search

TheButlah / makrl

makrl - modular algorithm kit for reinforcement learning
4 stars 1 forks source link

Merge version 0.0.1 into dev #22

Closed TheButlah closed 5 years ago

TheButlah commented 5 years ago

I have finished the framework directory structure and the API for Model and subclasses. Please review this PR and get it merged into dev.

TheButlah commented 5 years ago

Update: after review, PolicyModel was only partially updated to account for n-step methods. I will fix that and then re-open the pull request.