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

Fix: issue 198 return metrics after initialisation of repertoires #203

Open
wants to merge 14 commits into
base: develop
Choose a base branch
from

Conversation

hannah-jan
Copy link
Collaborator

@hannah-jan hannah-jan commented Oct 1, 2024

Related issues: Issue #198

Return metrics after initialisation of repertoires.

This PR introduces:

  • returning of metrics after initialisation of repertoires
  • updates to notebooks/tests to be compatible with this

@hannah-jan hannah-jan changed the title Fix/issue 198 Fix: issue 198 return metrics after initialisation of repertoires Oct 1, 2024
@Lookatator Lookatator changed the base branch from main to develop October 1, 2024 13:06
@Lookatator Lookatator self-requested a review October 6, 2024 12:43
Copy link
Member

@Lookatator Lookatator left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the PR! 😃 It seems the init_metrics are not used in the python notebooks. Do you think we should add them to the all_metrics dictionary, and logged in the csv_logger? (this way they would appear on the plot in the end)

Lookatator and others added 14 commits February 18, 2025 10:13
Removing the following useless dependencies from requirements.txt: gym, ipython, jupyter
* a new method for MAP-Elites repertoire, that enables to samples individuals with their corresponding descriptors.
* a new output extra_info for Emitter.emit methods that is similar to the extra_scores of the scoring function, and that enables to pass information from the emit step to the state_update (necessary for DCG-MAP-Elites).
* a new DCGTransition that add desc and desc_prime to the QDTransition.
* descriptor-conditioned TD3 loss, descriptor-conditioned scoring functions, descriptor-conditioned MLP
* two new reward wrappers to clip and offset the reward (necessary for DCG-MAP-Elites).
…nt-robotics#187)

- moving to Python 3.10
- Upgrade all library versions in requirements.txt and setup.py
- Remove DM-Haiku cause it is now deprecated.
- all networks now are based on a single MLP class
- jax.tree_map has been replaced with jax.tree_util.tree_map everywhere to avoid Deprecation Warnings.
- types -> custom_types
- add extra_require for jax[cuda12]
- fix all notebooks and update typing extensions (for running notebooks)
- fix dependabot security issues
- added instructions for pip install qdax[cuda12] in README
- fix observation space in jumanji test script
…rtain domains (adaptive-intelligent-robotics#186)

* feat: add reevaluation function to compute corrected archives in uncertain domains
…tive-intelligent-robotics#185)

* change indexing to retrieve current fitness in intra batch comp

Authored-by: Lisa <lc1021@ic.ac.uk>
…nt-robotics#200)

* Move cmaes.py from core to baselines
* Change corresponding api docs

---------

Authored-by: Lisa <lc1021@ic.ac.uk>
…#195)

* fix: Change to new-style-jax-rng-keys.

---------

Co-authored-by: Milton Montero <lleramontero@gmail.com>
…ent-robotics#194)

* fix: fix typos and add codespell pre-commit hook

---------

Co-authored-by: Jeroen Van Goey <j.vangoey@instadeep.com>
…usage (adaptive-intelligent-robotics#190)

* unify descriptor naming
* Stop returning random keys
* Rename all jax.tree_utils.tree_* into jax.tree.*
* Remove Brax scoring function with init_states as arguments
* Add AURORA to notebooks in README
* Remove examples/scripts/ directory
* Rename all arguments named "unused" by "_"
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.

6 participants