Commit Graph

3 Commits

Author SHA1 Message Date
mbchang
ffc87233a1
refactor GymnasiumAgent (#3927)
refactor GymnasiumAgent (for single-agent environments) to be extensible
to PettingZooAgent (multi-agent environments)
2023-05-01 20:25:03 -07:00
mbchang
3e1cb31f63
fix: add import for gymnasium (#3899) 2023-05-01 10:37:25 -07:00
mbchang
4eefea0fe8
new example: single agent, simulated environment (openai gym) (#3758)
For many applications of LLM agents, the environment is real (internet,
database, REPL, etc). However, we can also define agents to interact in
simulated environments like text-based games. This is an example of how
to create a simple agent-environment interaction loop with
[Gymnasium](https://github.com/Farama-Foundation/Gymnasium) (formerly
[OpenAI Gym](https://github.com/openai/gym)).
2023-04-28 19:52:05 -07:00