Your question is Precision and Recall Function. Start with the requirements on the right.
Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.
For a binary classification model used in a 7-Eleven customer offer pipeline, calculate precision and recall from true labels and predicted probability scores. A score greater than or equal to the supplied threshold is classified as positive.
Implement precision_recall_at_threshold without using machine learning libraries.
The function receives:
y_true: a list of integers containing only 0 and 1, where 1 is the positive class.scores: a list of numeric model scores, with one score for each label.threshold: a numeric cutoff.Return a dictionary with numeric keys precision and recall, each represented as a floating-point value. If precision or recall has a zero denominator, return 0.0 for that metric.
def precision_recall_at_threshold(y_true, scores, threshold):