🌿 CottonNET

CottonNET is a Convolutional Neural Network (CNN)-based deep learning project built with PyTorch to classify cotton leaf images into four categories:

• Diseased Cotton Leaf
• Healthy Cotton Leaf
• Diseased Cotton Plant
• Healthy Cotton Plant

The project includes image preprocessing, data augmentation, and both a custom CNN model and a fine-tuned EfficientNet model to achieve high classification accuracy.

---

🌐 Try It Online

👉 CottonNET – Live Demo on Hugging Face Spaces

Upload a photo of a cotton leaf or plant to instantly get its classification!

---

🧠 Models Used

🔨 Custom CNN (from scratch)

• Built with PyTorch
• Achieved 82% accuracy

⚡ EfficientNet (Pretrained)

• Fine-tuned on the same dataset
• Achieved 96% accuracy

---

🔄 Data Preprocessing & Augmentation

• Performed using torchvision.transforms
• Includes:
  • Resizing and cropping
  • Random horizontal flips
  • Normalization
  • Random rotation

---

📊 Model Evaluation

• Visual and tabular evaluation of training and test performance
• Metrics used:
  • Accuracy
  • Loss plots

---

🚀 Getting Started

Clone the repository:

git clone https://github.com/Bvn004/CottonNET.git
cd CottonNET