`neural_compressor.adaptor.torch_utils.mixed_precision`

Module Contents

Functions

ipex_mixed_precision(model[, example_inputs, device])

The function is used for ipex mixed precision quantization.

neural_compressor.adaptor.torch_utils.mixed_precision.ipex_mixed_precision(model, example_inputs=None, device='cpu')[source]: The function is used for ipex mixed precision quantization.