torch_tensorrt.ptq

class torch_tensorrt.ptq.CacheCalibrator(*args, **kwargs)

Bases: object

Constructs a calibrator class in TensorRT which directly uses pre-existing cache file for calibration. :param cache_file: path to cache file. :param algo_type: choice of calibration algorithm.

class torch_tensorrt.ptq.CalibrationAlgo(value)

Bases: enum.Enum

An enumeration.

ENTROPY_CALIBRATION = <CalibrationAlgo.ENTROPY_CALIBRATION: 1>

ENTROPY_CALIBRATION_2 = <CalibrationAlgo.ENTROPY_CALIBRATION_2: 2>

LEGACY_CALIBRATION = <CalibrationAlgo.LEGACY_CALIBRATION: 0>

MINMAX_CALIBRATION = <CalibrationAlgo.MINMAX_CALIBRATION: 3>

class torch_tensorrt.ptq.DataLoaderCalibrator(*args, **kwargs)

Bases: object

Constructs a calibrator class in TensorRT and uses pytorch dataloader to load/preproces data which is passed during calibration. :param dataloader: an instance of pytorch dataloader which iterates through a given dataset. :param algo_type: choice of calibration algorithm. :param cache_file: path to cache file. :param use_cache: flag which enables usage of pre-existing cache. :param device: device on which calibration data is copied to.

torch_tensorrt.ptq.get_batch(self, names)

torch_tensorrt.ptq.get_batch_size(self)

torch_tensorrt.ptq.get_cache_mode_batch(self)

torch_tensorrt.ptq.read_calibration_cache(self)

torch_tensorrt.ptq.write_calibration_cache(self, cache)

Classes

class torch_tensorrt.ptq.DataLoaderCalibrator(*args, **kwargs)

Constructs a calibrator class in TensorRT and uses pytorch dataloader to load/preproces data which is passed during calibration. :param dataloader: an instance of pytorch dataloader which iterates through a given dataset. :param algo_type: choice of calibration algorithm. :param cache_file: path to cache file. :param use_cache: flag which enables usage of pre-existing cache. :param device: device on which calibration data is copied to.

__init__(**kwargs)

class torch_tensorrt.ptq.CacheCalibrator(*args, **kwargs)

Constructs a calibrator class in TensorRT which directly uses pre-existing cache file for calibration. :param cache_file: path to cache file. :param algo_type: choice of calibration algorithm.

__init__(**kwargs)

Enums

class torch_tensorrt.ptq.CalibrationAlgo(value)

An enumeration.