What is the F1 score?

The F1 score is the harmonic mean of precision and recall: F1 = 2 * (Precision * Recall) / (Precision + Recall). It balances both metrics into a single number between 0 and 1, where 1 is perfect and 0 is worst.

Why use the harmonic mean instead of the arithmetic mean?

The harmonic mean penalizes extreme imbalances between precision and recall. A model with 100% precision but 0% recall would get an arithmetic mean of 50%, but an F1 of 0% — correctly reflecting that it is useless.

Do I need True Negatives to calculate F1?

No. The F1 score formula only uses True Positives, False Positives, and False Negatives. True Negatives do not factor into precision, recall, or F1.

What is a good F1 score?

It depends on the domain. Above 0.90 is typically excellent; 0.70–0.90 is good for many real-world tasks; below 0.50 usually indicates the model needs significant improvement. Always compare against a baseline appropriate to your dataset.

What is the difference between macro, micro, and weighted F1?

This tool computes binary F1 for a single class. Macro-F1 averages the per-class F1 scores equally; micro-F1 aggregates TP/FP/FN across all classes before computing; weighted-F1 weights each class by its support. For binary problems they are equivalent.

What if precision or recall is undefined (division by zero)?

Precision is undefined when TP + FP = 0 (no positive predictions). Recall is undefined when TP + FN = 0 (no actual positives). In both cases, F1 is also undefined. This tool displays "N/A" in those cases.

F1 Score Calculator

Name: F1 Score Calculator
Availability: InStock
Author: Nham Vu

Enter TP, FP, and FN counts to instantly compute precision, recall, and F1 score for your binary classifier.

Confusion Matrix Inputs

Enter counts from your classifier's confusion matrix.

True Positives (TP)

Positive cases correctly predicted as positive.

False Positives (FP)

Negative cases incorrectly predicted as positive.

False Negatives (FN)

Positive cases incorrectly predicted as negative.

Precision

—

TP / (TP + FP)

Recall

—

TP / (TP + FN)

F1 Score

—

2 × P × R / (P + R)

F1 Score Gauge —

0.00 0.50 1.00

Confusion Matrix Summary

Predicted Positive Predicted Negative

Actual Positive

True Positive

—

Actual Positive

False Negative

—

Actual Negative

False Positive

—

Actual Negative

True Negative

N/A

True Negatives are not required to compute F1.

Copied!

Summary

Enter TP, FP, and FN counts to instantly compute precision, recall, and F1 score for your binary classifier.

How it works

Enter the number of True Positives — correctly predicted positive cases.
Enter the number of False Positives — negative cases incorrectly predicted as positive.
Enter the number of False Negatives — positive cases incorrectly predicted as negative.
Precision, recall, and F1 score are computed and displayed instantly.
A confusion-matrix summary card shows all four quadrants for quick reference.
Copy any metric value to clipboard with the copy button.

Use cases

Evaluate spam-detection or fraud-detection models where class imbalance is common.
Compare two classifier configurations using a single F1 metric.
Check how recall and precision trade off as you adjust a decision threshold.
Quickly sanity-check model outputs after a training run.
Teaching or studying the relationship between precision, recall, and F1.
Debugging classifiers where false negatives are especially costly (e.g., medical diagnosis).
Reporting model performance in a research paper or project writeup.
Validating NLP named-entity-recognition or information-retrieval results.

Frequently Asked Questions

Related tools

Percentage Calculator Standard Deviation Mean Median Mode

Last updated: 2026-05-23 · Reviewed by Nham Vu