Pyca - Getting Started in Multi Agent RL

Pyca is a code toy that allow you to build and play with environments that support multiple learning agents.

Built on Pycolab and OpenAI’s Spinning Up

Overview

• environments
  • simple environments
• agents
  • agents that you can deploy
• experiments
  • tools
    • training
    • visualization
    • collecting demonstrations from human agents
  • data
• common
  • tools
    • tools that you can use to generate pretty pictures
  • IDEAS.md
    • send a pull request!
  • EXAMPLES.md
• README.md