tf.distribute.experimental.CommunicationImplementation | TensorFlow v2.16.1 (original) (raw)
tf.distribute.experimental.CommunicationImplementation
Stay organized with collections Save and categorize content based on your preferences.
Cross device communication implementation.
View aliases
Main aliases
tf.distribute.experimental.CollectiveCommunication
Compat aliases for migration
SeeMigration guide for more details.
tf.compat.v1.distribute.experimental.CollectiveCommunication, tf.compat.v1.distribute.experimental.CommunicationImplementation
AUTO: Automatically chosen by Tensorflow.RING: TensorFlow's ring algorithms for all-reduce and all-gather.NCCL: NVIDIA®'s NCCL library. This is now only used for all-reduce on GPUs; all-reduce on CPU, all-gather and broadcast fallbacks to RING.
| Class Variables | |
|---|---|
| AUTO | <CommunicationImplementation.AUTO: 'AUTO'> |
| NCCL | <CommunicationImplementation.NCCL: 'NCCL'> |
| RING | <CommunicationImplementation.RING: 'RING'> |
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.