Introduction

Introduction

This chapter walks through a complete machine learning project from start to finish. We'll use a classic example: predicting the median house price of a neighbourhood (a "district") in California from information such as income, house age, number of rooms and location.

Before writing any code, professionals spend time on questions like "What exactly are we trying to achieve?" and "How will we know if the model is good?". Skipping this is like starting to build a house without knowing who will live in it. A technically perfect model that solves the wrong problem is worthless.

Here is the full checklist we'll follow over the next topics:

1. Frame the problem and pick a metric ← this topic
2. Get the data and create a test set
3. Explore and visualise the data
4. Prepare the data
5. Engineer features
6. Train, fine-tune and evaluate models
7. Launch, monitor and maintain the system