There are other weighting schemes than F1 scores and Accuracy that checks on classification strength.
There are pair-counting metrics (if ground truth data pairs match similarly to model pairs) and information based (compare model structure to ground truth).
Describe your proposed solution
Adaptation and expansion of external valuation metrics, for possible ensembles.
Describe alternatives you've considered, if relevant
Describe the workflow you want to enable
There are other weighting schemes than F1 scores and Accuracy that checks on classification strength. There are pair-counting metrics (if ground truth data pairs match similarly to model pairs) and information based (compare model structure to ground truth).
Describe your proposed solution
Adaptation and expansion of external valuation metrics, for possible ensembles.
Describe alternatives you've considered, if relevant