THis repository presents a deep learning pipeline for binary classification of chest X-Ray images into Normal and Pneumonia classes using a fine-tuned ResNet-50 model.

Develop an end-to-end deep learning model to identify pneumonia in chest X-rays, assisting medical professionals in rapid and reliable diagnosis.

• main.ipynb: Full pipeline for loading data, training, evaluating, and visualizing the model.
• resnet_50.ipynb: Implementation and fine-tuning of the ResNet-50 architecture.

• Total Images: 5,21 chset X-ray images
• Classes: Pneumonia and Normal

Dataset is split into 'train', 'val' and 'test' directories.

• 🧠 Model: ResNet-50 with pretrained ImageNet weights
• 🧪 Task: Binary Classification - Normal vs Pneumonia
• 🌀 Transforms: Resize, Normalize, Augmentations using torchvision.transforms
• 💥 Loss Function: CrossEntropyLoss
• 🚀 Optimizer: Adam or SGD with scheduler
• 📊 Metrics: Accuracy, Precision, Recall, F1-score
• 📉 Visualizations: Training & validation curves using Matplotlib

• Python 3
• PyTorch
• Torchvision
• Matplotlib
• NumPy
• scikit-learn

• THis project uses transfer learning - leveraging a ResNet-50 pretrained on ImageNet
• Ideal for binary medical image classification tasks
• GPU support is recommended for faster training.