Deep Feature Synthesis (2D): Automated Feature Engineering

Before any model-selection or hyperparameter search runs, an AutoML pipeline has to turn raw relational tables into a flat feature matrix — and it does this automatically with Deep Feature Synthesis. This 2D companion simulator builds the same relational schema (Customers → Transactions → Categories) as its 3D twin, but renders it as three independently-drawn panels: a pannable/zoomable node graph, a live scatter plot of the best synthesized feature against a hidden target, and a ranked bar chart of every candidate feature's Pearson correlation. Tune signal strength, synthesis depth and customer count, spotlight a single aggregation primitive, and watch automated feature engineering either converge on the useful aggregate or, when the signal is weak, find nothing better than chance.