This PR addresses issue #419, which works perfectly for most environments. However, for some reason, this adjustment causes an error in environments with specific combinations of buildings and climates (though the vast majority function correctly).
Despite extensive debugging efforts, the origin cause remains unidentified. In these cases, the Stable Baselines 3 agent returns nans instead of valid values when the evaluation process begins.
Therefore, an alternative method to synchronize normalization between evaluation and training has been implemented, which appears to be error-free. If any other bug arise, an issue will be created to report the problem.
Types of changes
[x] Bug fix (non-breaking change which fixes an issue)
[ ] New feature (non-breaking change which adds functionality)
[ ] Breaking change (fix or feature that would cause existing functionality to change)
Description
This PR addresses issue #419, which works perfectly for most environments. However, for some reason, this adjustment causes an error in environments with specific combinations of buildings and climates (though the vast majority function correctly).
Despite extensive debugging efforts, the origin cause remains unidentified. In these cases, the Stable Baselines 3 agent returns nans instead of valid values when the evaluation process begins.
Therefore, an alternative method to synchronize normalization between evaluation and training has been implemented, which appears to be error-free. If any other bug arise, an issue will be created to report the problem.
Types of changes
Checklist:
autopep8second level aggressive.isort.cd docs && make spelling && make htmlpass (required if documentation has been updated.)pytest tests/ -vvpass. (required).pytype -d import-error sinergym/pass. (required)Changelog: