Hi. I’m looking for a simple basic example of inference model trained with Pytorch using C++.
I know how to convert Pytorch model to Onnx or with torch2trt converter. So i’m looking for code how to use these converted models using C++.
Can you give some helpful links?
Hi,
In general, we expect the workflow like this: pyTorch → onnx → TensorRT engine.
So please convert your model into onnx format first.
After that, you can follow this sample to convert the onnx model into TensorRT:
/usr/src/tensorrt/samples/sampleOnnxMNIST
Thanks.