PufferAI / PufferLib

Simplifying reinforcement learning for complex game environments
https://pufferai.github.io/
MIT License
1.23k stars 58 forks source link

Fix bugs in Stochastic and improve rendered Ocean runtime woes #90

Closed xinpw8 closed 2 months ago

xinpw8 commented 2 months ago

Fix Stochastic rendering and division-by-zero error, add imageio to setup.py, silently handle conflicts b/w runtime and agent render modes, and exit rendered ocean envs gracefully.

Steps to reproduce make new pyenv python version 3.11.7 git clone https://github.com/PufferAI/PufferLib.git -b dev cd Pufferlib pip install -e .[cleanrl] python demo.py --config ocean --mode evaluate --render raylib --env stochastic