Could not run 'aten::quantize_per_tensor' with arguments from the 'QuantizedCPU' backend

quantization
4

can you print the full quantized model? looks like there might be redundant QuantStub in layer1

1 Like