precision-recall-metrics — Amazon case

The team needs a reliable implementation for fraud alert evaluation in a production-oriented ML workflow. A small implementation error can change the model input or metric enough to make a downstream result misleading.

Given binary predictions and labels, compute precision and recall without using a metrics library.

Count true positives, false positives and false negatives (label 1 is positive) and return precision and recall. A ratio whose denominator is zero is reported as 0.0.

Implement solve(y_true,y_pred).

Returns. Return a tuple (precision, recall) of Python floats.

Examples

Example 1

Input:

solve([1, 1, 0, 0], [1, 0, 1, 0])

Output:

(0.5, 0.5)

Example 2

Input:

solve([1, 1, 0], [0, 0, 0])

Output:

(0.0, 0.0)