LearnFrenzy

Sign in Take a free test

Glossary · Machine Learning

What is Train-Test Split?

Dividing data into a part for training a model and a held-out part for checking how it performs on unseen data.

In more detail

Often 80/20, stratified for classification. The test set should be used once, at the end; tuning uses cross-validation or a validation set.

Example

train_test_split(X, y, test_size=0.2, stratify=y)

Learn it properlyHow Machine Learning Works: Data, Features, Training and Prediction10 min lesson · Machine Learning

Related terms

Cross-ValidationEvaluating a model by training and validating it several times on different splits of the data and averaging the results.Data LeakageWhen information unavailable at prediction time sneaks into training, making a model look better than it really is.Feature EngineeringCreating, transforming and selecting input variables so a model can learn patterns more easily.Linear RegressionA model that predicts a number as a weighted sum of input features — a best-fit straight line or plane.Precision and RecallTwo classification metrics: precision is how many flagged items were correct; recall is how many actual positives were caught.Random ForestAn ensemble model that combines many decision trees trained on random samples of data and features.