can it add language model(such as BERT) as a backbone? #37
Answered
by
WenbinLee
wendilinplay
asked this question in
Q&A
-
Hi, Just wondering if the fine-tune would work adding a pertained language model as a new backbone. |
Beta Was this translation helpful? Give feedback.
Answered by
WenbinLee
Apr 14, 2022
Replies: 1 comment
-
Hi, this is an interesting question. However, because this framework is mainly built for the vision task, I am afraid it will be difficult to directly add a pre-trained language model. If you want to conduct a multi-modality model based on LibFewShot, it needs to take effort to modify the entire framework. Hope this can help you. |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
wendilinplay
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi, this is an interesting question. However, because this framework is mainly built for the vision task, I am afraid it will be difficult to directly add a pre-trained language model. If you want to conduct a multi-modality model based on LibFewShot, it needs to take effort to modify the entire framework. Hope this can help you.