Update the chapter Vision Transformers - 0

[sezan92]

Objective

I was going through the course and I noticed some issues with the chapter of Intro to VT . It looked less organized to me. I think the following sections should be added

Tasks

• summary to the Vision transformer
• pros and cons of VT and CNN

[sezan92]

• Gaps i have found
• No concrete intuition / explanation behind Vision transformers
• pros and cons of cnn vs vt is written in very minimal language (not enough to explain)

[johko]

Ok, thanks. Do you want to work on it @sezan92 ?

[0xD4rky]

@johko shall I take up this issue as I have worked previously on vision transformers, you can refer this repo for example: https://github.com/0xD4rky/Vision-Transformers

[sezan92]

@johko i wanted to work on it. as @0xD4rky has interest i can set him as reviewer

[0xD4rky]

@johko i wanted to work on it. as @0xD4rky has interest i can set him as reviewer

sure that works for me!

[sezan92]

nice, @johko please assign me to this issue

[sezan92]

@johko can you upload this picture to the HF server?

[johko]

@sezan92 Sorry, but for some reason I cannot download/save the image from here.
But as always, feel free to raise a PR with the image here: https://huggingface.co/datasets/hf-vision/course-assets/discussions?status=open and ping me for a merge

[sezan92]

@johko please can you check if my pr is fine? https://huggingface.co/datasets/hf-vision/course-assets/discussions/95