AI Classification vs Regression Tutorial | Free Lesson

TL;DRClassification predicts a discrete label; regression predicts a continuous number.

The two core supervised tasks. Classification outputs a category — spam/not, cat/dog/bird, fraud/legit. Regression outputs a quantity — price, temperature, score. The distinction drives everything downstream: which models, which loss function, which metrics (accuracy/F1 for classification; MAE/RMSE for regression). Identify which you have before anything else.

Key points

Classification → discrete category
Regression → continuous value
Determines model, loss, and metrics
Identify the task type first

Common mistakes

Using regression metrics on a classification task
Treating ordered categories as plain classes
Forcing a continuous target into buckets needlessly

Try it: Label each as classification or regression: churn, revenue, sentiment.

Example code

<!doctype html><html><head><meta charset="utf-8"></head>
<body style="background:#06040d;color:#e6e0ff;font-family:monospace;padding:20px"><pre>classification: spam? · cat/dog · fraud?
regression:     price · temperature · score</pre></body></html>

Open the interactive lesson →

← Self-Supervised Learning · The Modern Shift Clustering · K-Means · Dbscan →