can not "polygraphy run camera.test.onnx " with pytorch-quantization

[XuechenCao]

Description

can not polygraphy run camera.test.onnx with pytorch-quantization
[E] 2: Assertion static_cast<size_t>(c) < mSet.size() failed.
[E] 2: [cgraph.h::assertIsValidSubscript::161] Error Code 2: Internal Error (Assertion static_cast<size_t>(c) < mSet.size() failed. )
[!] Invalid Engine. Please ensure the engine was built correctly
[E] FAILED | Runtime: 128.925s | Command: /root/miniconda3/bin/polygraphy run camera.test.onnx --trt --validate --trt-outputs mark all --save-results=trt_out.json --int8 --fp16

Environment

TensorRT Version:8.6

NVIDIA GPU:4090

NVIDIA Driver Version: 535.104.05

CUDA Version:12.2

CUDNN Version:

[zerollzeng]

Does it work if you don't specify --trt-outputs mark all?

[ttyio]

closing since no activity for more than 3 weeks per our policy, thanks all!

[J-xinyu]

I encountered the same problem when I used --trt-outputs mark all for my QDQ model，but it work if i don't specify --trt-outputs mark all, have any the effective solution?