I am trying to compile the inference code used in handler file and run the torchserve to load the handler file along with compiled code. But I am not able to load the compiled code after creating the MAR file. Anyway to load the compiled inference file in the handler file. This is to integrate the entire training and inferencing code with a windows based application for remote serving purpose.
Have you followed the process here for using torchserve on windows: 14. TorchServe on Windows — PyTorch/Serve master documentation?