This project provides a REST API to the OpenAI Gym library, enabling reinforcement learning agents to be developed in languages other than Python. It exposes Gym environments and their functionalities (step, reset, space information) via HTTP endpoints, allowing for cross-language development and integration.

How It Works

The project runs a local Python server that interfaces directly with the gym library. It exposes a RESTful API with endpoints for environment creation, stepping, resetting, and querying space information. A Python client is included to demonstrate interaction, and the architecture is designed to be language-agnostic, welcoming contributions for clients in other languages.

Quick Start & Requirements

• Install: pip install -r requirements.txt
• Run server: python gym_http_server.py
• Example client: python example_agent.py
• Lua client: cd binding-lua && lua example_agent.lua
• Prerequisites: Python, gym library.

Highlighted Details

• REST API for cross-language Gym interaction.
• Supports environment creation, stepping, resetting, and space querying.
• Includes Python client and Lua client examples.
• API specification detailed in README.

Maintenance & Community

The project is marked as "Archive" and no updates are expected. Contributions for clients in other languages were welcomed.

Licensing & Compatibility

The repository does not explicitly state a license in the README.

Limitations & Caveats

The project is archived and no longer maintained or updated. The env.monitor.start function's video_callable parameter is not implemented.