===============================================================================
FILES TO UPLOAD TO HUGGINGFACE SPACES
===============================================================================

✅ COPY THESE FILES TO YOUR SPACE (11 files total):

1.  app.py                    - Main application (REQUIRED - HF Spaces entry point)
2.  llm.py                    - LLM inference with local models
3.  extractors.py             - Document text extraction (DOCX/PDF)
4.  tagging.py                - Speaker tagging
5.  chunking.py               - Text chunking
6.  validation.py             - Quality validation
7.  reporting.py              - CSV/PDF report generation
8.  dashboard.py              - Dashboard generation
9.  production_logger.py      - Session logging
10. quote_extractor.py        - Quote extraction (optional but recommended)
11. requirements.txt          - Python dependencies

===============================================================================
OPTIONAL - NICE TO HAVE:
===============================================================================

- README.md                   - Documentation for your Space

===============================================================================
DO NOT UPLOAD:
===============================================================================

❌ .env                        - Contains secrets (use Spaces Variables instead)
❌ test_*.py                   - Test files
❌ *.log                       - Log files
❌ logs/                       - Log directory
❌ outputs/                    - Output directory
❌ __pycache__/                - Python cache

===============================================================================
HUGGINGFACE SPACES SETTINGS:
===============================================================================

Space SDK:       Gradio
Hardware:        GPU (T4 or better) ⚠️ IMPORTANT - CPU will be very slow!

Optional Variables (Settings → Variables):
- DEBUG_MODE = True              (to see detailed logs)
- LOCAL_MODEL = microsoft/Phi-3-mini-4k-instruct   (default, no need to set)

===============================================================================
DEPLOYMENT METHOD:
===============================================================================

Option 1: Direct Upload
- Go to your Space → Files → Upload files
- Drag and drop the 11 files above

Option 2: Git Repository
- Create a Git repo with these files
- Add .gitignore (already created)
- Connect repo to your Space
- Auto-deploys on push

===============================================================================
FIRST TIME STARTUP:
===============================================================================

1. Dependencies install: ~2-5 minutes
2. Model download: ~2-5 minutes (Phi-3-mini downloads automatically)
3. Total first startup: ~5-10 minutes

Subsequent starts: ~30-60 seconds (model is cached)

===============================================================================
VERIFICATION:
===============================================================================

Check the Logs tab - you should see:

✅ Configuration loaded for HuggingFace Spaces
🚀 TranscriptorAI Enterprise - LLM Backend: local
[Local Model] Loading microsoft/Phi-3-mini-4k-instruct...
[Local Model] ✅ Model loaded on cuda:0
Running on local URL:  http://0.0.0.0:7860

===============================================================================