RepAdapter

RepAdapter copied to clipboard

Hello, great work! Is it possible to extend this to object detection tasks? Is that something which was pursued during this research but hasn't been included in the paper/repo due...

chandan-wiai

Your work is truly impressive! After thoroughly reading your paper, I have a question I'd appreciate your insights on. Did you utilize 1000 images directly as the training set for...

Bingpospn

no set_RepAdapter in repadater.py

1

Thank you for your wonderful work! But, I find a bug in your code. Can you debug it. Thank you.

dunanyang

Could you please release the code for video classification?

1

Shallow2000

Bottleneck dimension used in Table 5

1

In the experiment in Table 5, the bottleneck dimension and number of parameters are not described. Could you please tell me the bottleneck dimension and number of parameters used in...

tjddus9597

About vtab-1k hyperparameter setting

1

Thank you for your great work! I would like to know which hyperparameters you have searched for for VTAB-1k. Is it convenient to provide hyperparameters setting.

123456789asdfjkl

Hi, thanks for your great work! I met some problems when I set the model as `swin_base_patch4_window7_224_in22k` in train.py. It seems like the dim does't match in RepAdapter, could you...

SNEAndy

Why forward the model for 50 times

1

Hi, it's interesting here: https://github.com/luogen1996/RepAdapter/blob/9d9e4fc52d028421894d1d68c9bb551ae0415534/utils.py#L38-L39, any insights?

Zhaoyi-Yan