{"id":45886652,"url":"https://github.com/sql-hkr/orlax","last_synced_at":"2026-02-27T16:04:16.462Z","repository":{"id":323849571,"uuid":"1094437686","full_name":"sql-hkr/orlax","owner":"sql-hkr","description":"ORLax - Offline Reinforcement Learning with JAX","archived":false,"fork":false,"pushed_at":"2025-11-12T11:57:12.000Z","size":162,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-11-12T13:23:19.060Z","etag":null,"topics":["jax","offline-reinforcement-learning"],"latest_commit_sha":null,"homepage":"","language":"Python","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/sql-hkr.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"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,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-11-11T17:49:49.000Z","updated_at":"2025-11-12T11:57:16.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/sql-hkr/orlax","commit_stats":null,"previous_names":["sql-hkr/orlax"],"tags_count":2,"template":false,"template_full_name":null,"purl":"pkg:github/sql-hkr/orlax","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sql-hkr%2Forlax","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sql-hkr%2Forlax/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sql-hkr%2Forlax/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sql-hkr%2Forlax/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sql-hkr","download_url":"https://codeload.github.com/sql-hkr/orlax/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sql-hkr%2Forlax/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29903617,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-27T14:46:13.553Z","status":"ssl_error","status_checked_at":"2026-02-27T14:46:10.522Z","response_time":57,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: 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":["jax","offline-reinforcement-learning"],"created_at":"2026-02-27T16:04:14.389Z","updated_at":"2026-02-27T16:04:16.454Z","avatar_url":"https://github.com/sql-hkr.png","language":"Python","readme":"# ORLax - Offline Reinforcement Learning with JAX\n\n[![PyPI version](https://img.shields.io/pypi/v/orlax)](https://pypi.org/project/orlax/)\n[![License](https://img.shields.io/github/license/sql-hkr/orlax)](LICENSE)\n\n**ORLax** is an extensible, research-friendly offline reinforcement learning framework built with JAX, Flax, and Optax. It provides clean, typed APIs optimized for editor autocompletion, modular algorithm implementations, and production-ready features like WandB logging and GPU acceleration.\n\n## Features\n\n- 🔥 **Modern JAX Stack**: Built on JAX, Flax, and Optax for high-performance GPU/TPU training\n- 📦 **Modular Design**: Clean separation of concerns with pluggable algorithms, models, and datasets\n- 🎯 **Type-Safe**: Comprehensive type hints with dataclasses instead of dict-heavy patterns\n- 📊 **Built-in Logging**: WandB integration with terminal progress bars (tqdm)\n- 🚀 **Production-Ready**: Checkpointing, multi-device training, and reproducible experiments\n- 🧪 **Research-Friendly**: Clear interfaces, and easy extensibility\n\n## Algorithms\n\n- **BC** (Behavioral Cloning) - Supervised learning from expert demonstrations\n- **CQL** (Conservative Q-Learning) - Conservative offline RL with Q-value penalties\n- **IQL** (Implicit Q-Learning) - Expectile regression-based offline RL\n\n## Installation\n\n### Using uv (Recommended)\n\n```bash\n# Clone the repository\ngit clone https://github.com/sql-hkr/orlax.git\ncd orlax\n\n# Install with uv\nuv sync\n```\n\n### Using pip\n\n```bash\n# Clone the repository\ngit clone https://github.com/sql-hkr/orlax.git\ncd orlax\n\n# Install in editable mode\npip install -e .\n```\n\n### GPU Support\n\nFor CUDA support, install JAX with CUDA:\n\n```bash\n# For CUDA 12\npip install --upgrade \"jax[cuda12]\"\n```\n\n## Quick Start\n\n### Training\n\n```bash\n# Train IQL on Hopper-Medium\nuv run orlax-train --config configs/iql_hopper.toml\n```\n\n## Citation\n\nIf you use ORLax in your research, please cite:\n\n```bibtex\n@software{orlax2025,\n  title = {ORLax: Offline Reinforcement Learning with JAX},\n  author = {sql-hkr},\n  year = {2025},\n  url = {https://github.com/sql-hkr/orlax}\n}\n```\n\n## Acknowledgments\n\n- Built with [JAX](https://github.com/google/jax), [Flax](https://github.com/google/flax), and [Optax](https://github.com/deepmind/optax)\n- Offline RL datasets from [Minari](https://github.com/Farama-Foundation/Minari) (successor to D4RL)\n\n## Contributing\n\nContributions are welcome! Please feel free to submit a Pull Request.\n\n1. Fork the repository\n2. Create your feature branch (`git checkout -b feat/amazing-feature`)\n3. Commit your changes (`git commit -m 'feat: add amazing feature'`)\n4. Push to the branch (`git push origin feat/amazing-feature`)\n5. Open a Pull Request\n\n## Contact\n\n- **Author**: sql-hkr\n- **Email**: sql.hkr@gmail.com\n- **GitHub**: [@sql-hkr](https://github.com/sql-hkr)\n- **Issues**: [GitHub Issues](https://github.com/sql-hkr/orlax/issues)\n\n---\n\n**Note**: This software is under active development. API stability is not guaranteed until version 1.0.0.\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsql-hkr%2Forlax","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsql-hkr%2Forlax","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsql-hkr%2Forlax/lists"}