Collating captions produces zeros about fashion-iq

[ramanishka]

@YupengGao Correct me if I'm wrong but it takes original captions and returns newly created tensor with zeros. This data loader ignores captions.

    target_images, candidate_images, captions, meta = zip(*data)
    ...
    captions = torch.zeros(len(captions), max(lengths)).long()
    for i, cap in enumerate(captions):
        end = lengths[i]
        captions[i, :end] = cap[:end]
    ...
    return target_images, candidate_images, captions, lengths, meta

[YupengGao]

@ramanishka Thanks for spotting the bug. I have fixed it.

[zxh19976666]

How to divide the datasets into training set and test set，I have download all images，but how to divide it？

[zxh19976666]

These lines apparently copy zeroes from the newly created tensor instead of actual token ids:

fashion-iq/start_kit/data_loader.py

Lines 89 to 92 in 529a8b9

	captions = torch.zeros(len(captions), max(lengths)).long()
	for i, cap in enumerate(captions):
	end = lengths[i]
	captions[i, :end] = cap[:end]

How to divide the datasets into training set and test set，I have download all images，but how to divide it？