def onnx_export_from_quantized_model()

in optimum/amd/brevitas/export.py [0:0]

• 18 lines of code
• 1 McCabe index (conditional complexity)

def onnx_export_from_quantized_model(
    quantized_model: Union["PreTrainedModel"],
    output: Union[str, Path],
    opset: Optional[int] = None,
    optimize: Optional[str] = None,
    monolith: bool = False,
    model_kwargs: Optional[Dict[str, Any]] = None,
    custom_onnx_configs: Optional[Dict[str, "OnnxConfig"]] = None,
    fn_get_submodels: Optional[Callable] = None,
    _variant: str = "default",
    preprocessors: List = None,
    device: str = "cpu",
    no_dynamic_axes: bool = False,
    task: str = "text-generation-with-past",
    use_subprocess: bool = False,
    do_constant_folding: bool = True,
    insert_matmulinteger: bool = True,
    **kwargs_shapes,