UFRN-URNAI / urnai-tools

A modular Deep Reinforcement Learning library that supports multiple environments, made with Python 3.6.

Apache License 2.0

5 stars 8 forks source link

Resolves #76 #92

Closed CinquilCinquil closed 2 months ago

CinquilCinquil commented 2 months ago

76

CinquilCinquil commented 2 months ago

I think the ActionSpace an agent receives could be replaced by an Strategy class (the class responsible for choosing the agent's next action), what do you think? @alvarofpp

It should at least receive a Strategy class if im understanding their relationship correctly.

CinquilCinquil commented 2 months ago

Hey, i tried rebasing to a newer version of v2 to make some adjustments in the code, but it appears that the old commits were duplicated when i commited the new changes. Any suggestions on how to fix this? or should i just ignore them?

alvarofpp commented 2 months ago

Any suggestions on how to fix this? or should i just ignore them?

I recommend creating a new branch from v2 and cherry-picking your commits about issue 76.

CinquilCinquil commented 2 months ago

To fix the duplicated commits i continued this issue on another branch (issue-76-2) on this pull request