YzM1a0 / MuDPT

Multi-modal Deep Prompt Tuning for Large Pre-trained Vision-Language Models
MIT License
5 stars 1 forks source link

some questions #7

Open TsingpekTao opened 4 months ago

TsingpekTao commented 4 months ago

Dear Yongzhu Miao:

Hello!

I have read your paper and found it to be an excellent piece of work in this field. It has been very helpful for my studies in this direction. I would like to ask you a few questions:

  1. I would like to know the differences between mpt, mudpt, umudpt, and uumudpt in your code.
  2. I noticed that in your code, when calculating cross-data using uumudpt, the vit-16/B uses its own .pt file. Was there some additional processing involved in this? When I ran the code, I encountered some dimension mismatches in certain datasets. If there was any extra processing, could you please explain the specific steps? How would using mudpt or umudpt make a difference? 1719280680272

I hope you can find time in your busy schedule to answer my questions.

Thank you very much!

Sincerely,