we have a min operation between the
top returning and Star Student Project relevant images. In a
large scale search system, there are always over hundreds of ground
truth relevant images to each Star Student Project query.
Therefore, dividing by Star Student Project relevant-images
would result in a very small mAP. A better alternative to be divided
by the Star Student Project number of returned images. We use
min Star Student Project relevant-images to calculate Star
Student Project As is at most in our evaluation and always
smaller than the number of Star Student Project labeled
ground truth, we simply replace min Star Student Project
with N in subsequent calculation.
https://www.linkedin.com/groups/Star-Student-Project-Star-Technology-6715168
http://www.cram.com/flashcards/star-student-project-4844694
http://www.cram.com/flashcards/star-student-project-4844694