-41.047 2897.0163,-40.3465"/>
+<path fill="none" stroke="#191970" d="M2891.0859,-184.8762C2874.3188,-157.5914 2826.1065,-79.1369 2801.631,-39.3086"/>
+<polygon fill="#191970" stroke="#191970" points="2804.5873,-37.4343 2796.3696,-30.7469 2798.6234,-41.0993 2804.5873,-37.4343"/>
</g>
<!-- Node35->Node15 -->
-<g id="edge42" class="edge">
+<g id="edge43" class="edge">
<title>Node35->Node15</title>
-<path fill="none" stroke="#191970" d="M2885.9054,-184.9093C2825.0459,-160.1425 2654.7219,-94.0936 2506,-67 2337.2478,-36.2573 1818.0919,-21.5555 1625.7077,-17.0569"/>
-<polygon fill="#191970" stroke="#191970" points="1625.777,-13.5577 1615.6986,-16.8252 1625.6149,-20.5558 1625.777,-13.5577"/>
+<path fill="none" stroke="#191970" d="M2865.5331,-184.9433C2818.7083,-170.8242 2727.8442,-143.8174 2650,-123 2549.1971,-96.0429 2525.185,-82.5167 2422,-67 2264.5831,-43.3281 1786.218,-24.5102 1602.7219,-17.9789"/>
+<polygon fill="#191970" stroke="#191970" points="1602.7526,-14.4779 1592.635,-17.6217 1602.5048,-21.4735 1602.7526,-14.4779"/>
</g>
<!-- Node35->Node16 -->
-<g id="edge43" class="edge">
+<g id="edge44" class="edge">
<title>Node35->Node16</title>
-<path fill="none" stroke="#191970" d="M2922.8735,-184.8376C2958.6147,-160.4791 3057.5291,-96.414 3150,-67 3217.8914,-45.4045 3432.0046,-25.2907 3513.8018,-18.1984"/>
-<polygon fill="#191970" stroke="#191970" points="3514.3303,-21.666 3523.994,-17.3229 3513.7311,-14.6916 3514.3303,-21.666"/>
+<path fill="none" stroke="#191970" d="M2914.2674,-184.9542C2958.6733,-160.8693 3081.073,-97.3933 3191,-67 3321.6339,-30.8815 3483.8904,-19.8555 3551.78,-16.6908"/>
+<polygon fill="#191970" stroke="#191970" points="3552.01,-20.1842 3561.8464,-16.2478 3551.7021,-13.191 3552.01,-20.1842"/>
</g>
<!-- Node35->Node17 -->
-<g id="edge44" class="edge">
+<g id="edge45" class="edge">
<title>Node35->Node17</title>
-<path fill="none" stroke="#191970" d="M2894.7728,-184.9745C2857.6629,-160.6544 2754.1213,-96.1374 2658,-67 2586.3257,-45.2733 2366.4499,-26.3028 2271,-18.8567"/>
-<polygon fill="#191970" stroke="#191970" points="2271.0446,-15.3498 2260.8045,-18.0683 2270.5049,-22.3289 2271.0446,-15.3498"/>
+<path fill="none" stroke="#191970" d="M2880.6712,-184.9157C2838.1194,-160.4557 2719.6478,-95.6337 2612,-67 2523.3156,-43.4105 2248.9974,-24.7852 2140.1625,-18.1353"/>
+<polygon fill="#191970" stroke="#191970" points="2140.0468,-14.6219 2129.8535,-17.511 2139.6237,-21.6091 2140.0468,-14.6219"/>
</g>
<!-- Node36->Node8 -->
-<g id="edge47" class="edge">
+<g id="edge48" class="edge">
<title>Node36->Node8</title>
-<path fill="none" stroke="#191970" d="M2752.9396,-313.4983C2748.7646,-295.8952 2739.9173,-266.5112 2724,-246 2712.218,-230.8175 2694.7671,-218.1794 2679.9966,-209.2226"/>
-<polygon fill="#191970" stroke="#191970" points="2681.6271,-206.1222 2671.221,-204.1316 2678.1145,-212.1771 2681.6271,-206.1222"/>
+<path fill="none" stroke="#191970" d="M2347.2659,-313.3656C2365.3485,-295.8989 2396.835,-266.9005 2427,-246 2447.7114,-231.6496 2472.8394,-218.1815 2492.0489,-208.6198"/>
+<polygon fill="#191970" stroke="#191970" points="2493.7882,-211.6653 2501.2286,-204.1226 2490.7085,-205.3791 2493.7882,-211.6653"/>
</g>
<!-- Node36->Node16 -->
-<g id="edge58" class="edge">
+<g id="edge59" class="edge">
<title>Node36->Node16</title>
-<path fill="none" stroke="#191970" d="M2802.6556,-313.4467C2830.9024,-303.9034 2867.4463,-290.7899 2899,-277 3084.791,-195.8038 3113.1214,-135.3848 3304,-67 3376.7456,-40.9379 3466.7536,-26.0732 3513.8408,-19.5384"/>
-<polygon fill="#191970" stroke="#191970" points="3514.3208,-23.0054 3523.7619,-18.1974 3513.3831,-16.0685 3514.3208,-23.0054"/>
+<path fill="none" stroke="#191970" d="M2388.8186,-322.9012C2463.5692,-315.061 2598.8378,-299.2349 2713,-277 2720.3734,-275.5639 2968.0213,-212.7798 2975,-210 3035.7173,-185.8149 3170.0964,-87.963 3232,-67 3345.3638,-28.6107 3488.8045,-18.8308 3551.8254,-16.3446"/>
+<polygon fill="#191970" stroke="#191970" points="3552.0659,-19.8383 3561.934,-15.9822 3551.815,-12.8428 3552.0659,-19.8383"/>
</g>
<!-- Node36->Node35 -->
-<g id="edge48" class="edge">
+<g id="edge49" class="edge">
<title>Node36->Node35</title>
-<path fill="none" stroke="#191970" d="M2812.8057,-314.3383C2834.8749,-306.4318 2858.9036,-294.5273 2876,-277 2892.9385,-259.6346 2901.5192,-232.4736 2905.6186,-214.1038"/>
-<polygon fill="#191970" stroke="#191970" points="2909.0692,-214.6966 2907.5854,-204.2062 2902.2034,-213.3323 2909.0692,-214.6966"/>
+<path fill="none" stroke="#191970" d="M2388.7692,-322.0015C2452.4442,-314.0775 2558.5251,-298.967 2648,-277 2726.4507,-257.7396 2815.7322,-225.561 2863.17,-207.6135"/>
+<polygon fill="#191970" stroke="#191970" points="2864.5378,-210.8379 2872.6386,-204.0095 2862.0477,-204.2958 2864.5378,-210.8379"/>
</g>
<!-- Node36->Node37 -->
-<g id="edge49" class="edge">
+<g id="edge50" class="edge">
<title>Node36->Node37</title>
-<path fill="none" stroke="#191970" d="M2699.3324,-322.0758C2591.5324,-309.8548 2357.074,-283.2749 2237.8159,-269.7549"/>
-<polygon fill="#191970" stroke="#191970" points="2238.0185,-266.2556 2227.6878,-268.6067 2237.2299,-273.211 2238.0185,-266.2556"/>
+<path fill="none" stroke="#191970" d="M2275.4735,-315.0221C2224.3449,-302.8314 2149.7821,-285.053 2100.8384,-273.3832"/>
+<polygon fill="#191970" stroke="#191970" points="2101.6427,-269.9769 2091.1036,-271.0621 2100.0191,-276.786 2101.6427,-269.9769"/>
</g>
<!-- Node36->Node42 -->
-<g id="edge57" class="edge">
+<g id="edge58" class="edge">
<title>Node36->Node42</title>
-<path fill="none" stroke="#191970" d="M2699.3697,-322.6204C2669.0261,-319.5608 2630.9931,-315.8705 2597,-313 2373.2257,-294.104 2315.3932,-308.1971 2093,-277 1964.5025,-258.9745 1815.0906,-223.4027 1743.5018,-205.4346"/>
-<polygon fill="#191970" stroke="#191970" points="1744.0683,-201.9679 1733.516,-202.9155 1742.3561,-208.7553 1744.0683,-201.9679"/>
+<path fill="none" stroke="#191970" d="M2300.7816,-313.4486C2260.4496,-294.6601 2187.6143,-262.9392 2122,-246 1997.9953,-213.9866 1846.458,-201.2609 1773.8254,-196.7283"/>
+<polygon fill="#191970" stroke="#191970" points="1773.8868,-193.2258 1763.6943,-196.1177 1773.4656,-200.2131 1773.8868,-193.2258"/>
</g>
<!-- Node37->Node10 -->
-<g id="edge53" class="edge">
+<g id="edge54" class="edge">
<title>Node37->Node10</title>
-<path fill="none" stroke="#191970" d="M2168.1937,-251.8402C2177.8289,-222.6978 2206.737,-135.2626 2219.6494,-96.208"/>
-<polygon fill="#191970" stroke="#191970" points="2223.0218,-97.1574 2222.8379,-86.5641 2216.3756,-94.9599 2223.0218,-97.1574"/>
+<path fill="none" stroke="#191970" d="M2058.0483,-251.8496C2072.0635,-232.2572 2103.6579,-186.0198 2122,-143 2128.4049,-127.9778 2133.1223,-110.0526 2136.1172,-96.6766"/>
+<polygon fill="#191970" stroke="#191970" points="2139.5877,-97.1814 2138.227,-86.6743 2132.7384,-95.7366 2139.5877,-97.1814"/>
</g>
<!-- Node37->Node11 -->
-<g id="edge50" class="edge">
+<g id="edge51" class="edge">
<title>Node37->Node11</title>
-<path fill="none" stroke="#191970" d="M2183.6078,-251.9331C2231.4099,-227.6531 2362.8535,-162.7549 2478,-123 2539.5429,-101.752 2556.967,-103.3049 2620,-87 2689.9587,-68.9036 2769.948,-47.7415 2824.7877,-33.1604"/>
-<polygon fill="#191970" stroke="#191970" points="2825.9625,-36.4697 2834.7267,-30.5166 2824.163,-29.7049 2825.9625,-36.4697"/>
+<path fill="none" stroke="#191970" d="M2079.8521,-251.8565C2190.1387,-214.9944 2586.1641,-82.6272 2732.313,-33.7785"/>
+<polygon fill="#191970" stroke="#191970" points="2733.6102,-37.0354 2741.9849,-30.5458 2731.3911,-30.3964 2733.6102,-37.0354"/>
</g>
<!-- Node37->Node18 -->
-<g id="edge55" class="edge">
+<g id="edge56" class="edge">
<title>Node37->Node18</title>
-<path fill="none" stroke="#191970" d="M2162.9134,-251.7587C2158.5973,-234.1499 2146.9326,-197.1678 2122,-179 2002.478,-91.9074 1940.7053,-141.6593 1794,-123 1300.5357,-60.2368 695.0423,-25.1484 548.8718,-17.2316"/>
-<polygon fill="#191970" stroke="#191970" points="548.9435,-13.7305 538.7699,-16.6883 548.5675,-20.7204 548.9435,-13.7305"/>
+<path fill="none" stroke="#191970" d="M1988.1989,-255.5948C1952.1712,-252.3755 1906.074,-248.541 1865,-246 1667.4382,-233.7782 1166.134,-261.4571 975,-210 973.9726,-209.7234 705.0034,-67.3539 704,-67 623.6228,-38.6491 523.6628,-24.5462 472.9002,-18.786"/>
+<polygon fill="#191970" stroke="#191970" points="473.1617,-15.2937 462.8394,-17.6813 472.3977,-22.2519 473.1617,-15.2937"/>
</g>
<!-- Node37->Node20 -->
-<g id="edge56" class="edge">
+<g id="edge57" class="edge">
<title>Node37->Node20</title>
-<path fill="none" stroke="#191970" d="M2159.2248,-251.8665C2147.9953,-234.092 2121.3656,-196.3714 2088,-179 2037.5039,-152.7097 1868.2014,-139.4814 1795.5631,-134.9058"/>
-<polygon fill="#191970" stroke="#191970" points="1795.7686,-131.4119 1785.573,-134.293 1795.3399,-138.3988 1795.7686,-131.4119"/>
+<path fill="none" stroke="#191970" d="M2045.2084,-251.8978C2033.9517,-234.1753 2007.2783,-196.538 1974,-179 1927.607,-154.5504 1773.3421,-140.5336 1704.126,-135.338"/>
+<polygon fill="#191970" stroke="#191970" points="1704.0061,-131.82 1693.777,-134.5797 1703.4944,-138.8012 1704.0061,-131.82"/>
</g>
<!-- Node37->Node28 -->
-<g id="edge51" class="edge">
+<g id="edge52" class="edge">
<title>Node37->Node28</title>
-<path fill="none" stroke="#191970" d="M2102.338,-257.5836C1915.9214,-245.9326 1365.4611,-211.5288 1166.1207,-199.07"/>
-<polygon fill="#191970" stroke="#191970" points="1166.244,-195.571 1156.0451,-198.4403 1165.8073,-202.5574 1166.244,-195.571"/>
+<path fill="none" stroke="#191970" d="M1988.1752,-255.9459C1952.1399,-252.84 1906.0429,-249.0018 1865,-246 1590.5204,-225.925 1263.8761,-206.7879 1120.5594,-198.6301"/>
+<polygon fill="#191970" stroke="#191970" points="1120.3305,-195.1115 1110.148,-198.0384 1119.9333,-202.1003 1120.3305,-195.1115"/>
</g>
<!-- Node37->Node35 -->
-<g id="edge52" class="edge">
+<g id="edge53" class="edge">
<title>Node37->Node35</title>
-<path fill="none" stroke="#191970" d="M2227.7918,-255.8454C2362.978,-243.6713 2681.3381,-215.0018 2829.9735,-201.6166"/>
-<polygon fill="#191970" stroke="#191970" points="2830.3003,-205.1015 2839.946,-200.7186 2829.6724,-198.1297 2830.3003,-205.1015"/>
+<path fill="none" stroke="#191970" d="M2113.552,-256.5461C2264.819,-244.5663 2650.2516,-214.0415 2817.5272,-200.7939"/>
+<polygon fill="#191970" stroke="#191970" points="2818.0506,-204.2635 2827.743,-199.9849 2817.4979,-197.2854 2818.0506,-204.2635"/>
</g>
<!-- Node37->Node42 -->
-<g id="edge54" class="edge">
+<g id="edge55" class="edge">
<title>Node37->Node42</title>
-<path fill="none" stroke="#191970" d="M2102.4229,-252.4641C2006.5854,-238.6255 1825.9362,-212.5404 1743.705,-200.6665"/>
-<polygon fill="#191970" stroke="#191970" points="1744.106,-197.1881 1733.7084,-199.223 1743.1055,-204.1163 1744.106,-197.1881"/>
+<path fill="none" stroke="#191970" d="M2005.4916,-251.9717C1943.6711,-239.028 1833.9628,-216.0578 1773.6566,-203.4312"/>
+<polygon fill="#191970" stroke="#191970" points="1774.2102,-199.9713 1763.7051,-201.3476 1772.7756,-206.8227 1774.2102,-199.9713"/>
</g>
<!-- Node43->Node10 -->
-<g id="edge84" class="edge">
+<g id="edge85" class="edge">
<title>Node43->Node10</title>
-<path fill="none" stroke="#191970" d="M1234.8917,-313.4724C1270.1134,-273.6495 1372.8539,-165.5502 1485,-123 1546.6694,-99.6016 1983.4657,-84.2118 2156.0708,-78.9829"/>
-<polygon fill="#191970" stroke="#191970" points="2156.4224,-82.474 2166.3128,-78.6752 2156.2121,-75.4772 2156.4224,-82.474"/>
+<path fill="none" stroke="#191970" d="M1330.955,-316.1994C1286.485,-303.7758 1236.0152,-281.3124 1257,-246 1315.4209,-147.6917 1373.2289,-155.06 1483,-123 1591.5238,-91.3043 1924.4282,-81.1279 2070.2976,-78.1379"/>
+<polygon fill="#191970" stroke="#191970" points="2070.5432,-81.6338 2080.4713,-77.9348 2070.4034,-74.6352 2070.5432,-81.6338"/>
</g>
<!-- Node43->Node11 -->
-<g id="edge65" class="edge">
+<g id="edge66" class="edge">
<title>Node43->Node11</title>
-<path fill="none" stroke="#191970" d="M1280.0438,-327.314C1530.7484,-322.0599 2510.8052,-300.1186 2572,-277 2592.3894,-269.2971 2590.9337,-256.5612 2610,-246 2653.523,-221.8918 2674.4581,-237.3812 2716,-210 2787.5768,-162.8221 2848.7284,-79.497 2875.9253,-38.9616"/>
-<polygon fill="#191970" stroke="#191970" points="2878.9032,-40.8037 2881.5031,-30.5328 2873.0656,-36.9406 2878.9032,-40.8037"/>
+<path fill="none" stroke="#191970" d="M1447.0912,-325.8676C1571.2329,-319.9239 1871.4828,-303.851 2122,-277 2173.7936,-271.4486 2539.2365,-232.9471 2586,-210 2591.4941,-207.304 2713.0442,-88.1769 2764.4094,-37.7144"/>
+<polygon fill="#191970" stroke="#191970" points="2767.0604,-40.0164 2771.7398,-30.5109 2762.154,-35.0236 2767.0604,-40.0164"/>
</g>
<!-- Node43->Node15 -->
-<g id="edge72" class="edge">
+<g id="edge73" class="edge">
<title>Node43->Node15</title>
-<path fill="none" stroke="#191970" d="M1163.7398,-322.8054C1082.1429,-314.2628 938.3853,-296.9834 890,-277 884.156,-274.5864 800.8673,-215.6353 798,-210 780.4651,-175.5379 773.9371,-153.267 798,-123 840.2308,-69.8808 1299.0585,-32.8957 1480.1262,-20.2718"/>
-<polygon fill="#191970" stroke="#191970" points="1480.6176,-23.7463 1490.3521,-19.5642 1480.1343,-16.763 1480.6176,-23.7463"/>
+<path fill="none" stroke="#191970" d="M1330.8084,-320.0517C1274.0307,-311.2717 1185.4337,-296.1403 1110,-277 1086.6702,-271.0804 918.1446,-229.4744 904,-210 895.9033,-198.8524 901.977,-192.6284 904,-179 911.6383,-127.5429 897.5965,-99.7678 938,-67 977.5157,-34.9523 1307.6478,-21.715 1457.053,-17.3279"/>
+<polygon fill="#191970" stroke="#191970" points="1457.5963,-20.8138 1467.4914,-17.0272 1457.3947,-13.8167 1457.5963,-20.8138"/>
</g>
<!-- Node43->Node16 -->
-<g id="edge88" class="edge">
+<g id="edge89" class="edge">
<title>Node43->Node16</title>
-<path fill="none" stroke="#191970" d="M1280.164,-327.8567C1527.952,-324.9087 2485.328,-311.3303 2539,-277 2580.1914,-250.6527 2550.6488,-209.3331 2589,-179 2605.9093,-165.6259 2945.12,-92.3684 2966,-87 2998.9958,-78.5165 3006.4981,-73.1897 3040,-67 3217.004,-34.2973 3432.9537,-21.0214 3513.5944,-16.9695"/>
-<polygon fill="#191970" stroke="#191970" points="3513.8052,-20.4635 3523.6221,-16.4786 3513.4629,-13.4719 3513.8052,-20.4635"/>
+<path fill="none" stroke="#191970" d="M1447.0459,-328.3449C1671.1141,-327.4256 2472.4229,-321.0593 2577,-277 2595.5444,-269.1871 2592.779,-256.4104 2610,-246 2818.0923,-120.2046 2889.8481,-120.7206 3127,-67 3283.6031,-31.5257 3476.1902,-20.0047 3551.5782,-16.695"/>
+<polygon fill="#191970" stroke="#191970" points="3551.988,-20.1809 3561.8324,-16.2644 3551.6943,-13.1871 3551.988,-20.1809"/>
</g>
<!-- Node43->Node17 -->
-<g id="edge90" class="edge">
+<g id="edge91" class="edge">
<title>Node43->Node17</title>
-<path fill="none" stroke="#191970" d="M1163.7864,-321.0565C1096.3937,-311.7911 990.0354,-294.9805 954,-277 875.6193,-237.8905 776.3054,-191.4221 831,-123 839.8194,-111.9671 1065.001,-68.8797 1079,-67 1297.87,-37.6114 1997.1168,-20.5157 2181.2251,-16.4491"/>
-<polygon fill="#191970" stroke="#191970" points="2181.4367,-19.9454 2191.3576,-16.227 2181.2832,-12.9471 2181.4367,-19.9454"/>
+<path fill="none" stroke="#191970" d="M1333.5926,-313.4033C1296.5753,-303.2572 1246.8199,-289.5041 1203,-277 1177.609,-269.7546 990.8214,-231.1396 975,-210 966.7445,-198.9694 968.1929,-190.9788 975,-179 1039.1859,-66.0487 1117.2851,-90.8059 1245,-67 1401.6558,-37.7996 1897.2282,-21.2584 2049.8294,-16.7625"/>
+<polygon fill="#191970" stroke="#191970" points="2050.321,-20.2498 2060.2146,-16.4596 2050.1168,-13.2528 2050.321,-20.2498"/>
</g>
<!-- Node43->Node18 -->
-<g id="edge91" class="edge">
+<g id="edge92" class="edge">
<title>Node43->Node18</title>
-<path fill="none" stroke="#191970" d="M1163.8274,-325.5158C1071.569,-319.9739 896.5529,-305.9287 842,-277 736.4035,-221.0035 771.3122,-136.4014 674,-67 635.9543,-39.8665 582.803,-26.4518 548.8906,-20.258"/>
-<polygon fill="#191970" stroke="#191970" points="549.0475,-16.7338 538.5997,-18.4924 547.8638,-23.633 549.0475,-16.7338"/>
+<path fill="none" stroke="#191970" d="M1330.9071,-322.1953C1228.4563,-310.8943 1023.4694,-287.3999 992,-277 810.2588,-216.9386 792.521,-141.0106 616,-67 567.4983,-46.6645 508.3061,-31.1291 472.4161,-22.6666"/>
+<polygon fill="#191970" stroke="#191970" points="473.0336,-19.217 462.5017,-20.3694 471.4536,-26.0363 473.0336,-19.217"/>
</g>
<!-- Node43->Node20 -->
-<g id="edge92" class="edge">
+<g id="edge93" class="edge">
<title>Node43->Node20</title>
-<path fill="none" stroke="#191970" d="M1280.0603,-313.8517C1321.3191,-302.893 1372.6951,-287.9692 1392,-277 1409.7796,-266.8975 1408.7856,-257.038 1426,-246 1442.0522,-235.7071 1565.9416,-185.1246 1584,-179 1633.3826,-162.2517 1692.2294,-148.1458 1728.3488,-140.1307"/>
-<polygon fill="#191970" stroke="#191970" points="1729.3311,-143.4985 1738.3498,-137.9385 1727.8323,-136.6609 1729.3311,-143.4985"/>
+<path fill="none" stroke="#191970" d="M1411.0216,-313.3298C1425.4225,-303.2677 1444.5452,-289.6371 1461,-277 1515.3393,-235.268 1522.7164,-216.5886 1580,-179 1598.6919,-166.7347 1621.0703,-155.2842 1638.8362,-146.8755"/>
+<polygon fill="#191970" stroke="#191970" points="1640.587,-149.9213 1648.1788,-142.5311 1637.6354,-143.574 1640.587,-149.9213"/>
... 496627 lines suppressed ...