Module: tfmot.quantization.keras.experimental.default_n_bit

Module containing N-bit default quantization scheme.

Modules

default_n_bit_transforms module: Module containing N-bit default transforms.

Classes

class DefaultNBitQuantizeLayoutTransform: Default model transformations.

class DefaultNBitQuantizeRegistry: QuantizationRegistry for built-in Keras classes for default 8-bit scheme.

class DefaultNBitQuantizeScheme: Default N-Bit Scheme supported by TFLite.