REQUIRED_FILES_FOR_SPACES.md

Required Files for HuggingFace Spaces Deployment

✅ CRITICAL - Must Upload These Files

Main Application

• app.py - Main Gradio application

Core Processing Modules

• llm.py - LLM inference (local model support)
• extractors.py - DOCX/PDF text extraction
• tagging.py - Speaker identification
• chunking.py - Semantic text chunking
• validation.py - Quality scoring and validation
• reporting.py - CSV/PDF report generation
• dashboard.py - Dashboard generation
• production_logger.py - Session logging

Optional but Recommended

• quote_extractor.py - Market research quote extraction (now optional)

Configuration

• requirements.txt - Python dependencies
• README.md - Documentation (optional but good practice)

---

❌ DO NOT Upload These Files

Local Development Only

• .env - Contains local secrets (use Spaces Variables instead)
• *.log - Log files
• logs/ - Log directory
• outputs/ - Output directory
• __pycache__/ - Python cache
• .git/ - Git repository

Test Files (Not Needed)

• test_*.py - All test scripts
• check_*.py - Check scripts
• debug_*.py - Debug scripts
• verify_*.py - Verification scripts
• fix_*.py - Fix scripts
• patch_*.py - Patch scripts
• create_sample_*.py - Sample creation

Documentation (Optional)

• *.md files - Helpful but not required for app to run
• You can upload them if you want documentation in your Space

---

📦 Minimal File List (Absolute Minimum)

If you want the smallest deployment, upload only these:

app.py
llm.py
extractors.py
tagging.py
chunking.py
validation.py
reporting.py
dashboard.py
production_logger.py
requirements.txt

Quote extraction will be disabled but everything else will work.

---

📋 Complete File List (Recommended)

Upload all core files plus quote extraction:

app.py
llm.py
extractors.py
tagging.py
chunking.py
validation.py
reporting.py
dashboard.py
production_logger.py
quote_extractor.py
requirements.txt
README.md (optional)

---

🔍 How to Check What's Missing

If you get ModuleNotFoundError: No module named 'xyz', you need to upload xyz.py.

Common missing modules:

• quote_extractor → Upload quote_extractor.py
• production_logger → Upload production_logger.py
• dashboard → Upload dashboard.py

---

📁 Folder Structure on HuggingFace Spaces

Your Space should look like:

your-space/
├── app.py
├── llm.py
├── extractors.py
├── tagging.py
├── chunking.py
├── validation.py
├── reporting.py
├── dashboard.py
├── production_logger.py
├── quote_extractor.py (optional)
├── requirements.txt
└── README.md (optional)

Do NOT create subdirectories - keep all Python files in the root.

---

🚀 Quick Upload Checklist

Before uploading to Spaces:

• app.py - Main file
• All imported modules (llm, extractors, etc.)
• requirements.txt - Dependencies
• Selected GPU hardware in Spaces settings
• No .env file included
• No test/debug files included

---

🔧 Troubleshooting Import Errors

Error: ModuleNotFoundError: No module named 'quote_extractor'

Fixed! This is now optional - app will work without it.

Error: ModuleNotFoundError: No module named 'extractors'

Solution: Upload extractors.py

Error: ModuleNotFoundError: No module named 'production_logger'

Solution: Upload production_logger.py

Error: ModuleNotFoundError: No module named 'transformers'

Solution: Check requirements.txt is uploaded and correct

---

📝 Alternative: Use Git Repository

Instead of manual upload, you can:

1. Create a Git repository with only required files
2. Connect it to your HuggingFace Space
3. Auto-deploy on push

Create .gitignore to exclude:

.env
*.log
logs/
outputs/
__pycache__/
test_*.py
debug_*.py
*.pyc

---

Last Updated

October 2025