Maybe it’s an issue with repositories in general but i think that, given the popularity of the repo (more than thousand stars) and the increase in interest in reinforcement learning in general, should think of ways to better maintain and improve . The github page has now several open issues and pull requests, some proposing very simple and useful changes, but the developers give very little feedback.

I think that it’s wonderful to have a single place to handle the environments interface and with some basic environments implemented, but I think that such place should be better , more complete and more efficient than is gym now.

Let me know what you think in the comments. I hope that someone at OpenAI will see this post and do something about this issue. A good start would be to check and eventually merge more often the pull requests.

Source link
thanks you RSS link


Please enter your comment!
Please enter your name here