{"id":18613967,"url":"https://github.com/enthought/nevergrad","last_synced_at":"2026-01-24T13:15:41.210Z","repository":{"id":66039664,"uuid":"236988560","full_name":"enthought/nevergrad","owner":"enthought","description":"Default Repo description from terraform module","archived":false,"fork":false,"pushed_at":"2020-03-04T08:36:09.000Z","size":6019,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-05-17T01:09:56.147Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/enthought.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":".github/CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2020-01-29T13:29:25.000Z","updated_at":"2025-02-18T20:22:28.000Z","dependencies_parsed_at":"2023-05-02T15:16:32.540Z","dependency_job_id":null,"html_url":"https://github.com/enthought/nevergrad","commit_stats":null,"previous_names":[],"tags_count":12,"template":false,"template_full_name":null,"purl":"pkg:github/enthought/nevergrad","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/enthought%2Fnevergrad","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/enthought%2Fnevergrad/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/enthought%2Fnevergrad/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/enthought%2Fnevergrad/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/enthought","download_url":"https://codeload.github.com/enthought/nevergrad/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/enthought%2Fnevergrad/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28728580,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-24T10:24:43.181Z","status":"ssl_error","status_checked_at":"2026-01-24T10:24:36.112Z","response_time":89,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-07T03:24:31.872Z","updated_at":"2026-01-24T13:15:41.194Z","avatar_url":"https://github.com/enthought.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"[![CircleCI](https://circleci.com/gh/facebookresearch/nevergrad/tree/master.svg?style=svg)](https://circleci.com/gh/facebookresearch/nevergrad/tree/master)\n\n# Nevergrad - A gradient-free optimization platform\n\n`nevergrad` is a Python 3.6+ library. It can be installed with:\n\n```\npip install nevergrad\n```\n\nYou can also install the master branch instead of the latest release with:\n\n```\npip install git+https://github.com/facebookresearch/nevergrad@master#egg=nevergrad\n```\n\nAlternatively, you can clone the repository and run `pip install -e .` from inside the repository folder.\n\nBy default, this only installs requirements for the optimization and instrumentation subpackages. If you are also interested in the benchmarking part,\nyou should install with the `[benchmark]` flag (example: `pip install 'nevergrad[benchmark]'`), and if you also want the test tools, use\nthe `[all]` flag (example: `pip install -e '.[all]'`).\n\n\nYou can join Nevergrad users Facebook group [here](https://www.facebook.com/groups/nevergradusers/).\n\n\n## Goals and structure\n\nThe goals of this package are to provide:\n- **gradient/derivative-free optimization algorithms**, including algorithms able to handle noise.\n- **tools to parametrize any code**, making it painless to optimize your parameters/hyperparameters, whether they are continuous, discrete or a mixture of continuous and discrete parameters.\n- **functions** on which to test the optimization algorithms.\n- **benchmark routines** in order to compare algorithms easily.\n\nThe structure of the package follows its goal, you will therefore find subpackages:\n- `optimization`: implementing optimization algorithms\n- `parametrization`: specifying what are the parameters you want to optimize\n- `functions`: implementing both simple and complex benchmark functions\n- `benchmark`: for running experiments comparing the algorithms on benchmark functions\n- `common`: a set of tools used throughout the package\n\n![Example of optimization](TwoPointsDE.gif)\n\n*Convergence of a population of points to the minima with two-points DE.*\n\n\n## Documentation\n\nThe following README is very general, here are links to find more details on:\n- [how to perform optimization](docs/optimization.md) using `nevergrad`, including using parallelization and a few recommendation on which algorithm should be used depending on the settings\n- [how to parametrize](docs/parametrization.md) your problem so that the optimizers are informed of the problem to solve. This also provides a tool to instantiate a script or non-python code into a Python function and be able to tune some of its parameters.\n- [how to benchmark](docs/benchmarking.md) all optimizers on various test functions.\n- [benchmark results](docs/benchmarks.md) of some standard optimizers an simple test cases.\n- examples of [optimization for machine learning](docs/machinelearning.md).\n- how to [contribute](.github/CONTRIBUTING.md) through issues and pull requests and how to setup your dev environment.\n- guidelines of how to contribute by [adding a new algorithm](docs/adding_an_algorithm.md).\n\n\n## Basic optimization example\n\n**All optimizers assume a centered and reduced prior at the beginning of the optimization (i.e. 0 mean and unitary standard deviation). They are however able to find solutions far from this initial prior.**\n\n\nOptimizing (minimizing!) a function using an optimizer (here `OnePlusOne`) can be easily run with:\n\n```python\nimport nevergrad as ng\n\ndef square(x):\n    return sum((x - .5)**2)\n\noptimizer = ng.optimizers.OnePlusOne(instrumentation=2, budget=100)\nrecommendation = optimizer.minimize(square)\nprint(recommendation)  # optimal args and kwargs\n\u003e\u003e\u003e Candidate(args=(array([0.500, 0.499]),), kwargs={})\n```\n\n`recommendation` holds the optimal attributes `args` and `kwargs` found by the optimizer for the provided function.\nIn this example, the optimal value will be found in `recommendation.args[0]` and will be a `np.ndarray` of size 2.\n\n`instrumentation=n` is a shortcut to state that the function has only one variable, of dimension `n`,\nSee the [instrumentation tutorial](docs/instrumentation.md) for more complex instrumentations.\n\n\nYou can print the full list of optimizers with:\n```python\nimport nevergrad as ng\nprint(list(sorted(ng.optimizers.registry.keys())))\n```\n\nThe [optimization documentation](docs/optimization.md) contains more information on how to use several workers,\ntake full control of the optimization through the `ask` and `tell` interface, perform multiobjective optimization,\nas well as pieces of advice on how to choose the proper optimizer for your problem.\n\n\n## Citing\n\n```bibtex\n@misc{nevergrad,\n    author = {J. Rapin and O. Teytaud},\n    title = {{Nevergrad - A gradient-free optimization platform}},\n    year = {2018},\n    publisher = {GitHub},\n    journal = {GitHub repository},\n    howpublished = {\\url{https://GitHub.com/FacebookResearch/Nevergrad}},\n}\n```\n\n## License\n\n`nevergrad` is released under the MIT license. See [LICENSE](LICENSE) for additional details about it.\nLGPL code is however also included in the multiobjective subpackage.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fenthought%2Fnevergrad","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fenthought%2Fnevergrad","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fenthought%2Fnevergrad/lists"}