pandas and Polars, joins, group-bys, time series, and finding the problem in the data before it becomes a problem in the model.
10h · 8 lessons · 1 challenge
Before this: NumPy & Vectorization
The first thing you do with a new dataset is find out how it is broken. This module is about that: missingness that is not random, duplicated keys, leaked columns, timestamps in three timezones, categories that appear once.
Covers pandas because that is what the ecosystem runs on, and Polars because that is increasingly what it runs on when the data is large.
A model scores 0.99 AUC. One column is cheating. Find it across ten datasets of increasing subtlety.
Worth 450 XP.
