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.

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

Feature/p2e dv3 #113

Merged
merged 37 commits into from
Nov 9, 2023
Merged

Feature/p2e dv3 #113

merged 37 commits into from
Nov 9, 2023

Conversation

michele-milesi
Copy link
Member

Summary

Describe the purpose of the pull request, including:

  • Added P2E algorithm based on Dreamer V3
  • Documentation updated

Type of Change

Please select the one relevant option below:

  • New feature (non-breaking change that adds functionality)
  • Documentation update

Checklist

Please confirm that the following tasks have been completed:

  • I have tested my changes locally and they work as expected. (Please describe the tests you performed.)
  • I have added unit tests for my changes, or updated existing tests if necessary.
  • I have updated the documentation, if applicable.
  • I have installed pre-commit and run locally for my code changes.

@michele-milesi michele-milesi changed the base branch from main to feature/split-p2e November 9, 2023 14:34
@michele-milesi michele-milesi merged commit 4ad20e8 into feature/split-p2e Nov 9, 2023
1 check passed
@michele-milesi michele-milesi deleted the feature/p2e_dv3 branch November 9, 2023 16:35
belerico added a commit that referenced this pull request Nov 13, 2023
* p2e dv1 and p2e dv2 split into exploration and finetuning

* fix: exploration amount

* fix: change actor from exploration to task when starting training

* fix: from __future__ import annotations

* fix: exploration amount

* docs: added p2e readme

* Feature/p2e dv3 (#113)

* feat: implemented p2e_dv3

* feat: added the possibility to have more critics for exploration

* tests: added p2e_dv3 test

* docs: update p2e_dv3 docs

* docs: update

* fix: p2e_dv3 refactoring

* fix: checkpoint

* Fix missing 0.5 value

* feat: add validate args to p2e_dv3

* feat: uniform p2e_dv3 with last improvements

* fix: ppo tests

* feat: split exploration and finetuning

* fix: resume from checkpoint controls

* fix: bugs

* tests: added p2e_dv3 and resume from checkpoint tests

* fix: p2e dv3 resume from checkpoint

* tests: update p2e dv3 test

* feat: added p2e_dv3 evaluation

* fix: evaluate and __init__

* fix: cli controls

* fix: added detach() when learning world model in exploration

* fix: checks in cli

* fix: exploration amount

* fix: removed minedojo test cfgs

---------

Co-authored-by: belerico_t <federico.belotti@orobix.com>

* fix: buffer load

---------

Co-authored-by: belerico_t <federico.belotti@orobix.com>
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