Skip to content

Episode reset condition #621

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

Merged
merged 18 commits into from
Jul 1, 2022
Merged

Episode reset condition #621

merged 18 commits into from
Jul 1, 2022

Conversation

HenriDeh
Copy link
Member

@HenriDeh HenriDeh commented Apr 20, 2022

This PR adds a reset_condition argument to the run method and built-in conditions. It is defaulted to ResetAtTerminal() to avoid breaking changes, which will trigger the end of the episode loop when is_terminated is true.

The new possibility is to use ResetAfterNSteps(n) instead. I made a doc page for more information.

  • Update NEWS.md?
  • Before this PR can be merged, we must wait to see the final design of Trajectories.jl. A priori, it should easily integrate. The main point to be careful about is that dummy states should no longer exist, as planned in the new design.

@HenriDeh HenriDeh marked this pull request as ready for review June 30, 2022 09:36
@HenriDeh
Copy link
Member Author

As expected, this did not require many changes to work. Are you okay with this addition @findmyway ?

@findmyway
Copy link
Member

LGTM

@HenriDeh HenriDeh merged commit 89fe660 into JuliaReinforcementLearning:master Jul 1, 2022
@HenriDeh HenriDeh deleted the EpisodeResetCondition branch July 1, 2022 13:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants