Transform library for non-TFX users
Stay organized with collections
Save and categorize content based on your preferences.
Transform is available as a standalone library.
The tft module documentation is the only module that is relevant to TFX users.
The tft_beam module is relevant only when using Transform as a standalone library. Typically, a TFX user constructs a preprocessing_fn, and the rest of the
Transform library calls are made by the Transform component.
You can also use the Apache Beam MLTransform class to preprocess data
for training and inference. The MLTransform class wraps multiple TFX data
processing transforms in one class. For more information, see
Preprocess data with MLTransform
in the Apache Beam documentation.
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 2024-04-30 UTC.
[null,null,["Last updated 2024-04-30 UTC."],[],[],null,["# Transform library for non-TFX users\n\n\u003cbr /\u003e\n\nTransform is available as a standalone library.\n\n- [Getting Started with TensorFlow Transform](https://www.tensorflow.org/tfx/transform/get_started)\n- [TensorFlow Transform API Reference](https://www.tensorflow.org/tfx/transform/api_docs/python/tft)\n\nThe [`tft`](https://www.tensorflow.org/tfx/transform/api_docs/python/tft) module documentation is the only module that is relevant to TFX users.\nThe [`tft_beam`](https://www.tensorflow.org/tfx/transform/api_docs/python/tft_beam) module is relevant only when using Transform as a standalone library. Typically, a TFX user constructs a `preprocessing_fn`, and the rest of the\nTransform library calls are made by the Transform component.\n\nYou can also use the Apache Beam `MLTransform` class to preprocess data\nfor training and inference. The `MLTransform` class wraps multiple TFX data\nprocessing transforms in one class. For more information, see\n[Preprocess data with MLTransform](https://beam.apache.org/documentation/ml/preprocess-data)\nin the Apache Beam documentation."]]
