vqa.pytorch icon indicating copy to clipboard operation
vqa.pytorch copied to clipboard

Information regarding training time

Open backpropper opened this issue 8 years ago • 6 comments

Can you please provide some information regarding the training time (with the hardware specifications) and the size of the dataset used?

Thanks!

backpropper avatar Oct 16 '17 20:10 backpropper

On one TitanX Pascal, on VQA2, when data loading time is closed to 0, you can expect 5 hours before convergence for a NoAtt model and 12 hours for a Att model.

Cadene avatar Oct 16 '17 22:10 Cadene

Same for VQA1 too?

backpropper avatar Oct 16 '17 22:10 backpropper

I don't remember for VQA1, but as it is composed of half of the questions, I would say 3 hours for NoAtt and 7 hours for Att. Let me know :P

Cadene avatar Oct 16 '17 22:10 Cadene

And you used the whole training set for training 100 epochs?

backpropper avatar Oct 16 '17 22:10 backpropper

train/val or trainval/test But in both cases, always the full dataset

Cadene avatar Oct 16 '17 22:10 Cadene

But you only train with the examples whose multiple choice answer is in the top n% (n determined by nans) and discard other examples. Is that correct?

backpropper avatar Oct 17 '17 05:10 backpropper