{"id":19401035,"url":"https://github.com/google-research/hyperbo","last_synced_at":"2025-04-06T10:14:07.570Z","repository":{"id":38065400,"uuid":"419072745","full_name":"google-research/hyperbo","owner":"google-research","description":"Pre-trained Gaussian processes for Bayesian optimization","archived":false,"fork":false,"pushed_at":"2024-11-25T20:57:15.000Z","size":150,"stargazers_count":90,"open_issues_count":5,"forks_count":8,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-03-30T09:08:46.100Z","etag":null,"topics":["bayesian-methods","bayesian-optimization","deep-learning","hyperparameter-tuning","machine-learning"],"latest_commit_sha":null,"homepage":"https://arxiv.org/abs/2109.08215","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/google-research.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"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":"2021-10-19T19:57:05.000Z","updated_at":"2025-02-10T17:03:53.000Z","dependencies_parsed_at":"2023-12-24T23:34:02.929Z","dependency_job_id":"86e42d12-5129-4dc0-840a-b196a282441c","html_url":"https://github.com/google-research/hyperbo","commit_stats":{"total_commits":65,"total_committers":7,"mean_commits":9.285714285714286,"dds":"0.19999999999999996","last_synced_commit":"892a4b6f8f4be87da161c33b90e3659027cf2387"},"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Fhyperbo","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Fhyperbo/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Fhyperbo/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Fhyperbo/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/google-research","download_url":"https://codeload.github.com/google-research/hyperbo/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247464226,"owners_count":20942970,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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":["bayesian-methods","bayesian-optimization","deep-learning","hyperparameter-tuning","machine-learning"],"created_at":"2024-11-10T11:16:47.837Z","updated_at":"2025-04-06T10:14:07.549Z","avatar_url":"https://github.com/google-research.png","language":"Python","readme":"# HyperBO - Prior Discovery\nA Jax/Flax codebase for the algorithm in HyperBO described in *[Pre-trained Gaussian processes for Bayesian optimization](https://www.jmlr.org/papers/v25/23-0269.html)* published in the Journal of Machine Learning Research (JMLR).\n\n**[PDF](https://arxiv.org/pdf/2109.08215.pdf)** | **[Blog post](https://ai.googleblog.com/2023/04/pre-trained-gaussian-processes-for.html)** | **[NeurIPS (Journal To Conference Track)](https://neurips.cc/virtual/2024/poster/98319) ** \n\n**[Colab Notebook](https://colab.research.google.com/github/google-research/hyperbo/blob/main/hyperbo/hyperbo_demo.ipynb)** | **[PD1 benchmark](https://github.com/google-research/hyperbo#pd1-benchmark)**\n\nDisclaimer: This is not an officially supported Google product.\n\n## Tutorial\nFollow [HyperBO's Colab Notebook](https://colab.research.google.com/github/google-research/hyperbo/blob/main/hyperbo/hyperbo_demo.ipynb) or [Jupyter Notebook](https://github.com/google-research/hyperbo/blob/main/hyperbo/hyperbo_demo.ipynb).\n\nAlso see tests for a more comprehensive understanding of the usage.\n\n## Installation\nWe recommend using Python 3.7 or 3.9 for stability.\n\nTo install the latest development version inside a virtual environment, run\n```\npython3 -m venv env-pd\nsource env-pd/bin/activate\npip install --upgrade pip\npip install \"git+https://github.com/google-research/hyperbo.git#egg=hyperbo\"\n```\n\n## PD1 benchmark\nPD1 is a new hyperparameter tuning benchmark for optimizing deep learning models. To download the PD1 dataset, please copy and paste the following link to your browser's address bar.\n```\nhttp://storage.googleapis.com/gresearch/pint/pd1.tar.gz\n```\nSee pd1/README.txt for more information. The data is licensed under the CC-BY 4.0 license.\n\nIf you'd like to use the evaluations at each training step, the relevant columns of the data frame are\n```\n'valid/ce_loss'\n'train/ce_loss',\n'train/error_rate',\n```\netc. They will hold arrays aligned with the global_step column that indicates what training step the measurement was taken at.\n\nSee the \"best_\\*\" columns for the best measurement achieved over training.\n\n\n## GPax\n[GPax](https://github.com/google-research/gpax) is a modular implementation of Gaussian processes used by HyperBO based on [Tensorflow Probability](https://www.tensorflow.org/probability) with Jax backend.\n\n## Citation\nPlease cite our work if you would like to use the code.\n```\n@article{JMLR:v25:23-0269,\n  author  = {Zi Wang and George E. Dahl and Kevin Swersky and Chansoo Lee and Zachary Nado and Justin Gilmer and Jasper Snoek and Zoubin Ghahramani},\n  title   = {Pre-trained Gaussian Processes for Bayesian Optimization},\n  journal = {Journal of Machine Learning Research},\n  year    = {2024},\n  volume  = {25},\n  number  = {212},\n  pages   = {1--83},\n  url     = {http://jmlr.org/papers/v25/23-0269.html}\n}\n```\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgoogle-research%2Fhyperbo","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgoogle-research%2Fhyperbo","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgoogle-research%2Fhyperbo/lists"}