---
language: en
tags:
  - image-classification
  - vision
model-index:
  - name: ViT Image Classification Model
    sources:
      - https://huggingface.co/SupremoUGH/image-classification-model
    results:
      - task: 
            name: image-classification
            type: image-classification
        metrics:
          - name: Accuracy
            value: 98.0%
            type: float
library_name: transformers
license: mit
---
# Image Classification Model (ViT)

This is an image classification model based on **Vision Transformer (ViT)**, fine-tuned on the **MNIST** dataset. The model is designed to classify images into one of 10 possible classes (digits 0-9). The code is compatible with Hugging Face's inference providers and can be easily deployed.

## Model Details

- **Model Type**: Vision Transformer (ViT)
- **Base Model**: `google/vit-base-patch16-224`
- **Task**: Image Classification
- **Dataset**: MNIST (handwritten digits)
- **Labels**: 10 classes (0-9)

## How to Use

### Install Requirements

Make sure you have the following dependencies installed:

```bash
pip3 install requirements.txt
```

### Run unit tests
```bash
python3 -m unittest discover -s tests
```