Fine tunning of a ViT transformers model for image classification

ViT is a transformers model that can be used for image classification. This repository will show you how to fine tune this model with your own database.

Requirements

Python dependencies

First you need to install all python libraries :

python -m pip install -r requirements.txt

Data folder structure

Create 3 subfolders : test, train, val.

Into this three folders, create a folder for each category of your data.

Usage

python runner.py

Inference Test

python inferenceTest.py