training_feat_all.lmdb and caption_train.json not found

[HubHop]

Hi, thanks for your excellent work.

I'm trying to run your code with Pretraining on Conceptual Captions, but some files do not exist, and I didn't find any script to generate them. I've tried to set up data as specified here, but the step 3 only gives me a data.mdb and a lock.mdb. Can you provide training_feat_all.lmdb and caption_train.json ?

Many thanks.

[yangapku]

The training_feat_all.lmdb should be the name of the directory storing data.mdb and lock.mdb.

[HubHop]

Hi @yangapku , thanks a lot, it works. But how do you prepare caption_train.json?

[vedanuj]

Hi @HubHop

For caption_train.json you will need to download the captions for Conceptual Captions dataset.

[zaynmi]

@HubHop Hi, where could I get the caption_train.json? Conceptual Captions dataset only provide .tsv files officially. Or should I generate images and captions by myself?

[HubHop]

Hi @zaynmi, I prepared my data based on this script:
https://github.com/jackroos/VL-BERT/blob/master/data/conceptual-captions/ReadMe.txt

Cheers

[leyuan]

Hi @HubHop what did you get after generating the Json? I got a weird format:

{"image": "val_image.zip@/00000000.jpg", "caption": ["author", ":", "a", "life", "in", "photography", "--", "in", "pictures"]}
{"image": "val_image.zip@/00000002.jpg", "caption": ["photograph", "of", "the", "sign", "being", "repaired", "by", "brave", "person"]}
{"image": "val_image.zip@/00000003.jpg", "caption": ["the", "player", "staring", "intently", "at", "a", "computer", "screen", "."]}

Did you get the caption_train.py work? Thanks.

[wu-zhonghua]

@HubHop Hi, where could I get the caption_train.json? Conceptual Captions dataset only provide .tsv files officially. Or should I generate images and captions by myself?

Have you found the caption_train.json? I still haven't found. Thanks

[wu-zhonghua]

Hi @HubHop what did you get after generating the Json? I got a weird format:

{"image": "val_image.zip@/00000000.jpg", "caption": ["author", ":", "a", "life", "in", "photography", "--", "in", "pictures"]}
{"image": "val_image.zip@/00000002.jpg", "caption": ["photograph", "of", "the", "sign", "being", "repaired", "by", "brave", "person"]}
{"image": "val_image.zip@/00000003.jpg", "caption": ["the", "player", "staring", "intently", "at", "a", "computer", "screen", "."]}

Did you get the caption_train.py work? Thanks.

Have you found the caption_train.json? I still haven't found. Thanks

[wu-zhonghua]

@leyuan @zaynmi Hi, have you found the caption_train.json files? Can you please tell me how to get it? Thanks