Which evaluation metric should I use?
It depends on what the errors cost. Recall where missing a positive case is expensive, precision where false alarms are, F1 where you need a balance, and ROC AUC for ranking quality across thresholds. Accuracy alone is almost never sufficient, particularly with imbalanced classes.
This comes up on Machine Learning Coursework, where it is answered in the context of the work itself.
Not what you asked?
Ask us the actual question
Send the draft, the rubric or the score report with it. You get a real answer and a price before you commit to anything.