search

NVIDIA / TensorRT

NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.
https://developer.nvidia.com/tensorrt
Apache License 2.0
10.55k stars 2.1k forks source link

TensorRT-8.4.1.4 BUG #2051

Closed triple-Mu closed 2 years ago

triple-Mu commented 2 years ago

Description

TensorRT-8.2 onnx export engine succeed. TensorRT-8.4 onnx export engine failed.

Environment

TensorRT Version: TensorRT-8.4.1.4 NVIDIA GPU: A10 NVIDIA Driver Version:510.73.08 CUDA Version: 11.6 CUDNN Version: 8.4 Operating System: ubuntu20.04 Python Version (if applicable): python3.8 Tensorflow Version (if applicable): PyTorch Version (if applicable): Baremetal or Container (if so, version):

Relevant Files

I have given my onnx.

Steps To Reproduce

This is my cmd and log:

&&&& RUNNING TensorRT.trtexec [TensorRT v8401] # trtexec --onnx=encoder_quan.onnx --saveEngine=encoder.plan --minShapes=speech:1x16x80,speech_lengths:1 --optShapes=speech:4x64x80,speech_lengths:4 --maxShapes=speech:16x256x80,speech_lengths:16 --workspace=24000 --verbose --int8
since creation.
[06/09/2022-22:40:45] [E] Error[10]: [optimizer.cpp::computeCosts::3628] Error Code 10: Internal Error (Could not find any implementation for node onnx::Conv_3304 + PPQ_Operation_58 + (Unnamed Layer* 2015) [Shuffle] + Conv_213 + PWN(Sigmoid_214, Mul_215).)
[06/09/2022-22:40:45] [E] Error[2]: [builder.cpp::buildSerializedNetwork::636] Error Code 2: Internal Error (Assertion engine != nullptr failed. )
[06/09/2022-22:40:45] [E] Engine could not be created from network
[06/09/2022-22:40:45] [E] Building engine failed
[06/09/2022-22:40:45] [E] Failed to create engine from model or file.
[06/09/2022-22:40:45] [E] Engine set up failed
&&&& FAILED TensorRT.trtexec [TensorRT v8401] # trtexec --onnx=encoder_quan.onnx --saveEngine=encoder.plan --minShapes=speech:1x16x80,speech_lengths:1 --optShapes=speech:4x64x80,speech_lengths:4 --maxShapes=speech:16x256x80,speech_lengths:16 --workspace=24000 --verbose --int8
ttyio commented 2 years ago

@triple-Mu Could you raise the trt hackathon 2022 issues in https://github.com/NVIDIA/trt-samples-for-hackathon-cn/issues , thanks!