onnx convert to TensorRT failed

[Rnaker]

Description

Environment

TensorRT Version: TensorRT-8.2.5.1 NVIDIA GPU: V100 32G NVIDIA Driver Version: 470.57.02 CUDA Version: 11.1 CUDNN Version: 8.0.5 Operating System: Centos7 Python Version (if applicable): 3.8 Tensorflow Version (if applicable): None PyTorch Version (if applicable): 1.10 Baremetal or Container (if so, version):

Relevant Files

Steps To Reproduce

commons:./trtexec --onnx=./model.onnx --saveEngine=./model.trt --workspace=8012 --device=0 --fp16 output: Timing Runner:(FareignNode(545...(Unnamed Layer*6352)[Shuffle]])(Myelin) trtexec:/root/gpgpu/MachineLearning/myelin/src/compiler/optimizer/autotuner.cpp:840:myelin::ir::tactic_attribute_t myelin::autotuner::au otuner t::get best tactic(): Assertion 'tactics.size()>0 &&"No tactics to implement operation"'failed.

This problem will not occur when I reduce the batch size from 32 to 20，I don't know why occur this problem?

[zerollzeng]

Can you try TRT 8.4? if it still doesn't work please provide the onnx model to us. Thanks

[aeeeeeep]

I also faced this problem, did you solve it?

[ttyio]

closing since no activity for more than 14 days, please reopen if you still have question, thanks!