Convert onnx to trt error on jetson xavier nv

AI & Data Science Deep Learning (Training & Inference) TensorRT
1

Description

A clear and concise description of the bug or issue.

Environment

TensorRT Version: 7.1.3
GPU Type: Tegra PCIe x4/x8 Endpoint/Root Complex
Nvidia Driver Version: nvidia-l4t-3d-core 32.4.4
CUDA Version: 10.2
CUDNN Version:
Operating System + Version: Ubuntu 18.04.5 LTS
Python Version (if applicable):
TensorFlow Version (if applicable):
PyTorch Version (if applicable): 1.0
Baremetal or Container (if container which image + tag):

Relevant Files

onnx model information：

ONNX IR version: 0.0.3
Opset version: 9
Producer name: pytorch
Producer version: 0.4
Domain:
Model version: 0
Doc string:

error information：

ERROR: …/builder/cudnnBuilderGraphShapeAnalyzer.cpp (2467) - Assertion Error in updateExtent: 0 (layer validation and shape analyzer disagree about dimensions)

Assertion `engine’ failed.

Steps To Reproduce

this is my deploy code:

        nvinfer1::IBuilder* builder = nvinfer1::createInferBuilder(iLogger_);

        // nvinfer1::INetworkDefinition* network = builder->createNetwork();
        const auto explicitBatch = 1U << static_cast<uint32_t>(nvinfer1::NetworkDefinitionCreationFlag::kEXPLICIT_BATCH);
        nvinfer1::INetworkDefinition* network = builder->createNetworkV2(explicitBatch);

        LOG(INFO)<<"Begin parsing "<<model_.name<<" model from "<<model_file_;
        auto parser = nvonnxparser::createParser(*network, iLogger_);
        // parser = nvonnxparser::createParser(*network, iLogger_);
        int verbosity = static_cast<int>(nvinfer1::ILogger::Severity::kWARNING);
        if (!parser->parseFromFile(model_file_.c_str(), verbosity))
        {
            // std::cout<<"Failed to parse onnx file"<<std::endl;
            LOG(ERROR)<<"Failed to parse model_file from "<<model_file_<<"!";
            exit(EXIT_FAILURE);
        }
        LOG(INFO)<<"End parsing "<<model_.name<<" model from "<<model_file_<<".";

        builder->setMaxBatchSize(1);
        builder->setMaxWorkspaceSize(1 << 15);

        nvinfer1::ICudaEngine* engine = builder->buildCudaEngine(*network);
        assert(engine);

when i run it , it would crash.i think maybe pytorch version is too low.

2

Hi,
Request you to share the ONNX model and the script if not shared already so that we can assist you better.
Alongside you can try few things:

  1. validating your model with the below snippet

check_model.py

import sys
import onnx
filename = yourONNXmodel
model = onnx.load(filename)
onnx.checker.check_model(model).
2) Try running your model with trtexec command.
https://github.com/NVIDIA/TensorRT/tree/master/samples/opensource/trtexec
In case you are still facing issue, request you to share the trtexec “”–verbose"" log for further debugging
Thanks!

5

Thank you，i check my model definition, there is a problem with my maxpooling op