PS33: End-to-End ML Pipeline

Expected source dataset: Churn_Modelling.csv. The supplied upload with that name was an Apple shortcut file containing a Google Drive URL, not a real CSV/XLSX data table. Put the downloaded real data in this dataset/ folder; the program uses it when valid. If none is present, it creates a clearly labelled practice dataset so all steps can be demonstrated.

Run: python solution.py
Includes preprocessing, train/test split, model fit, required metrics, and plot/tuning where applicable.
