Supporting multiple frameworks with TFLite (original) (raw)

The machine learning (ML) models you use with LiteRT can be trained using JAX, PyTorch or TensorFlow and then converted to a TFLite flatbuffer format.

See the following pages for more details:

• Converting from JAX
• Converting from PyTorch
• Converting from TensorFlow

An overview of the TFLite Converter which is an important component of supporting different frameworks with TFLite is on Model conversion overview.

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 2024-08-30 UTC.