Spaces:
Sleeping
Sleeping
Delete README.md
Browse files
README.md
DELETED
|
@@ -1,65 +0,0 @@
|
|
| 1 |
-
# Enhanced Arabic Document Chatbot
|
| 2 |
-
|
| 3 |
-
Advanced Arabic document chatbot with persistent knowledge base and semantic search.
|
| 4 |
-
|
| 5 |
-
## Quick Start
|
| 6 |
-
|
| 7 |
-
1. **Install dependencies**:
|
| 8 |
-
```bash
|
| 9 |
-
pip install -r requirements.txt
|
| 10 |
-
```
|
| 11 |
-
|
| 12 |
-
2. **Set up environment variables** (create `.env` file):
|
| 13 |
-
```bash
|
| 14 |
-
GEMINI_API_KEY=your_gemini_api_key_here
|
| 15 |
-
OPENAI_API_KEY=your_openai_api_key_here
|
| 16 |
-
```
|
| 17 |
-
|
| 18 |
-
3. **Run the application**:
|
| 19 |
-
```bash
|
| 20 |
-
python enhanced_main.py
|
| 21 |
-
```
|
| 22 |
-
|
| 23 |
-
4. **Access the web interface** at `http://localhost:7860`
|
| 24 |
-
|
| 25 |
-
## Features
|
| 26 |
-
|
| 27 |
-
- Automatic document indexing from configured folders
|
| 28 |
-
- Persistent vector database - no re-indexing needed
|
| 29 |
-
- Advanced semantic chunking for Arabic legal documents
|
| 30 |
-
- Context-aware retrieval with improved accuracy
|
| 31 |
-
- Background monitoring for new documents
|
| 32 |
-
- OpenAI embeddings with Gemini AI responses
|
| 33 |
-
|
| 34 |
-
## Configuration
|
| 35 |
-
|
| 36 |
-
The application monitors these directories by default:
|
| 37 |
-
- `data_cmp/data_cmp/CBK/` - Central Bank documents
|
| 38 |
-
- `data_cmp/data_cmp/CMA/` - Capital Markets documents
|
| 39 |
-
- `data_cmp/data_cmp/Legal_Principles/` - Legal principles documents
|
| 40 |
-
|
| 41 |
-
## Command Line Options
|
| 42 |
-
|
| 43 |
-
```bash
|
| 44 |
-
python enhanced_main.py --help
|
| 45 |
-
|
| 46 |
-
Options:
|
| 47 |
-
--test Run installation test
|
| 48 |
-
--reindex Force reindex all documents
|
| 49 |
-
--clear Clear knowledge base and start fresh
|
| 50 |
-
--chunking [type] Choose chunking strategy (semantic, late, hierarchical, fixed)
|
| 51 |
-
--test-chunking Test chunking strategy on sample documents
|
| 52 |
-
```
|
| 53 |
-
|
| 54 |
-
## System Requirements
|
| 55 |
-
|
| 56 |
-
- Python 3.8 or higher
|
| 57 |
-
- 8GB+ RAM recommended
|
| 58 |
-
- Internet connection for API access
|
| 59 |
-
- ~500MB disk space for vector index
|
| 60 |
-
|
| 61 |
-
## API Keys
|
| 62 |
-
|
| 63 |
-
You need API keys from:
|
| 64 |
-
- **Gemini**: Get from [Google AI Studio](https://makersuite.google.com/)
|
| 65 |
-
- **OpenAI**: Get from [OpenAI Platform](https://platform.openai.com/)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|