The new operator would take a model and an example set at the input and remove attributes that are not used by the model. The produced example set would then be more comfortable for further manipulation (e.g. export or visual inspection).
how is this different from Feature Selection?
pending comment from user
The desired functionality can be done with the current "Select by Weights" operator if models return attribute weights.
Hence, once "All supervised models should, if possible, return attribute weights" idea gets released, this idea can be marked as released. Alternatively, this idea can be closed as a duplicate.
ok thanks @yzan - will close as duplicate.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.