-
Notifications
You must be signed in to change notification settings - Fork 348
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.
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
Add Meta-World example #144
Conversation
Update to newest version
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @Kostis-S-Z,
Thanks for the PR, it looks great. I've added a small request for documentation, so that we can also include MetaWorld on the website. Maybe we should also think about including graphs for those examples on this page -- let's sync on slack to discuss that.
Thanks a lot for this contribution, it looks great! |
MAML-TRPO on Meta-world (#105 )
Meta-World Wrapper (adds to core library)
Since the Meta-World environment does not use the functionality of done (see more in this issue) but the cherry Runner assumes that at the end of the horizon (end of the episode length) the environment sends
done=True
a small modification to re-enable this feature was needed in order to make it compatible.Example
A simple MAML-TRPO example on the meta-learning multi-tasking benchmark environment Meta-world. This example inherits its structure from the
maml_trpo.py
file.In the
main()
function through thebenchmark
parameter, the user can choose to test the code on the 3 benchmark environments of Meta-world: ML1, ML10, ML45.Contribution Checklist
If your contribution modifies code in the core library (not docs, tests, or examples), please fill the following checklist.