Build and interpret regression models in Python with scikit-learn — from a single-feature line to multiple, polynomial and regularised models. You will fit models, read their coefficients, and judge them honestly with MAE, RMSE and R2 so you can predict continuous values and explain what drives them.
Watch the free preview
What supervised regression actually is — free to watch, no account needed.
What you'll learn
- Build and interpret regression models with scikit-learn end to end
- Explain the cost function and the gradient-descent idea behind fitting a line
- Fit multiple and polynomial regression models and read their coefficients
- Apply Ridge and Lasso regularisation to control overfitting
- Evaluate models honestly with MAE, RMSE and R2 on held-out data
Syllabus
Linear Regression Foundations
What supervised regression actually isFree preview
The cost function and the gradient-descent idea
Fitting and Interpreting with scikit-learn
A disciplined fit — train/test split and workflow
Interpreting coefficients
Multiple Features, Metrics and Curves
Multiple regression and honest metrics
Polynomial features and overfitting
Regularisation and a Robust Workflow
Ridge and Lasso regularisation
Putting it together — a robust regression workflow
Lab — 10 Exercises & Solutions
Exercises 1–5
Exercises 6–10