F1 Score
ELI5 — The Vibe Check
The F1 Score is the balanced average of precision and recall — a single number that captures both. If you want a model that's both precise (doesn't cry wolf) AND has high recall (finds everything), the F1 score tells you how well it balances these. It's the metric that doesn't let a bad precision hide behind a good recall.
Real Talk
The F1 Score is the harmonic mean of precision and recall: 2 * (precision * recall) / (precision + recall). It ranges from 0 to 1 and balances the tradeoff between false positives and false negatives. It is particularly useful for imbalanced datasets. The F-beta score generalizes this to weight precision and recall differently.
When You'll Hear This
"Use F1 score when class distribution is uneven." / "F1 score dropped when we tuned for precision."
Related Terms
Accuracy
Accuracy is the simplest way to score a model — what percentage of predictions were correct.
Classification
Classification is teaching an AI to sort things into categories. Is this email spam or not? Is this image a cat, dog, or bird?
Precision
Precision asks: 'Of all the times the AI said YES, how often was it actually right?
Recall
Recall asks: 'Of all the actual YES cases in the world, how many did the AI catch?' High recall means the model finds almost everything it should.