tfmot.clustering.keras.ClusteringSummaries

Helper class to create tensorboard summaries for the clustering progress.

tfmot.clustering.keras.ClusteringSummaries(
    log_dir='logs', cluster_update_freq='epoch', **kwargs
)

This class is derived from tf.keras.callbacks.TensorBoard and just adds functionality to write histograms with batch-wise frequency.

Arguments
`log_dir`	The path to the directory where the log files are saved
`cluster_update_freq`	determines the frequency of updates of the clustering histograms. Same behaviour as parameter update_freq of the base class, i.e. it accepts `'batch'`, `'epoch'` or integer.

Methods

set_model(
    model
)

Sets Keras model and writes graph if specified.

set_params(
    params
)

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2023-05-26 UTC.