{"id":31173286,"url":"https://github.com/jrcalgo/relayrl-prototype","last_synced_at":"2025-09-19T12:48:00.309Z","repository":{"id":302365718,"uuid":"1004421942","full_name":"jrcalgo/RelayRL-prototype","owner":"jrcalgo","description":"Unstable Single-Agent Distributed Reinforcement Learning Framework written in Rust \u0026 exported to Python","archived":false,"fork":false,"pushed_at":"2025-07-10T15:39:08.000Z","size":1030,"stargazers_count":1,"open_issues_count":4,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-09-16T03:48:58.947Z","etag":null,"topics":["client-server","distributed-ml","python-bindings","pytorch","reinforcement-learning","tch-rs"],"latest_commit_sha":null,"homepage":"","language":"Rust","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/jrcalgo.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,"zenodo":null}},"created_at":"2025-06-18T15:43:16.000Z","updated_at":"2025-07-18T00:22:08.000Z","dependencies_parsed_at":"2025-07-02T02:30:49.506Z","dependency_job_id":"24c54fc9-1d5d-4506-946a-155fe3f1f371","html_url":"https://github.com/jrcalgo/RelayRL-prototype","commit_stats":null,"previous_names":["jrcalgo/relayrl-prototype"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/jrcalgo/RelayRL-prototype","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jrcalgo%2FRelayRL-prototype","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jrcalgo%2FRelayRL-prototype/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jrcalgo%2FRelayRL-prototype/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jrcalgo%2FRelayRL-prototype/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jrcalgo","download_url":"https://codeload.github.com/jrcalgo/RelayRL-prototype/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jrcalgo%2FRelayRL-prototype/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":275941525,"owners_count":25556976,"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","status":"online","status_checked_at":"2025-09-19T02:00:09.700Z","response_time":108,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":["client-server","distributed-ml","python-bindings","pytorch","reinforcement-learning","tch-rs"],"created_at":"2025-09-19T12:47:58.871Z","updated_at":"2025-09-19T12:48:00.298Z","avatar_url":"https://github.com/jrcalgo.png","language":"Rust","funding_links":[],"categories":[],"sub_categories":[],"readme":"# RelayRL, a Distributed Reinforcement Learning Framework\n\n\u003e **Warning:**  \n\u003e This project is a **prototype** and is **unstable during training**. Use at your own risk and expect breaking changes.\n\u003e \n\u003e **Platform Support:**\n\u003e RelayRL runs on **MacOS, Linux, and Windows** (x86_64). Some dependencies may require additional setup on Windows.\n\nRelayRL is a high-performance, distributed reinforcement learning (RL) framework designed for research and experimentation. By using proven transport-layer communication protocols, RelayRL integrates a client-server architecture. It provides a robust Rust backend with Python bindings, enabling seamless integration with modern ML workflows and high-speed system-level orchestration. \n\n## Single-Agent Focus\n\nRelayRL is **primarily designed for single-agent RL**.  \nWhile it is theoretically possible to launch multiple agent and training server processes for multi-agent scenarios, the framework does **not** natively support scalable multi-agent orchestration at this time.\n\n## Features\n\n- **Distributed RL Orchestration:**\n  Run agents and training servers as separate processes, communicating via gRPC or ZeroMQ.\n- **Python \u0026 Rust Interoperability:**\n  Natively supports built-in or custom Python PyTorch for RL algorithms and Rust for orchestration and performance-critical components.\n- **Extensible Algorithm/Environment Support:**\n  Plug in your own RL algorithms/environments in Python (see `examples/` for Jupyter notebooks).\n- **Benchmarking \u0026 Profiling:**\n  Tools and scripts for measuring network and runtime performance.\n- **Configurable Experiments:**\n  JSON-based configuration for easy experiment management.\n\n## Quick Start\n\n**Install RelayRL Framework:**\n  ```sh\n  pip install relayrl_framework\n  ```\n\n## Build From Source\n\n**Supported Platforms:** MacOS, Linux, Windows (x86_64)\n\n1. **Clone the repository:**\n   ```sh\n   git clone https://github.com/jrcalgo/RelayRL-prototype.git\n   cd RelayRL-prototype\n   ```\n\n2. **Build the Rust framework:**\n   ```sh\n   cd relayrl_framework\n   cargo build --release\n   ```\n\n3. **Install Python bindings (from the `relayrl_framework` directory):**\n   ```sh\n   pip install torch==2.5.1\n   pip install maturin\n   maturin develop --release\n   ```\n\n4. **Run an example (see `examples/`):**\n   - Jupyter notebooks demonstrate how to use the Python API for training and evaluation.\n  \n## Tutorial\n\n[See the examples' README!](https://github.com/jrcalgo/RelayRL-proto/blob/main/examples/README.md#how-to-use-in-novel-environments)\n\n## Directory Structure\n\n- `relayrl_framework/` — Core Rust framework and Python bindings\n- `examples/` — Example scripts and Jupyter notebooks (Python API usage)\n- `README.md` — This file (project overview)\n- `CONTRIBUTING.md` — OSS Contribution instructions for relayrl_framework crate\n- `relayrl_framework/README.md` — Framework-level and development instructions\n- `examples/README.md` — Python integration and API utilization instructions\n\n## Repository Feature Roadmap\n\n- Improved training stability and error handling for gRPC and ZeroMQ implementations\n- Cleaned up user-level APIs (RelayRLAgent, TrainingServer, ConfigLoader, etc.)\n- Better documentation and API reference\n- Benchmarking against known frameworks (RayRLLib, Stable-Baselines3, CleanRL, etc.)\n- Preservation of most critical components/logic\n\n## Related Work (Currently in-progress in a separate repository)\n\n- Improved multi-agent support and orchestration by enhancing component modularity\n- Enhanced stability, scalability and error handling during training through comprehensive network error handling, backpressure, configurable components, etc.\n- Real-time configuration updates and shutdown/init signals through lifecycle control\n- Exposing scalable components for containerized automation\n- Easing execution and external application integration through compilation into CLI executables\n- More built-in RL algorithms available\n\n## Disclaimer\n\nThis project is a **prototype** and is **unstable during training**.\nAs of now, although both transport loops are functional, ZMQ is provably more stable than gRPC in training tasks.\nAPIs and features are subject to change.  \nContributions and feedback are welcome!\n\n## Contributions\n\n[See contribution instructions!](CONTRIBUTING.md)\n\n## License\n\n[Apache License 2.0](LICENSE)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjrcalgo%2Frelayrl-prototype","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjrcalgo%2Frelayrl-prototype","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjrcalgo%2Frelayrl-prototype/lists"}