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
Hi, I want to use lens on Birds classification task. But the released vocab_tags obtained from laion does not contain all the bird classnames of my dataset, thus the final accuracy is poor.
I want to ask how to build my own vocab_tags for my datasets.
The text was updated successfully, but these errors were encountered:
Hi @kai-wen-yang , you might need first to extend the vocabulary with these new classes and then generate the CLIP weights associated with the new vocabulary. We will release code for all these steps too!
Hi @kai-wen-yang , you might need first to extend the vocabulary with these new classes and then generate the CLIP weights associated with the new vocabulary. We will release code for all these steps too!
Excuse me, when will these steps be released?
Hi, I want to use lens on Birds classification task. But the released vocab_tags obtained from laion does not contain all the bird classnames of my dataset, thus the final accuracy is poor.
I want to ask how to build my own vocab_tags for my datasets.
The text was updated successfully, but these errors were encountered: