Rate this Page

lower_pt2e_quantized_to_x86#

class torch.ao.quantization.pt2e.lowering.lower_pt2e_quantized_to_x86(model, example_inputs)[source]#

Lower a PT2E-qantized model to x86 backend.

Args: * model (torch.fx.GraphModule): a model quantized by PT2E quantization flow. * example_inputs (tuple[torch.Tensor, …]): example inputs for the model.

Return: A GraphModule lowered to x86 backend.

Return type

GraphModule