neural_compressor.tensorflow.quantization.autotune

Intel Neural Compressor Tensorflow quantization AutoTune API.

Functions

get_all_config_set(...)

Get all config set.

autotune(...)

The main entry of auto-tune.

Module Contents

neural_compressor.tensorflow.quantization.autotune.get_all_config_set() → neural_compressor.common.base_config.BaseConfig | List[neural_compressor.common.base_config.BaseConfig][source]

Get all config set.

neural_compressor.tensorflow.quantization.autotune.autotune(model: str | tensorflow.keras.Model | neural_compressor.tensorflow.utils.BaseModel, tune_config: neural_compressor.common.base_tuning.TuningConfig, eval_fn: Callable, eval_args: Tuple[Any] | None = None, calib_dataloader: Callable = None, calib_iteration: int = 100, calib_func: Callable = None) → neural_compressor.tensorflow.utils.BaseModel | None[source]

The main entry of auto-tune.