Can existing large datasets be used to fine tune the blip'large_caption task?
#29
by shams123321 - opened
Can existing large datasets be used to fine tune the blip'large_caption task?
The dataset used is UFine6926 dataset, as its text description of images is very fine-grained, with an average of 80.8 words per image. Can this dataset be used to fine tune the caption task of blip?
Looking forward to your reply!
Thank you!