Closed ondratybl closed 2 years ago
Hi @ondratybl, you would like to apply the model to see its f1 score on unseen data, for instance?
That's kinda ML based, and I haven't actually got around to implementing such a thing. I'm not sure how difficult it would be, like it would just be how well that model runs on new data and look at what predictions match and what don't.
If you want to submit a PR please feel free to do so.
I think this may be what you're after: https://github.com/Rambatino/CHAID/pull/91
@Rambatino thanks a lot. It seems to be what I need. tree.predict() has been removed for some reason in the current version?
Yeah it never got merged, I dunno there may have been some issues with it, will have a look back into it when I get some free time
how do I predict using this class for data which have not been used for training? I see only model_predictions which only retrieves predictions for the training data... thanks a lot :)