How to use polygraphy run tools when there are custom third-party operators used?

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

user122251 October 23, 2024, 7:47am 1

I am confused when using TensorRT’s polygraphy tool. How do I use polygraphy run to compare the inference output of each layer of the onnx model and tensorRT model when there are custom third-party operators?
I have read the official instruction manual of polygraphy, but have not found the answer yet. can anyone help, thanks.

1 Like

Topic		Replies	Views
Tensorrt8.5 inference different with origin onnx model TensorRT	5	1389	January 23, 2023
Different engines give different inference results when using the same onnx model and giving the same input TensorRT	4	937	December 31, 2023
Onnx vs tensorrt different inference result TensorRT	3	3056	November 29, 2022
Big difference between infer results of onnxruntime and tensorrt TensorRT cudnn	2	54	March 20, 2025
What a difference in tensort api and onnx to trt? TensorRT	2	430	July 27, 2022
When converting BERT onnx to TensorRT engine, get different num_layers TensorRT	12	1849	February 21, 2021
TensorRT 8 : C++ inference gives different results compared to tensorflow python inference TensorRT	7	1346	October 5, 2021
8bit quantized onnx file and its 8bit engine inference results differ TensorRT tensorrt	2	683	November 21, 2021
TensorRT Engine TensorRT tensorrt	1	233	June 10, 2024
Tensorrt Model accuracy problem TensorRT tensorrt	0	96	July 3, 2024

How to use polygraphy run tools when there are custom third-party operators used?

Related topics