search

YiLunLee / missing_aware_prompts

Multimodal Prompting with Missing Modalities for Visual Recognition, CVPR'23
https://yilunlee.github.io/missing_aware_prompts/
160 stars 9 forks source link

multi-layer prompt query #4

Closed ChosenOne23 closed 1 year ago

ChosenOne23 commented 1 year ago

image hello,I am interested in your work,but when I read the code,I have some questions,as u can see in the picture,why choose prompts[:,self.prompt_layers.index(i)] (shape:prompt_number,emb_dim) instead of prompts [self.prompt_layers.index(i),:] (shape:prompt_length,emb_dim).because I think we need to attach shape of (prompt_length,emb_dim) to input embedding.