This guide will help you get started with the PathRAG application quickly.
- Open the application in your browser
- Use one of the default accounts to log in:
- Username:
user1, Password:Pass@123 - Username:
user2, Password:Pass@123 - Username:
user3, Password:Pass@123
- Username:
- Alternatively, you can register a new account by clicking "Register" on the login page
After logging in, you'll see the dashboard with an overview of:
- Total chat conversations
- Uploaded documents
- Knowledge graph statistics
- Click on "Chats" in the sidebar
- Type your question or message in the input field at the bottom
- Press Enter or click the send button
- The AI will respond based on the knowledge in the system
- You can scroll through your chat history
- Ask specific questions for more accurate responses
- Reference documents you've uploaded for context-specific answers
- Try multi-hop questions that require connecting information across documents
- The AI uses the knowledge graph to provide more comprehensive responses
Simple questions:
- "What are the main features of PathRAG?"
- "How does PathRAG compare to traditional RAG systems?"
Complex questions:
- "What techniques developed by researchers at Stanford have been applied to knowledge graph construction?"
- "How do the approaches used in PathRAG relate to earlier work on knowledge representation?"
- Click on "Documents" in the sidebar
- Click the "Upload Document" button
- Drag and drop a file or click to select a file
- Supported formats: PDF, DOCX, MD
- The document will upload and be processed automatically
- You'll see the upload progress in real-time
- Once processed, the document will appear in the list
- Larger documents may take longer to process
- The system extracts entities and relationships from documents to build the knowledge graph
- After uploading documents, you can ask questions about them in the chat
- Upload related documents to create a richer knowledge graph with more connections
- Click on "Knowledge Graph" in the sidebar
- You'll see a visualization of the knowledge graph
- Nodes represent entities (people, organizations, concepts, etc.)
- Edges represent relationships between entities
- You can:
- Drag nodes to rearrange the graph
- Zoom in/out using the mouse wheel
- Hover over nodes to see details
- Use the search box to query specific parts of the knowledge graph
- Different colors represent different entity types
- The thickness of connections indicates the strength of relationships
- Try specific queries to focus on particular areas of knowledge
The application is fully responsive and works on mobile devices:
- Access the same URL on your mobile browser
- Log in with your credentials
- Use the menu button to access the sidebar
- All features work the same as on desktop
You can install the application as a Progressive Web App (PWA):
- Click the install icon in the address bar
- Follow the prompts to install
- Tap the menu button in your browser
- Select "Add to Home Screen" or "Install App"
- Follow the prompts
- Tap the share button
- Select "Add to Home Screen"
- Tap "Add"
Enter: Send a message in chatShift + Enter: Add a new line in the chat inputEsc: Close modals
If you encounter any issues:
- Check the README.md file for detailed documentation
- Refer to the INSTALLATION.md for setup troubleshooting
- Contact the system administrator for account-related issues