# keras-team/keras-tuner

A Hyperparameter Tuning Library for Keras

Repository: https://github.com/keras-team/keras-tuner
Canonical: https://ross.abutalabs.com/products/keras-tuner
Homepage: https://keras.io/keras_tuner/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: automl, deep-learning, hyperparameter-optimization, keras, machine-learning, tensorflow
Last push: 2025-12-01T22:23:25+00:00

## Health v2 (maintenance only)
Score: 53/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 55, release rhythm 24, longevity 100
- inputs: {"age_days": 2645, "days_push": 275, "days_rel": 295, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2923, forks 404 (observed 2026-08-28T04:07:30.309239+00:00)

## What it is
KerasTuner is a hyperparameter optimization framework for Keras models with a define-by-run syntax for configuring search spaces. It ships with Random Search, Bayesian Optimization, Hyperband, and Grid Search algorithms and is extensible for custom search strategies.

## Use cases
- tune hyperparameters of a keras neural network
- find the best number of units and learning rate for my model
- run hyperparameter search with bayesian optimization
- use hyperband to prune bad trials early
- define a search space inline while building the model
- extend a custom search algorithm for hyperparameter tuning
- tune sklearn models

## When to choose
- you train Keras or TensorFlow models and want automated hyperparameter search
- you want define-by-run search space configuration in Python
- you need built-in Bayesian Optimization, Hyperband, or Random Search
- you are a researcher prototyping new tuning algorithms

## When to avoid
- you use PyTorch or JAX without Keras
- you need general AutoML beyond hyperparameter tuning (e.g., architecture search pipelines)
- you want a framework-agnostic tuner like Optuna

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, benchmarking
- domain: machine-learning, deep-learning, artificial-intelligence
- platform: python, cross-platform
- tags: hyperparameter-optimization, automl, keras, tensorflow, bayesian-optimization, hyperband, random-search

## Member repositories
- keras-team/keras-tuner (main) score 53

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:30.309239+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 2026-08-30T07:33:45.072421+00:00, confidence not recorded.
  - readme: https://github.com/keras-team/keras-tuner (fetched 2026-08-28T04:07:30.309239+00:00, sha 92b8709f003c)
  - homepage: https://keras.io/keras_tuner/ (fetched 2026-08-29T09:49:03.802419+00:00, sha b37c7a95dbf3)
  - registry_pypi: https://pypi.org/pypi/keras-tuner/json (fetched 2026-08-29T09:49:03.816471+00:00, sha 98e568e5379c)
  - site_page: https://keras.io/api (fetched 2026-08-29T09:49:03.811460+00:00, sha 642b042cb2af)
  - site_page: https://keras.io/2/api (fetched 2026-08-29T09:49:03.813356+00:00, sha 62a5b85795d8)
  - site_page: https://keras.io/keras_tuner/api (fetched 2026-08-29T09:49:03.814993+00:00, sha 07c2dda8b6f8)
- Data as of 2026-08-30T08:39:29.467469+00:00.