(v3.3.1) - Sinergym observation normalization improved #408
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
The normalization wrapper in Sinergym, based on the default Gymnasium wrapper, has encountered a functionality issue.
When a model is trained using this normalization, the calibration is lost post-process. Consequently, when the model is loaded for evaluation, the absence of normalization calibration results in poor agent performance.
Motivation and Context
This PR records the averages and deviations updated during this process, and offer the ability to load them either in the constructor or via a method to use it in evaluation tasks. These evaluations are configurable, whose normalization calibrations can be not updated during that process.
Why is this change required? What problem does it solve? Please, reference issue or issues opened previously.
Fixes #407
Types of changes
Checklist:
autopep8
second level aggressive.isort
.cd docs && make spelling && make html
pass (required if documentation has been updated.)pytest tests/ -vv
pass. (required).pytype -d import-error sinergym/
pass. (required)Changelog:
NormalizeObservation
wrapper with this new functionality.Extra: