The Machine Learning Classifier is difficult to learn, and the documentation is skimpy. I’ve got as far as creating a SPECIAL type of MLSkill (regular ML Skills won’t work), but am confused as how to do the initial training.
The doc only shows this:
Does this mean the training happens at runtime? Or do I train this type of ML Skill using datasets and pipelines like a regular ML Skill? Can I train it using the same datasets I will use for data extraction?