I would like to come back on this "When do we require to calculate the confidence Interval?" since recently a reviewer asked me to provide confidence intervals for metrics regarding my work about object detection (precision, recall, F-score). I have never had this kind of request before and never seen anything like that on papers I had the chance to read or review. My idea it to follow this approach but was wondering if it's really necessary given a test set sufficiently large (how large to be considered "safe"?).
Asked
Active
Viewed 44 times
1
-
Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. – Community Aug 23 '23 at 13:29