File size: 2,409 Bytes
eab4de1
 
 
 
 
 
 
 
 
 
 
b8a537c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
---

title: Chat with Wiki Documents
emoji: πŸ“š
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: "5.49.1"
app_file: app.py
pinned: false
---


# Chat with Wiki Documents

A Gradio application that allows you to chat with documents from any online wiki using Retrieval Augmented Generation (RAG).

## Features

- πŸ”— Connect to any online wiki
- πŸ”‘ Secure API key input
- πŸ’¬ Interactive chat interface
- πŸ“š Automatic document fetching and processing
- πŸ” Semantic search across wiki content
- πŸ€– Powered by OpenAI's GPT models

## Setup

1. **Install dependencies:**
   ```bash

   pip install -r requirements.txt

   ```

2. **Get an OpenAI API key:**
   - Sign up at [OpenAI](https://platform.openai.com/)
   - Create an API key from your dashboard

3. **Run the application:**
   ```bash

   python app.py

   ```

4. **Open your browser:**
   - The app will be available at `http://localhost:7860`
   - Or use the public URL provided in the terminal

## Usage

1. **Enter Wiki URL:** Paste the URL of the wiki you want to chat with
2. **Enter API Key:** Paste your OpenAI API key
3. **Click "Initialize RAG":** This will fetch and process the wiki content
4. **Start Chatting:** Ask questions about the wiki content in the chat interface

## How It Works

1. **Document Fetching:** The app fetches content from the specified wiki URL and follows links to gather multiple pages
2. **Text Processing:** Documents are split into chunks for efficient processing
3. **Embedding:** Text chunks are converted to embeddings using OpenAI's embedding model
4. **Vector Store:** Embeddings are stored in a ChromaDB vector database
5. **Retrieval:** When you ask a question, relevant chunks are retrieved using semantic search
6. **Generation:** The retrieved context is used to generate an answer using GPT

## Notes

- The app fetches up to 10 wiki pages by default
- Vector database is stored locally in `./chroma_db`
- Your API key is only used for API calls and not stored permanently
- Make sure the wiki URL is publicly accessible

## Troubleshooting

- **"Error fetching wiki content":** Check that the wiki URL is correct and publicly accessible
- **"Error initializing RAG":** Verify your API key is valid and has credits
- **Slow initialization:** Large wikis may take time to process. Be patient!

## License

MIT