Parse onnx model throwing unhandled exception

huge909 · July 22, 2020, 9:23am

Description

I’am using TensorRT to load my ONNX model in C++ with VS2017, but throw Unhandled Exception:Microsoft C++ exception: std::out_of_range, in the line:

auto parsed = parser->parseFromFile(
		locateFile(mParams.onnxFileName, mParams.dataDirs).c_str(), static_cast<int>(gLogger.getReportableSeverity()));

The code is from tensorrt/samples/sampleONNXMNIST. I have tested with the mnist.onnx, it works fine.

My ONNX model is converted from pytorch1.5.1 in linux platform, with the following code:

torch.onnx.export(model, input_tensor,'model.onnx', input_names=input_name, output_names=output_name, verbose=True,opset_version=12)

Can you give me some help? Thanks!

Environment

TensorRT Version: TensorRT-7.0.0.11.Windows10.x86_64.cuda-10.2.cudnn7.6
GPU Type: GTX1050TI
Nvidia Driver Version: 441.22
CUDA Version: 10.2
CUDNN Version: 7.6
Operating System + Version: Windows 10.x86_64
PyTorch Version (if applicable): 1.5.1

Relevant Files

model.onnx

Others

Output of trtexec.exe --explicitBatch --onnx=model.onnx --verbose:

&&&& RUNNING TensorRT.trtexec # trtexec.exe --explicitBatch --onnx=model.onnx --verbose
[07/22/2020-16:29:00] [I] === Model Options ===
[07/22/2020-16:29:00] [I] Format: ONNX
[07/22/2020-16:29:00] [I] Model: model.onnx
[07/22/2020-16:29:00] [I] Output:
[07/22/2020-16:29:00] [I] === Build Options ===
[07/22/2020-16:29:00] [I] Max batch: explicit
[07/22/2020-16:29:00] [I] Workspace: 16 MB
[07/22/2020-16:29:00] [I] minTiming: 1
[07/22/2020-16:29:00] [I] avgTiming: 8
[07/22/2020-16:29:00] [I] Precision: FP32
[07/22/2020-16:29:00] [I] Calibration:
[07/22/2020-16:29:00] [I] Safe mode: Disabled
[07/22/2020-16:29:00] [I] Save engine:
[07/22/2020-16:29:00] [I] Load engine:
[07/22/2020-16:29:00] [I] Inputs format: fp32:CHW
[07/22/2020-16:29:00] [I] Outputs format: fp32:CHW
[07/22/2020-16:29:00] [I] Input build shapes: model
[07/22/2020-16:29:00] [I] === System Options ===
[07/22/2020-16:29:00] [I] Device: 0
[07/22/2020-16:29:00] [I] DLACore:
[07/22/2020-16:29:00] [I] Plugins:
[07/22/2020-16:29:00] [I] === Inference Options ===
[07/22/2020-16:29:00] [I] Batch: Explicit
[07/22/2020-16:29:00] [I] Iterations: 10
[07/22/2020-16:29:00] [I] Duration: 3s (+ 200ms warm up)
[07/22/2020-16:29:00] [I] Sleep time: 0ms
[07/22/2020-16:29:00] [I] Streams: 1
[07/22/2020-16:29:00] [I] ExposeDMA: Disabled
[07/22/2020-16:29:00] [I] Spin-wait: Disabled
[07/22/2020-16:29:00] [I] Multithreading: Disabled
[07/22/2020-16:29:00] [I] CUDA Graph: Disabled
[07/22/2020-16:29:00] [I] Skip inference: Disabled
[07/22/2020-16:29:00] [I] Inputs:
[07/22/2020-16:29:00] [I] === Reporting Options ===
[07/22/2020-16:29:00] [I] Verbose: Enabled
[07/22/2020-16:29:00] [I] Averages: 10 inferences
[07/22/2020-16:29:00] [I] Percentile: 99
[07/22/2020-16:29:00] [I] Dump output: Disabled
[07/22/2020-16:29:00] [I] Profile: Disabled
[07/22/2020-16:29:00] [I] Export timing to JSON file:
[07/22/2020-16:29:00] [I] Export output to JSON file:
[07/22/2020-16:29:00] [I] Export profile to JSON file:
[07/22/2020-16:29:00] [I]
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::GridAnchor_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::NMS_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::Reorg_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::Region_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::Clip_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::LReLU_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::PriorBox_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::Normalize_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::RPROI_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::BatchedNMS_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::FlattenConcat_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::CropAndResize
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::DetectionLayer_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::Proposal
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::ProposalLayer_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::PyramidROIAlign_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::ResizeNearest_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::Split
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::SpecialSlice_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ::InstanceNormalization_TRT
----------------------------------------------------------------
Input filename:   model.onnx
ONNX IR version:  0.0.6
Opset version:    12
Producer name:    pytorch
Producer version: 1.5
Domain:
Model version:    0
Doc string:
----------------------------------------------------------------
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::GridAnchor_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::NMS_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::Reorg_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::Region_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::Clip_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::LReLU_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::PriorBox_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::Normalize_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::RPROI_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::BatchedNMS_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::FlattenConcat_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::CropAndResize
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::DetectionLayer_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::Proposal
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::ProposalLayer_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::PyramidROIAlign_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::ResizeNearest_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::Split
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::SpecialSlice_TRT
[07/22/2020-16:29:01] [V] [TRT] Plugin creator registration succeeded - ONNXTRT_NAMESPACE::InstanceNormalization_TRT
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:203: Adding network input: skeleton with dtype: float32, dimensions: (1, 3, 300, 18, 2)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: skeleton for ONNX tensor: skeleton
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: A
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: data_bn.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: data_bn.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: data_bn.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: data_bn.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: data_bn.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.0
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.1
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.2
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.3
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.4
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.5
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.6
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.7
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.8
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: edge_importance.9
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: fcn.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: fcn.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.gcn.conv.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.gcn.conv.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.0.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.0.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.0.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.0.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.2.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.2.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.3.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.3.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.3.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.3.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.0.tcn.3.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.gcn.conv.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.gcn.conv.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.0.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.0.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.0.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.0.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.2.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.2.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.3.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.3.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.3.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.3.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.1.tcn.3.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.gcn.conv.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.gcn.conv.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.0.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.0.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.0.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.0.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.2.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.2.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.3.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.3.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.3.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.3.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.2.tcn.3.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.gcn.conv.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.gcn.conv.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.0.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.0.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.0.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.0.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.2.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.2.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.3.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.3.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.3.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.3.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.3.tcn.3.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.gcn.conv.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.gcn.conv.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.residual.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.residual.0.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.residual.1.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.residual.1.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.residual.1.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.residual.1.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.residual.1.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.0.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.0.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.0.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.0.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.2.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.2.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.3.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.3.num_batches_tracked
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.3.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.3.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.4.tcn.3.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.5.gcn.conv.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.5.gcn.conv.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.5.tcn.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:90: Importing initializer: st_gcn_networks.5.tcn.0.num_batches_tracked


......


[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Gather_709 [Gather] inputs: [880 -> (5)], [881 -> ()],
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:942: Using Gather axis: 0
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Gather_709 for ONNX node: Gather_709
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 882 for ONNX tensor: 882
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Gather_709 [Gather] outputs: [882 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Shape_710 [Shape]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 847
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Shape_710 [Shape] inputs: [847 -> (3, 18, 18)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Shape_710 for ONNX node: Shape_710
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 883 for ONNX tensor: 883
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Shape_710 [Shape] outputs: [883 -> (3)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Constant_711 [Constant]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Constant_711 [Constant] inputs:
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Constant_711 [Constant] outputs: [884 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Gather_712 [Gather]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 883
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 884
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Gather_712 [Gather] inputs: [883 -> (3)], [884 -> ()],
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:942: Using Gather axis: 0
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Gather_712 for ONNX node: Gather_712
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 885 for ONNX tensor: 885
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Gather_712 [Gather] outputs: [885 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Shape_713 [Shape]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 847
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Shape_713 [Shape] inputs: [847 -> (3, 18, 18)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Shape_713 for ONNX node: Shape_713
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 886 for ONNX tensor: 886
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Shape_713 [Shape] outputs: [886 -> (3)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Constant_714 [Constant]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Constant_714 [Constant] inputs:
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Constant_714 [Constant] outputs: [887 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Gather_715 [Gather]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 886
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 887
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Gather_715 [Gather] inputs: [886 -> (3)], [887 -> ()],
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:942: Using Gather axis: 0
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Gather_715 for ONNX node: Gather_715
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 888 for ONNX tensor: 888
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Gather_715 [Gather] outputs: [888 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Shape_716 [Shape]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 847
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Shape_716 [Shape] inputs: [847 -> (3, 18, 18)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Shape_716 for ONNX node: Shape_716
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 889 for ONNX tensor: 889
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Shape_716 [Shape] outputs: [889 -> (3)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Constant_717 [Constant]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Constant_717 [Constant] inputs:
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Constant_717 [Constant] outputs: [890 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Gather_718 [Gather]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 889
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 890
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Gather_718 [Gather] inputs: [889 -> (3)], [890 -> ()],
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:942: Using Gather axis: 0
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Gather_718 for ONNX node: Gather_718
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 891 for ONNX tensor: 891
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Gather_718 [Gather] outputs: [891 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Mul_719 [Mul]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 876
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 879
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Mul_719 [Mul] inputs: [876 -> ()], [879 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Mul_719 for ONNX node: Mul_719
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 892 for ONNX tensor: 892
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Mul_719 [Mul] outputs: [892 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Mul_720 [Mul]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 892
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 882
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Mul_720 [Mul] inputs: [892 -> ()], [882 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Mul_720 for ONNX node: Mul_720
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 893 for ONNX tensor: 893
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Mul_720 [Mul] outputs: [893 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Mul_721 [Mul]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 885
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 888
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Mul_721 [Mul] inputs: [885 -> ()], [888 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Mul_721 for ONNX node: Mul_721
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 894 for ONNX tensor: 894
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Mul_721 [Mul] outputs: [894 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_722 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 893
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_722 [Unsqueeze] inputs: [893 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_722 for ONNX node: Unsqueeze_722
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 895 for ONNX tensor: 895
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_722 [Unsqueeze] outputs: [895 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_723 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 894
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_723 [Unsqueeze] inputs: [894 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_723 for ONNX node: Unsqueeze_723
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 896 for ONNX tensor: 896
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_723 [Unsqueeze] outputs: [896 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Concat_724 [Concat]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 895
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 896
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Concat_724 [Concat] inputs: [895 -> (1)], [896 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Concat_724 for ONNX node: Concat_724
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 897 for ONNX tensor: 897
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Concat_724 [Concat] outputs: [897 -> (2)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Reshape_725 [Reshape]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 873
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 897
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Reshape_725 [Reshape] inputs: [873 -> (-1, -1, -1, -1, -1)], [897 -> (2)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Reshape_725 for ONNX node: Reshape_725
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 898 for ONNX tensor: 898
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Reshape_725 [Reshape] outputs: [898 -> (-1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Mul_726 [Mul]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 885
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 888
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Mul_726 [Mul] inputs: [885 -> ()], [888 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Mul_726 for ONNX node: Mul_726
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 899 for ONNX tensor: 899
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Mul_726 [Mul] outputs: [899 -> ()],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_727 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 899
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_727 [Unsqueeze] inputs: [899 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_727 for ONNX node: Unsqueeze_727
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 900 for ONNX tensor: 900
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_727 [Unsqueeze] outputs: [900 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_728 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 891
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_728 [Unsqueeze] inputs: [891 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_728 for ONNX node: Unsqueeze_728
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 901 for ONNX tensor: 901
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_728 [Unsqueeze] outputs: [901 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Concat_729 [Concat]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 900
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 901
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Concat_729 [Concat] inputs: [900 -> (1)], [901 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Concat_729 for ONNX node: Concat_729
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 902 for ONNX tensor: 902
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Concat_729 [Concat] outputs: [902 -> (2)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Reshape_730 [Reshape]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 847
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 902
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Reshape_730 [Reshape] inputs: [847 -> (3, 18, 18)], [902 -> (2)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Reshape_730 for ONNX node: Reshape_730
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 903 for ONNX tensor: 903
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Reshape_730 [Reshape] outputs: [903 -> (-1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Gemm_731 [Gemm]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 898
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 903
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Gemm_731 [Gemm] inputs: [898 -> (-1, -1)], [903 -> (-1, -1)],
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:1027: Using opA: 0 opB: 0
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:1028: GEMM: A, after squeezing: (-1, -1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Gemm_731 for ONNX node: Gemm_731
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 904 for ONNX tensor: 904
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Gemm_731 [Gemm] outputs: [904 -> (-1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_732 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 876
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_732 [Unsqueeze] inputs: [876 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_732 for ONNX node: Unsqueeze_732
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 905 for ONNX tensor: 905
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_732 [Unsqueeze] outputs: [905 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_733 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 879
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_733 [Unsqueeze] inputs: [879 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_733 for ONNX node: Unsqueeze_733
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 906 for ONNX tensor: 906
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_733 [Unsqueeze] outputs: [906 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_734 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 882
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_734 [Unsqueeze] inputs: [882 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_734 for ONNX node: Unsqueeze_734
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 907 for ONNX tensor: 907
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_734 [Unsqueeze] outputs: [907 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Unsqueeze_735 [Unsqueeze]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 891
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Unsqueeze_735 [Unsqueeze] inputs: [891 -> ()],
[07/22/2020-16:29:01] [V] [TRT] onnx2trt_utils.cpp:1411: Original shape: (), unsqueezing to: (1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Unsqueeze_735 for ONNX node: Unsqueeze_735
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 908 for ONNX tensor: 908
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Unsqueeze_735 [Unsqueeze] outputs: [908 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Concat_736 [Concat]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 905
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 906
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 907
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 908
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Concat_736 [Concat] inputs: [905 -> (1)], [906 -> (1)], [907 -> (1)], [908 -> (1)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Concat_736 for ONNX node: Concat_736
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 909 for ONNX tensor: 909
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Concat_736 [Concat] outputs: [909 -> (4)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Reshape_737 [Reshape]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 904
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 909
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Reshape_737 [Reshape] inputs: [904 -> (-1, -1)], [909 -> (4)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Reshape_737 for ONNX node: Reshape_737
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 910 for ONNX tensor: 910
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Reshape_737 [Reshape] outputs: [910 -> (-1, -1, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: BatchNormalization_738 [BatchNormalization]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 910
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.0.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.0.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.0.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.0.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: BatchNormalization_738 [BatchNormalization] inputs: [910 -> (-1, -1, -1, -1)], [st_gcn_networks.9.tcn.0.weight -> (256)], [st_gcn_networks.9.tcn.0.bias -> (256)], [st_gcn_networks.9.tcn.0.running_mean -> (256)], [st_gcn_networks.9.tcn.0.running_var -> (256)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: BatchNormalization_738 for ONNX node: BatchNormalization_738
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 911 for ONNX tensor: 911
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: BatchNormalization_738 [BatchNormalization] outputs: [911 -> (-1, -1, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Relu_739 [Relu]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 911
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Relu_739 [Relu] inputs: [911 -> (-1, -1, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Relu_739 for ONNX node: Relu_739
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 912 for ONNX tensor: 912
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Relu_739 [Relu] outputs: [912 -> (-1, -1, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Conv_740 [Conv]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 912
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.2.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.2.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Conv_740 [Conv] inputs: [912 -> (-1, -1, -1, -1)], [st_gcn_networks.9.tcn.2.weight -> (256, 256, 9, 1)], [st_gcn_networks.9.tcn.2.bias -> (256)],
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:442: Convolution input dimensions: (-1, -1, -1, -1)
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:524: Using kernel: (9, 1), strides: (1, 1), padding: (4, 0), dilations: (1, 1), numOutputs: 256
[07/22/2020-16:29:01] [V] [TRT] builtin_op_importers.cpp:525: Convolution output dimensions: (-1, 256, -1, -1)
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Conv_740 for ONNX node: Conv_740
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 913 for ONNX tensor: 913
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Conv_740 [Conv] outputs: [913 -> (-1, 256, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: BatchNormalization_741 [BatchNormalization]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 913
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.3.weight
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.3.bias
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.3.running_mean
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: st_gcn_networks.9.tcn.3.running_var
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: BatchNormalization_741 [BatchNormalization] inputs: [913 -> (-1, 256, -1, -1)], [st_gcn_networks.9.tcn.3.weight -> (256)], [st_gcn_networks.9.tcn.3.bias -> (256)], [st_gcn_networks.9.tcn.3.running_mean -> (256)], [st_gcn_networks.9.tcn.3.running_var -> (256)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: BatchNormalization_741 for ONNX node: BatchNormalization_741
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 914 for ONNX tensor: 914
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: BatchNormalization_741 [BatchNormalization] outputs: [914 -> (-1, 256, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Add_742 [Add]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 914
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 846
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Add_742 [Add] inputs: [914 -> (-1, 256, -1, -1)], [846 -> (-1, 256, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Add_742 for ONNX node: Add_742
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 915 for ONNX tensor: 915
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Add_742 [Add] outputs: [915 -> (-1, 256, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Relu_743 [Relu]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 915
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Relu_743 [Relu] inputs: [915 -> (-1, 256, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:122: Registering layer: Relu_743 for ONNX node: Relu_743
[07/22/2020-16:29:01] [V] [TRT] ImporterContext.hpp:97: Registering tensor: 916 for ONNX tensor: 916
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Relu_743 [Relu] outputs: [916 -> (-1, 256, -1, -1)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Constant_744 [Constant]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Constant_744 [Constant] inputs:
[07/22/2020-16:29:01] [W] [TRT] onnx2trt_utils.cpp:198: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:180: Constant_744 [Constant] outputs: [917 -> (8)],
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:107: Parsing node: Pad_745 [Pad]
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 916
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:123: Searching for input: 917
[07/22/2020-16:29:01] [V] [TRT] ModelImporter.cpp:129: Pad_745 [Pad] inputs: [916 -> (-1, 256, -1, -1)], [917 -> (8)],

AakankshaS · July 22, 2020, 10:06am

Hi @huge909,
Would you mind sharing your custom script for the same.
Thanks!

huge909 · July 23, 2020, 12:30am

Hi, @AakankshaS,
I have posted the code twice, but hidden by spam bot, don’t know why.

My script

The code is just from here.

huge909 · July 23, 2020, 11:23am

Hi, @AakankshaS,
Any ideas? I have simplified the ONNX model using onnx-simplifier, the model is here.

huge909 · July 24, 2020, 2:48am

Solved by downgrade opset_version from 12 to 10:

torch.onnx.export(model, input_tensor,'model.onnx', input_names=input_name, output_names=output_name, verbose=True,opset_version=10)

Topic		Replies	Views
Convert onnx to trt format using trtexec TensorRT tensorrt	5	5373	August 17, 2021
Run the Tacotron2 meet the problem TensorRT	7	1834	February 21, 2022
TensorRT cannot parse ONNX model TensorRT	5	1799	June 18, 2020
[TensorRT] ERROR: Network must have at least one output [TensorRT] ERROR: Network validation failed TensorRT tensorrt , cuda , onnx	10	2775	October 16, 2020
ONNX to TensorRT conversion Jetson Nano tensorrt , pytorch , onnx	4	361	July 3, 2024
[onnx->TensorRT]Parser problem about padding TensorRT	5	2125	October 19, 2021
Pytorch model convert to TensorRT engine failed TensorRT tensorrt , pytorch , onnx	5	1213	December 28, 2020
Run out of memory when creating TensorRT engine from onnx model Jetson Xavier NX tensorrt	7	2752	October 18, 2021
Parse onnx file failed: Parameter check failed. condition:condition: allDimsGtEq(windowSize, 1) && volume(windowSize) < MAX_KERNEL_DIMS_PRODUCT(nbSpat TensorRT tensorrt	3	948	May 17, 2022
Error converting gru model to TRT engine TensorRT	3	1008	October 25, 2021

Parse onnx model throwing unhandled exception

Description

Environment

Relevant Files

Others

Related topics