New! Use Simple ML for Sheets to apply machine learning to the data in your Google Sheets
Read More
Developer Manual
Stay organized with collections
Save and categorize content based on your preferences.
Internally, TD-DF relies on
Yggdrasil Decision Forests
(YDF). Depending on the change, reading YDF's user and developer manual might be
beneficial.
The library's dependency structure is organized in layers:
- Keras
- TensorFlow
- Python utility
- Yggdrasil
New logic should be implemented where relevant. When several layers are
possibly relevant, the most generic layer should be favored.
The directory structure of TF-DF and
YDF
is a good start.
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 2022-07-08 UTC.
[null,null,["Last updated 2022-07-08 UTC."],[],[],null,["# Developer Manual\n\n\u003cbr /\u003e\n\nInternally, TD-DF relies on\n[Yggdrasil Decision Forests](https://github.com/google/yggdrasil-decision-forests)\n(YDF). Depending on the change, reading YDF's user and developer manual might be\nbeneficial.\n\nThe library's dependency structure is organized in layers:\n\n1. Keras\n2. TensorFlow\n3. Python utility\n4. Yggdrasil\n\nNew logic should be implemented where relevant. When several layers are\npossibly relevant, the most generic layer should be favored.\n\nThe directory structure of [TF-DF](/decision_forests/directory_structure) and\n[YDF](https://github.com/google/yggdrasil-decision-forests/blob/main/documentation/directory_structure.md)\nis a good start."]]
