{"library":"optax","type":"library","category":null,"description":"Optax is a gradient processing and optimization library designed for JAX. It provides a rich set of optimizers (Adam, SGD, etc.), learning rate schedules, and gradient transformations that can be composed to build custom optimization pipelines. It's actively developed by DeepMind/Google, with the current stable version being 0.2.8, and follows a release cadence tied to JAX ecosystem developments, often releasing minor versions for bug fixes and new features.","language":"python","status":"active","version":"0.2.8","tags":["jax","optimization","machine-learning","gradients","deep-learning"],"install":[{"cmd":"pip install optax jax jaxlib","imports":["import optax","import optax\noptimizer = optax.adam(...)","from optax import chain\noptimizer = chain(...)"]},{"cmd":"pip install 'optax[accelerate]' jax[cuda12_pip] -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html","imports":[]}],"homepage":null,"github":"https://github.com/google-deepmind/optax","docs":"https://optax.readthedocs.io/","changelog":null,"pypi":"https://pypi.org/project/optax/","npm":null,"openapi_spec":null,"status_page":null,"smithery":null,"compatibility":{"summary":{"python_range":"3.10–3.9","success_rate":43,"avg_install_s":18.6,"avg_import_s":2.67,"wheel_type":"wheel"},"url":"https://checklist.day/v1/registry/optax/compatibility"},"provenance":{"verified_status":"install_fail","verified_at":"Sun Jun 28","last_verified":"Sun Jun 28","next_check":"Sun Jul 05","install_tag":null}}