Different result between onnx and tensorrt in pow layer. Is it reasonable?


I convert my onnx model to tensorrt model. But when I do tensorrt inference, I got a significant different result compare to onnx result. I print two intermedia result from the ends of a simple pow layer. I want to know whether is it reasonable

the convert command line:trtexec --onnx="$onnx" \ --saveEngine="$engine_file" \ --verbose --explicitBatch --workspace=262144 --fp16=disable

tensorrt version : TensorRT v8204


Could you please try on the latest TensorRT version 8.4 EA. If still face this issue, please share with us the issue repro model/script and steps to try from our end.

Thank you.