Module: mediapipe_model_maker.model_util

View source on GitHub

Utilities for models.

Modules

dataset module: Common dataset for model training and evaluation.

quantization module: Libraries for post-training quantization.

Classes

class LiteRunner: A runner to do inference with the TFLite model.

class WarmUp: Applies a warmup schedule on a given learning rate decay schedule.

Functions

convert_to_tflite(...): Converts the input Keras model to TFLite format.

convert_to_tflite_from_file(...): Converts the input Keras model to TFLite format.

get_default_callbacks(...): Gets default callbacks.

get_lite_runner(...): Returns a LiteRunner from flatbuffer of the TFLite model.

get_steps_per_epoch(...): Gets the estimated training steps per epoch.

load_keras_model(...): Loads a tensorflow Keras model from file and returns the Keras model.

load_tflite_model_buffer(...): Loads a TFLite model buffer from file.

save_tflite(...): Saves TFLite file to tflite_file.