Nadun102/medical_leaves

🌿 Indian Medicinal Leaf Classifier An AI-powered application that identifies 80 different types of Indian medicinal leaves using deep learning.

Model Details Architecture: ResNet18 with Transfer Learning Framework: PyTorch Training Data: 6,900+ images of Indian medicinal leaves Number of Classes: 80 different medicinal plants Image Size: 224x224 pixels Validation Accuracy: Check the model checkpoint for exact value Supported Medicinal Plants The model can identify 80 types of Indian medicinal leaves including:

Aloevera, Amla, Neem, Tulsi, Mint, Curry, Ginger Turmeric, Lemongrass, Hibiscus, Jasmine, Rose And 68 more Indian medicinal plants! How It Works Upload a clear image of a medicinal leaf AI Analysis - The model processes the image using computer vision Results - Get top 5 predictions with confidence scores Usage Tips For best results:

Use clear, well-lit photographs Focus on the leaf (minimize background clutter) Capture the entire leaf when possible Try multiple angles if confidence is low Technical Stack Deep Learning: PyTorch Model: ResNet18 (Transfer Learning) Interface: Gradio Deployment: Hugging Face Spaces Training Details Data Augmentation: Random flips, rotations, color jittering Optimizer: Adam Loss Function: Cross-Entropy Loss Train/Val/Test Split: 83% / 15% / 2% License MIT License

Citation If you use this model, please cite:

Indian Medicinal Leaf Classifier PyTorch ResNet18 Transfer Learning Model Year: 2025 Acknowledgments Dataset: Indian Medicinal Leaves Dataset Framework: PyTorch, Gradio Platform: Hugging Face Spaces

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference