while using torch.export.onnx(…) [torch.onnx — PyTorch 1.9.0 documentation]
is there a way to get all the tensors along with the jit trace? this is primarily for layerwise validation for the model conversion …
while using torch.export.onnx(…) [torch.onnx — PyTorch 1.9.0 documentation]
is there a way to get all the tensors along with the jit trace? this is primarily for layerwise validation for the model conversion …