deepchecks / deepchecks

Deepchecks: Tests for Continuous Validation of ML Models & Data. Deepchecks is a holistic open-source solution for all of your AI & ML validation needs, enabling to thoroughly test your data and models from research to production.

Home Page:https://docs.deepchecks.com/stable

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

[BUG][Tabular] bad validation on model classes

Nadav-Barak opened this issue · comments

We should validate the the model classes contain all observed labels + is of the same length of the proba

Currently it is done partially and spraded in different places in the package.

Suggestion:

In context add functions of get_predictions() and get_probabilities() and use them instead of calling the context.model.predict_proba functionality. Validation should be done on the first time those functions are called