F1 score

F1 score is the harmonic mean of precision and recall.

The F1 score

F1 is the harmonic mean of precision and recall: 2·P·R / (P + R). If either is low, F1 is low too; an arithmetic mean would hide this.

P, R = 1.0, 0.1
print((P + R) / 2)           # 0.55  misleading
print(2 * P * R / (P + R))   # 0.18  realistic

It is often used as a single summary metric in imbalanced problems where both false alarms and misses matter.

classification_report

classification_report gives the precision, recall, F1 and support (example count) for each class in a single table:

from sklearn.metrics import classification_report
print(classification_report(y_test, y_pred, digits=3))

In the bottom rows, macro avg weights classes equally and weighted avg by class size.

Exercise Four metrics

For the given labels, print accuracy, precision, recall and F1 with round(..., 3) in a single print, separated by spaces.

Show example solution

from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score

y_true = [1, 1, 1, 1, 0, 0, 0, 0, 0, 0]
y_pred = [1, 1, 1, 0, 1, 0, 0, 0, 0, 0]

print(round(accuracy_score(y_true, y_pred), 3), round(precision_score(y_true, y_pred), 3), round(recall_score(y_true, y_pred), 3), round(f1_score(y_true, y_pred), 3))

Test yourself

If precision is 0.5 and recall is 0.5, what is F1?
- 1.0
- 0.75
- 0.5
- 0.25
Show answer

0.5 — 2 · 0.5 · 0.5 / (0.5 + 0.5) = 0.5; if both are equal, F1 is the same.

Compute the F1 score.

f = ____(y_true, y_pred)

Show answer

f = f1_score(y_true, y_pred)

Match the report rows.

precision

Hit rate when saying positive
recall

Share of positives caught
f1-score

Harmonic mean
support

Examples in the class

Next lesson: Summary - Model evaluation

We Are Enhancing Your Experience

Cookie Preferences

Strictly Necessary Cookies

Analytical Cookies