New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

How train Mario game with OpenAI gym in the cloud， and download this agent， and run this agent on myself computer ？ #15

Open

jinpengchina opened this issue Oct 6, 2017 · 3 comments

jinpengchina commented Oct 6, 2017

No description provided.

hughperkins commented Apr 23, 2018

Does the lack of answers mean it's not possible?

In the worst case we can probably use xfvb perhaps? https://en.wikipedia.org/wiki/Xvfb

If it's opengl based, we could also use something like the spring headless stubs, https://github.com/hughperkins/springheadless/tree/cdb02269c0df1b492e92040682173c521e5d24d3/tools/HeadlessSpring , but I suspect that mario being 2d is not opengl-based :)

hughperkins commented Apr 23, 2018

seems like @valiro21 got it working with xfvb #9 (comment)

danielnbarbosa commented Nov 14, 2018

@hughperkins FWIW I got it training in AWS using the Unity ML Agents AMI, which does all the X Server setup for you: https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Training-on-Amazon-Web-Service.md

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment