Method _get_quant_cfg

tensorrt_llm/models/modeling_utils.py:199–207

(self, module_name=None)

            return False

    def _get_quant_cfg(self, module_name=None):
        if self.exclude_modules is not None:
            for exclude_module in self.exclude_modules:
                if exclude_module == module_name or (
                        exclude_module.endswith('*')
                        and module_name.startswith(exclude_module[:-1])):
                    return LayerQuantConfig(quant_algo=None,
                                            quantized_layers={})
        return self

    def _get_modelopt_qformat(self):
        algo_to_modelopt_map = {