Machine Learning / AI

You are analyzing the effectiveness of a drug, you measured its effect only on the patients who completed the study. Why might this be incorrect?

Which algorithm is less prone to overfitting: a forest of shallow trees or a forest of deep trees?

Why can't a list be used as a dictionary key in Python, but a tuple can?

Have you had experience with uplift modeling?

If you multiply one of the vectors by 10, will the cosine similarity change?

How well do you know Python?

What gradient boosting tools have you worked with? CatBoost, LightGBM, XGBoost?