Compute AUROC from Predictions

Problem

Given two arrays of equal length, y_true and y_score, write a function that returns the Area Under the ROC Curve (AUROC) for a binary classifier. y_true[i] is either 0 or 1, and y_score[i] is the predicted probability or score for the positive class. Do not use external machine learning libraries.

Constraints

• 1 <= len(y_true) == len(y_score) <= 10^5
• y_true[i] is either 0 or 1
• 0.0 <= y_score[i] <= 1.0
• Return 0.0 if there are no positive examples or no negative examples