search

eleurent / rl-agents

Implementations of Reinforcement Learning and Planning algorithms
MIT License
589 stars 153 forks source link

display of env 'IntersectionEnv': #87

Open swb19 opened 2 years ago

swb19 commented 2 years ago

When I run the following code, I find that although I can train normally, the visualization part always fails to work (black screen is displayed). It seems to be the problem of self.env.unwrapped.viewer.set_agent_display in evaluation.py? Have you encountered similar problems?

python experiments.py evaluate configs/IntersectionEnv/env_multi_agent.json \
                               configs/IntersectionEnv/agents/DQNAgent/ego_attention_2h.json \
                               --train --episodes=3000

image

eleurent commented 2 years ago

Hi, No that is strange, it seems to be working fine for me: image

Could you try to run a separate script where you instantiate and render highway-env manually, to see if that works?