Installation

Installing tensorflow_hub

The tensorflow_hub library can be installed alongside TensorFlow 1 and TensorFlow 2. We recommend that new users start with TensorFlow 2 right away, and current users upgrade to it.

Use with TensorFlow 2

Use pip to install TensorFlow 2 as usual. (See there for extra instructions about GPU support.) Then install a current version of tensorflow-hub next to it (must be 0.5.0 or newer).

$pipinstall"tensorflow>=2.0.0"
$pipinstall--upgradetensorflow-hub

The TF1-style API of TensorFlow Hub works with the v1 compatibility mode of TensorFlow 2.

Legacy use with TensorFlow 1

TensorFlow 1.15 is the only version of TensorFlow 1.x still supported by the tensorflow_hub library (as of release 0.11.0). TensorFlow 1.15 defaults to TF1-compatible behavior but contains many TF2 features under the hood to allow some use of TensorFlow Hub's TF2-style APIs.

$pipinstall"tensorflow>=1.15,<2.0"
$pipinstall--upgradetensorflow-hub

Use of pre-release versions

The pip packages tf-nightly and tf-hub-nightly are built automatically from the source code on github, with no release testing. This lets developers try out the latest code without building from source.

$pipinstalltf-nightly
$pipinstall--upgradetf-hub-nightly

Next Steps

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 2024年04月15日 UTC.