Whenever a skill gets saved/updated, it should be automatically evaluated on all the datasets (with the default metric of the dataset) that were selected for the skill.
To be discussed: What happens when the skill gets deleted? Delete results? Keep them?
Whenever a skill gets saved/updated, it should be automatically evaluated on all the datasets (with the default metric of the dataset) that were selected for the skill.
To be discussed: What happens when the skill gets deleted? Delete results? Keep them?