Lifelong_Relation_Detection icon indicating copy to clipboard operation
Lifelong_Relation_Detection copied to clipboard

Published 20 hours ago • verified publisher icon hongwang600

Why is the predicted label only compared against 10 randomly selected candidate labels instead of the full set of labels?

Open zhuang-li opened this issue 3 years ago • 0 comments

In the evaluation, I found that you only selected 10 candidate labels for evaluation. I this this evaluation method is quite odd. Can you explain why you chose this evaluation approach?

zhuang-li avatar Oct 31 '21 17:10 zhuang-li