pytorch-vsumm-reinforce icon indicating copy to clipboard operation
pytorch-vsumm-reinforce copied to clipboard

Published 20 hours ago • verified publisher icon KaiyangZhou

Stopping criteria

Open wjb123 opened this issue 6 years ago • 3 comments

As you said in the paper, there is a description on stopping criteria. "For all our models, we stop training after K consecutive epochs with descending summarization F-score on the validation set. We set K = 5", but I can not see any clue on this strategy from your code. Moreover, I don't see validation set in your code. Can you make some explanation ?

wjb123 avatar Sep 12 '18 02:09 wjb123

the released code is not exactly the same as the one implemented for the paper

KaiyangZhou avatar Sep 12 '18 08:09 KaiyangZhou

So how do you choose the validation set ? For SumMe dataset, there are 20 videos for training and 5 videos for testing.

wjb123 avatar Sep 12 '18 13:09 wjb123

I find the test F-score (on the test set) is the same when the model is training on different epochs, Can you tell me the reason ?

wjb123 avatar Sep 20 '18 06:09 wjb123