Knowledge Maxxing← Back to topic library
ML MaxxingBeginner75 sec
F1 score
The F1 score is a single metric combining precision and recall into one number, using their harmonic mean, so both must be reasonably high for the score to be high.
Prerequisite: Recall
What matters
Because it penalizes a large gap between precision and recall, F1 is useful when you need one balanced number rather than tracking two metrics separately.
Example
A model with 90% precision but only 10% recall gets a low F1 score, reflecting that it is missing most real positive cases despite being precise on the few it flags.
Common misconception
F1 score is not simply the average of precision and recall; its harmonic mean punishes a large imbalance between the two more heavily than a plain average would.
Recall question
Why does F1 score drop sharply when precision and recall are very unequal?
Keywords and related concepts
F1 scoreharmonic meanprecision and recallprecision