About the Flickr-30k dataset

salesforce / ALBEF

Code for ALBEF: a new vision-language pre-training method

BSD 3-Clause "New" or "Revised" License

1.53k stars 195 forks source link

About the Flickr-30k dataset #136

Open rhyhck opened 8 months ago

rhyhck commented 8 months ago

Hi,Thank you for your good codes When I have downloaded the Flickr-30k from the URL "https://www.kaggle.com/datasets/hsankesara/flickr-image-dataset" and run the code, I meet the problem as follow: "FileNotFoundError: [Errno 2] No such file or directory: '/data/renhongyi/ALBEF/ALBEF-main/data/flickr30k_images/flickr30k-images/4858070078.jpg" I found the pic don't exist in "flickr30k-images/" but exist in flickr30k_train.json. Would you tell me the reason?

shyammarjit commented 7 months ago

The code adds this folder extra flickr30k-images/ after your given path, So make sure your data path is according to this format. Suppose, for example, I have data within /home/shyam/flickr30k-images/ then I should only put /home/shyam/ in the config file.