hill-a / stable-baselines

A fork of OpenAI Baselines, implementations of reinforcement learning algorithms
http://stable-baselines.readthedocs.io/
MIT License
4.16k stars 725 forks source link

How do ppo2 reset environment #1083

Closed dongfangliu closed 3 years ago

dongfangliu commented 3 years ago

I check through the ppo2 script for a whole day, add i did not see any code that resets the environment when the step() function return dones=True

That's really strange

Miffyli commented 3 years ago

This is done by vectorized environments. Any default environments will be wrapped into a vectorized environment of one, which then automatically resets episodes when they hit done=True.