You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for your studying and sharing!
I am just wondering what dataset is used for pre-trained model
Which one(MS-COCO or Visual Genome) is used?
if MS-COCO used, can I get the model pre-trained on visual genome dataset?
The text was updated successfully, but these errors were encountered:
The Visual Genome was most likely used to train the given pre-trained model as the Visual Genome provides captions which are of paragraph-length, while the MS-COCO dataset does not.
Thanks for your studying and sharing!
I am just wondering what dataset is used for pre-trained model
Which one(MS-COCO or Visual Genome) is used?
if MS-COCO used, can I get the model pre-trained on visual genome dataset?
The text was updated successfully, but these errors were encountered: