tf.nn.scale_regularization_loss | TensorFlow v2.16.1 (original) (raw)

tf.nn.scale_regularization_loss

Scales the sum of the given regularization losses by number of replicas.

View aliases

Compat aliases for migration

tf.compat.v1.nn.scale_regularization_loss

tf.nn.scale_regularization_loss(
    regularization_loss
)

Used in the notebooks

Used in the guide	Used in the tutorials
Distributed training with TensorFlow	Custom training with tf.distribute.Strategy Custom training loop with Keras and MultiWorkerMirroredStrategy Parameter server training with ParameterServerStrategy

Usage with distribution strategy and custom training loop:

with strategy.scope():
  def compute_loss(self, label, predictions):
    per_example_loss = tf.keras.losses.sparse_categorical_crossentropy(
        labels, predictions)

    # Compute loss that is scaled by sample_weight and by global batch size.
    loss = tf.nn.compute_average_loss(
        per_example_loss,
        sample_weight=sample_weight,
        global_batch_size=GLOBAL_BATCH_SIZE)

    # Add scaled regularization losses.
    loss += tf.nn.scale_regularization_loss(tf.nn.l2_loss(weights))
    return loss

Args
regularization_loss	Regularization loss.

Returns
Scalar loss value.

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. Some content is licensed under the numpy license.

Last updated 2024-04-26 UTC.