fahmidhasan/plant-pathology-expert

PlantDoc 🌱 - Plant Pathology Expert

An AI-powered plant disease expert chatbot using Retrieval-Augmented Generation (RAG) with DeepSeek AI. Ask questions about plant pathology and get expert answers from comprehensive textbook knowledge.

🚀 Features

• —🤖 AI-Powered: Uses DeepSeek AI via OpenRouter for intelligent responses
• —📚 Expert Knowledge: Based on comprehensive Plant Pathology textbook
• —🔍 Source Citations: Provides page references for all answers
• —💬 Conversational: Maintains context across follow-up questions
• —🌿 Specialized: Focused on plant diseases, symptoms, and treatments
• —⚡ Fast & Accurate: Quick responses with reliable information

💡 How to Use

1. 1.Ask Questions: Type your plant pathology questions in the chat
2. 2.Get Expert Answers: Receive detailed responses with source citations
3. 3.Follow Up: Ask related questions for deeper understanding

🔬 Example Questions

• —"What causes tomato blight disease?"
• —"How to identify fungal diseases in plants?"
• —"What are the symptoms of bacterial wilt?"
• —"Treatment options for powdery mildew?"
• —"How do viral diseases spread in plants?"

🛠️ Technical Details

• —AI Model: DeepSeek Chat via OpenRouter
• —Knowledge Base: Plant Pathology Fifth Edition textbook
• —Vector Search: ChromaDB for semantic search
• —Embedding Model: Agricultural-specialized sentence transformer
• —Framework: Streamlit for web interface

🔐 Configuration

This Space requires an OpenRouter API key to function. The key should be set in the Space's repository secrets as OPENROUTER_API_KEY.

📖 About the Knowledge Base

The system is trained on "Plant Pathology Fifth Edition" - a comprehensive textbook covering:

• —Plant disease fundamentals
• —Pathogen biology and identification
• —Disease diagnosis and management
• —Integrated pest management
• —Resistance mechanisms
• —Environmental factors

🏗️ Architecture

The application uses a sophisticated RAG (Retrieval-Augmented Generation) pipeline:

1. 1.Document Processing: PDF content is chunked and embedded
2. 2.Vector Search: Semantic search finds relevant content
3. 3.AI Generation: DeepSeek AI generates contextual responses
4. 4.Source Attribution: Provides page references for transparency

🛠️ Local Development

To run this locally:

# Clone and install
git clone https://huggingface.co/spaces/fahmidhasan/plant-pathology-expert/
cd plant-pathology-expert
pip install -r requirements.txt

# Set up environment
echo "OPENROUTER_API_KEY=your-key-here" > .env

# Run the app
streamlit run app.py

📊 System Information

• —Knowledge Base: 2,870 text chunks from 948 pages
• —Embedding Model: Agricultural-specialized sentence transformer
• —Vector Database: ChromaDB for efficient search
• —Response Time: ~2-5 seconds per query
• —Language Support: English

🔒 Privacy & Security

• —No Data Storage: Conversations are not stored permanently
• —Secure API: Uses OpenRouter for AI processing
• —Local Processing: Embeddings and search happen locally

📚 Use Cases

Perfect for:

• —Students: Learning about plant diseases
• —Farmers: Identifying crop problems
• —Researchers: Quick reference lookup
• —Extension Workers: Disease diagnosis support
• —Gardeners: Home plant care

🤝 Contributing

This is an open-source project. Feel free to contribute improvements, report bugs, or suggest new features!

📄 License

This project is open source. Please ensure compliance with content licensing when using with copyrighted materials.

Built with ❤️ using Streamlit, DeepSeek AI, and ChromaDB

1. 1.The system will automatically adapt to your content domain

License

Software Code: MIT License (see LICENSE file)

Third-Party Content: Not included - users must provide their own PDF documents and ensure proper licensing compliance.