Regression

Predict a number by fitting a line to data.

Lessons

• Linear regression — Fit a straight line through data to predict a number, like price from size.
• Cost function — A cost function turns the model’s errors into a single number to minimise.
• Gradient descent — Gradient descent is how models learn: step downhill on the cost curve until you reach the bottom.
• Evaluation metrics — A trained model needs a report card: MAE, RMSE and R² each tell a slightly different story.
• Predict house prices — Assemble the whole pipeline: features → split → fit → predict → evaluate.
• Regression boss — The final boss of Machine Learning — the line, the cost, and the whole story together.