tensorflowtensorflow
121

Tfx Addons

screenshot of Tfx Addons

Developers helping developers. TFX-Addons is a collection of community projects to build new components, examples, libraries, and tools for TFX. The projects are organized under the auspices of the special interest group, SIG TFX-Addons. Join the group at http://goo.gle/tfx-addons-group

Github

Overview

TFX Addons is an open source community-led project that provides additional functionality, bug fixes, and documentation for TensorFlow. The project follows the TensorFlow Code of Conduct and encourages contributions from the public. The maintainers of TFX Addons can be found in the CODEOWNERS file of the repository.

Features

  • TFX Addons package: Provides additional functionality and bug fixes for TensorFlow.
  • CI for examples: Continuous integration for example projects.
  • PyPI: Available on PyPI for easy installation on all operating systems.
  • Multiple ownership: Encourages multiple owners for all submodules to ensure active maintenance.
  • Feast example generator: Provides an example generator for the Feast library.
  • Feature selection: Includes functionality for feature selection in machine learning projects.
  • Firebase publisher: Allows for publishing to Firebase.
  • Huggingface pusher: Provides a pusher for Huggingface models.
  • Message exit handler: Includes an exit handler for messages.
  • MLMD client: Provides a client for accessing the MLMD database.
  • Model card generator: Allows for generating model cards.
  • Pandas transform: Provides transformations for Pandas data.
  • Predictions to BigQuery: Allows for storing predictions in BigQuery.
  • Sampling: Includes functionality for data sampling.
  • Schema curation: Provides tools for curating schemas.
  • XGBoost evaluator: Includes an evaluator for XGBoost models.

Summary

TFX Addons is a community-led project that enhances TensorFlow with additional functionality, bug fixes, and documentation. It provides a wide range of features, including example generators, feature selection, publishing to Firebase, Huggingface model support, MLMD client, model card generation, transformations for Pandas data, storing predictions in BigQuery, data sampling, schema curation, and XGBoost model evaluation. The project encourages contributions and multiple ownership for active maintenance. TFX Addons can be easily installed from PyPI and is available for all operating systems.