Class probabilities in prediction analysis

[koenderks]

Add the option to add the predicted class probabilities to the data set in the prediction analysis

• Closes [Feature Request]: Add probabilities to the predictions of the ML algorithms jasp-issues#3340
• Closes [Feature Request]: probability of prediction within a machine learning model jasp-issues#3127

Steps to test:

1. Train a machine learning classification model
2. Save it to a .jaspML file using the interface
3. Load it into the prediction analysis
4. Add the predictions (including class probabilities) to the dataset

[FBartos]

The Save training model checkbox below the path specification is a bit unintuitive to use. I.e., I would assume that the model get's save automatically once I specify the path. Wouldn't it be better to just check whether the path == "" and save if it doesn't?
Or is it to prevent re-saving the model with every analysis update? Then nesting the path within the checkbox (so it is unavailble if not selected) would be better design IMO
(the add predictions checkbox works exactly in this way)

[FBartos]

otherwise works and looks well!

[koenderks]

Thanks, good point about the design. I'll think a bit about it and then do it in a future PR.