Transfer_ViT

Pretraining and finetuning different vision transformer models on the ImageNet and Ham10000 dataset.

This repository constians implementation of 11 image classifier models.
List of implemented models:

Carry out the following steps to download and preprocess the dataset:

Download the HAM10000 dataset from the following link: https://www.kaggle.com/datasets/kmader/skin-cancer-mnist-ham10000
Extract the zip file and put all put the HAM10000 folder in the parent directory.
Copy all the images from the HAM10000_images_part_2 folder and paste them into the HAM10000_images_part_1 folder
Run the data_preprocessing.py script

To implement each model run the python script with the model name

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
BeiT_pretrained.py		BeiT_pretrained.py
CaiT_basic.py		CaiT_basic.py
DeepViT_basic.py		DeepViT_basic.py
DeiT_pretrained.py		DeiT_pretrained.py
README.md		README.md
T2TViT_basic.py		T2TViT_basic.py
ViT_basic.py		ViT_basic.py
ViT_pretrained.py		ViT_pretrained.py
data_preprocesssing.py		data_preprocesssing.py
full_image_Densenet.py		full_image_Densenet.py
full_image_Inception.py		full_image_Inception.py
full_image_ResNet.py		full_image_ResNet.py
full_image_VGG.py		full_image_VGG.py