sukhmani1303/pneumonia-vit-model

Pneumonia ViT Classifier

This repository contains a pretrained Vision Transformer (ViT) model for classifying chest X-ray images as Normal or Pneumonia.

Dataset

• Source: Chest X-ray Pneumonia Dataset
• Input: 224x224 RGB images
• Classes: ['Normal', 'Pneumonia']
• Accuracy: ~90-93%

Usage

from handler import PneumoniaClassifier
from PIL import Image

classifier = PneumoniaClassifier()
image = Image.open("path/to/xray.jpg")
result = classifier.predict(image)
print(f"Prediction: {result['prediction']}, Confidence: {result['confidence']:.4f}")

Model Details

• Framework: PyTorch 2.0.1, transformers 4.31.0
• Base Model: ViT-base-patch16-224-in21k
• Format: Hugging Face transformers
• Preprocessing: RGB, resized to 224x224, normalized per ViTImageProcessor
• Output: Binary classification

Medical Disclaimer

For educational use only. Consult healthcare professionals for medical diagnosis.

Tags

Medical Imaging, Deep Learning, Vision Transformer, Pneumonia