After the model training is complete, save the model file and export the ONNX model by using the APIs provided by PyTorch. Then use the ATC tool to convert the model into an .om file that adapts to the Ascend AI Processor for offline inference.

This section describes how to convert the trained .pth or .pth.tar file into the ONNX model. For details about how to convert the ONNX model into an .om file that adapts to the Ascend AI Processor, see "ATC Tool Instructions" in the CANN Development Auxiliary Tool Guide (Inference).

For details about how to use the Auto Tune function, see section "Auto Tune Tool Instructions" in the CANN Development Auxiliary Tool Guide (Inference).

For details about how to build an offline inference application, see the CANN Application Software Development Guide (C and C++, Inference). The process is as follows: