← Back to Portfolio

02 — Feature Engineering

Transforms raw demand and weather data into ML-ready features including humidex, cooling degree hours, demand momentum, and peak context variables.

feature-engineering humidex cooling-degree-hours temporal-features
Pythonpandasnumpymatplotlibseaborn

Key Findings

  • Humidex and daily CDH emerge as stronger predictors than raw temperature alone
  • Previous day's max demand captures system-level momentum that weather features miss
  • Peak context features (current threshold, peaks so far) provide essential operational framing

Notebook