# tensorflow/addons

Useful extra functionality for TensorFlow 2.x maintained by SIG-addons

Repository: https://github.com/tensorflow/addons
Canonical: https://ross.abutalabs.com/products/addons
Language: Python
License: Apache-2.0
License Family: permissive
Topics: machine-learning, deep-learning, tensorflow, neural-network, tensorflow-addons, python
Archived: true
Last push: 2025-09-04T15:43:46+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 40, release rhythm 8, longevity 100
- inputs: {"age_days": 2837, "days_push": 363, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1699, forks 620 (observed 2026-08-28T04:05:24.142023+00:00)

## What it is
TensorFlow Addons is a collection of extra layers, optimizers, losses, metrics, activations, callbacks, and image ops that conform to TensorFlow 2.x API patterns but are not part of core TensorFlow. It is maintained by SIG-addons and is now in minimal maintenance mode with a planned end of life, with users directed to Keras, KerasCV, and KerasNLP instead.

## Use cases
- use extra optimizers like AdamW or Lookahead with TensorFlow 2.x
- add specialized loss functions not available in core TensorFlow
- apply additional image processing ops in a TensorFlow pipeline
- use extra RNN cell types in a Keras model
- add custom activation functions and layers to TensorFlow models

## When to choose
- you are on an existing TensorFlow 2.x project that already depends on TFA
- you need a specific op, loss, or optimizer that only exists in Addons and cannot migrate yet

## When to avoid
- you are starting a new project - use Keras, KerasCV, or KerasNLP instead
- you need long-term support or new features - the project has ended development and reached end of life

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: tensorflow, neural-networks, optimizers, losses, layers, deprecated, linux, macos

## Member repositories
- tensorflow/addons (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:24.142023+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:37:48.410089+00:00, confidence not recorded.
  - readme: https://github.com/tensorflow/addons (fetched 2026-08-28T04:05:24.142023+00:00, sha cd559dea347a)
- Data as of 2026-08-30T08:39:29.467469+00:00.