Skip to content
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

Refactor #51

Closed
3 tasks done
Trinkle23897 opened this issue May 20, 2020 · 2 comments
Closed
3 tasks done

Refactor #51

Trinkle23897 opened this issue May 20, 2020 · 2 comments
Assignees
@Trinkle23897
Labels
enhancement Feature that is not a new algorithm or an algorithm enhancement

Comments

@Trinkle23897
Copy link
Collaborator

Trinkle23897 commented May 20, 2020 •
edited
Loading

  • DQN/DDPG/TD3/SAC with n-step return, in process_fn
  • PER interface
  • Batch over batch: do not copy
@Trinkle23897 Trinkle23897 added the enhancement Feature that is not a new algorithm or an algorithm enhancement label May 20, 2020
@Trinkle23897 Trinkle23897 self-assigned this May 20, 2020
Trinkle23897 added a commit that referenced this issue May 27, 2020
@Trinkle23897
item3 of #51
de556fd
Trinkle23897 added a commit that referenced this issue Jun 2, 2020
@Trinkle23897
compute_nstep_returns (item 2 of #51)
ff81a18
@mtaohuang
Copy link

maybe the n-step return can be moved into sample method of collector, with n-step parameter?
In this way the policy doesn't need access to the buffer or indices in the buffer.
I think this should be the way to do it, can you think of any drawbacks?

@Trinkle23897
Copy link
Collaborator Author

No, I think it is a part of the algorithm, like GAE.

BFAnas pushed a commit to BFAnas/tianshou that referenced this issue May 5, 2024
@Trinkle23897
item3 of thu-ml#51
9eaaa5c
BFAnas pushed a commit to BFAnas/tianshou that referenced this issue May 5, 2024
@Trinkle23897
compute_nstep_returns (item 2 of thu-ml#51)
0c883c7
BFAnas pushed a commit to BFAnas/tianshou that referenced this issue May 5, 2024
@Trinkle23897
nstep all (fix thu-ml#51)
28a1e82
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@Trinkle23897 Trinkle23897

Labels
enhancement Feature that is not a new algorithm or an algorithm enhancement
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@Trinkle23897 @mtaohuang