huggingface / huggingface/simulate

Question about n_show and n_maps in ParallelRLEnv

Open
#369 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
195
Forks
13
Avg merge
58m
Merged PRs (30d)
1

Description

In the `ParallelRLEnv` class, I am having a doubt if the environment instances above the number set for n_show perform a function? Because in the class when you inherit from VecEnv of stable-baselines3 you set the num_envs to n_show. So an agent from SB3 will be able to interact only with those agents that are in n_show. Even in the step function of ParallelRLEnv, there is a check that the actions are given only for n_show envs.

So if n_maps is 20 and n_show is 12, is there any reason to have the extra 8 env instances since you can neither interact with them nor get observations from them.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.