How do i get bare bones of ViT transformers

I want to have the ViT model’s version where it isnt trained.So basically that just the code for it.How can i do this? Im new so i have no clue help me out please

I bet you want the default setting of ViT model in huggingface. You may try:

from transformers import ViTModel, ViTConfig


Yeah I figured that out, but how to I proceed from there for training ?

Maybe you can take a look on this tutorial?

This seems to be using a pre-trained version by what I could see