uta-smile / TCL

code for TCL: Vision-Language Pre-Training with Triple Contrastive Learning, CVPR 2022

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Question about Data augmentation for MoCo.

celestialxevermore opened this issue · comments

Dear author, I feel thankful of your great masterpiece, and I really appreciate about your work these days.

Reading your paper with comprehending your code,

I got in my mind about data augmentation.

I found that you give data augmentation on Image, then, did you do same thing on text modalities?

If it is right, then can you check out that line in code?

Thx.

commented

Hi, thanks for your interest in our paper.
No, there is not data augmentation for text input, but we follow simCSE and
consider standard dropout as minimal data augmentation for the text.

Thx. your reply is helpful. Hope you have good days :)