afe.apis.loaded_net

Attributes

GroundTruth

Classes

LoadedNet

Functions

load_model(→ LoadedNet)

Module Contents

afe.apis.loaded_net.GroundTruth

class afe.apis.loaded_net.LoadedNet(mod: afe._tvm._defines.TVMIRModule, layout: str, target: sima_utils.common.Platform, *, custom_param_data: sima_utils.common.CustomPlatformParams | None = None, output_labels: list[str] | None, model_path: str | None)

execute(inputs: afe.apis.defines.InputValues, *, log_level: int = logging.NOTSET) → list[numpy.ndarray]

quantize(calibration_data: Iterable[afe.apis.defines.InputValues] | None, quantization_config: afe.apis.defines.QuantizationParams, *, automatic_layout_conversion: bool = False, arm_only: bool = False, simulated_arm: bool = False, model_name: str | None = None, any_shape_on_mla: bool = False, log_level: int = logging.NOTSET) → afe.apis.model.Model

quantize_with_accuracy_feedback(calibration_data: Iterable[afe.apis.defines.InputValues], evaluation_data: Iterable[tuple[afe.apis.defines.InputValues, GroundTruth]], quantization_config: afe.apis.defines.QuantizationParams, *, accuracy_score: afe.driver.statistic.Statistic[tuple[list[numpy.ndarray], GroundTruth], float], target_accuracy: float, automatic_layout_conversion: bool = False, max_optimization_steps: int | None = None, model_name: str | None = None, any_shape_on_mla: bool = False, log_level: int = logging.NOTSET) → afe.apis.model.Model

afe.apis.loaded_net.load_model(params: afe.load.importers.general_importer.ImporterParams, *, target: sima_utils.common.Platform = gen1_target, custom_param_data: sima_utils.common.CustomPlatformParams | None = None, log_level: int = logging.NOTSET) → LoadedNet