The ML Pipeline

A machine-learning pipeline is the sequence of steps that transforms raw data into a model:

1. Data collection
2. Data preparation (including dataset splitting)
3. Training
4. Evaluation
5. Deployment

This demo focuses on the dataset split and why it matters.

Datasets and splits

A dataset is a structured collection of examples used to train or evaluate a model.

Common splits

Using the test set to make repeated decisions leaks information and makes your reported performance too optimistic.