YFCC subset taking very long to process

[kshama2705]

Hi
I've been following the instructions given but it takes very long to read subset for YFCC. I tried to run it on a single CPU and it has been taking upwards of 8 hours to create the SQL database of the subset. Is there anyway to speed this up or any approximate time it should be processed in?
Update: the code create_subset.py has been running for 2 days now.

@xvjiarui

[YitingQu]

Same problem here

[ShunZuo-AI]

Hi I've been following the instructions given but it takes very long to read subset for YFCC. I tried to run it on a single CPU and it has been taking upwards of 8 hours to create the SQL database of the subset. Is there anyway to speed this up or any approximate time it should be processed in? Update: the code create_subset.py has been running for 2 days now.

@xvjiarui
I encountered difficulties downloading the data cc12m and cc3m, and the vast majority of the images cannot be downloaded. Could you please publish these data to hugging_face and provide us with the opportunity to download them? Thank you very much

[ShunZuo-AI]

I encountered difficulties downloading the data cc12m and cc3m, and the vast majority of the images cannot be downloaded. Could you please publish these data to hugging_face and provide us with the opportunity to download them? Thank you very much