31. F1 score
f1_score()Asked in screens shaped like: OpenAI, Anthropic
Binary F1: the harmonic mean of precision and recall for labels y and predictions pred in {0,1}. Round to 4 decimals. If precision + recall is 0, return 0.
Examples
Example 1
Input: { "y": [1, 0, 1, 1], "pred": [1, 0, 0, 1] } Output: 0.8
Constraints
- 1 <= n <= 10**5, where n is the length of the input
- Input is non-empty
Implement f1_score(). Types below are the NumPy contract; Python and PyTorch use lists or tensors with the same names.
| Name | Type | I/O |
|---|---|---|
| y | array | Input |
| pred | array | Input |
| value | float | Output |
Same tests, three APIs. Pick a language with the chips above the editor.
NumPy
Vectorized arrays. Default for analyst, scientist, and DE screens.
Revealed one at a time. The reference implementation stays in the Solution tab.
Hint 1
Keep the function signature. Fill the body — do not rename parameters.
Hint 2
Match the rounding in the examples. Tests compare with a small numeric tolerance.
Hint 3
Switch NumPy / Python / PyTorch with the chips above the editor. Each language has its own tests.
Run your code to see stdout.
Submit to run the checks.
All tests passed on this language.
Next problem: KL divergence →Pro problem
Hard problems unlock with Pro
Easy pads stay free. Medium and hard SQL and Python problems — editor, tests, hints, and solutions — open after you upgrade to Pro or coaching.
See plansPractice free easy problems