sgrvinod
on Pytorch==1.4.0
I am trying to run this project on Pytorch==1.4.0, torchvision==0.5.0
Before training, I had done some changes in train.py:
scores, _ = pack_padded_sequence(scores, decode_lengths, batch_first=True)
targets, _ = pack_padded_sequence(targets, decode_lengths, batch_first=True)
to:
scores = pack_padded_sequence(scores, decode_lengths, batch_first=True)
targets = pack_padded_sequence(targets, decode_lengths, batch_first=True)
scores=scores.data
targets=targets.data
But when I run caption.py, the output is always ['<start>', 'a', 'white', 'bird', 'in', 'the', 'air', '<end>'].
why?
Have you solved this problem? I'm having the same issue
This error may be caused by improper preprocessing.
