requirements.txt

# Core Flask and API
flask==3.0.3
gunicorn==23.0.0  # Production WSGI server
pydantic==2.9.2  # Data validation for schemas
python-dotenv==1.0.1  # Environment variables

# Database and ORM
sqlalchemy==2.0.35  # ORM for models
alembic==1.13.3  # Database migrations
psycopg2-binary==2.9.9  # PostgreSQL driver (adjust for other DBs)

# Async Tasks
celery==5.4.0
redis==5.0.8  # Redis for Celery and caching

# AI and NLP Models
transformers==4.44.2  # Hugging Face for LLaMA, T5, BERT, etc.
torch==2.4.1  # PyTorch for model inference
torchvision==0.19.1  # For CLIP
sentence-transformers==3.1.1  # Sentence-BERT for embeddings
fasttext==0.9.3  # Language detection
doctr==0.8.1  # OCR for document extraction
camelot-py==0.11.0  # Table extraction
opencv-python==4.10.0.84  # Image processing for charts/images
peft==0.12.0  # LoRA for fine-tuning
onnx==1.16.2  # Model optimization
openai==1.50.0  # Optional OpenAI embeddings (requires API key)

# Vector Storage
chromadb==0.5.5  # Chroma for vector search

# AWS Integration
boto3==1.35.24  # AWS SDK for S3 and Sagemaker
sagemaker==2.231.0  # Sagemaker model hosting

# Monitoring and Logging
prometheus-flask-exporter==0.23.1  # Prometheus metrics
structlog==24.4.0  # Structured logging

# Testing
pytest==8.3.3
pytest-cov==5.0.0  # Coverage reports
pytest-asyncio==0.24.0  # Async tests
factory-boy==3.3.1  # Test fixtures

# Development Tools
black==24.8.0  # Code formatting
ruff==0.6.8  # Linting
isort==5.13.2  # Import sorting
mypy==1.11.2  # Type checking

# Security
pyjwt==2.9.0  # JWT for authentication
cryptography==43.0.1  # Secure token handling

# Utilities
requests==2.32.3  # HTTP requests
numpy==1.26.4  # Numerical operations
pandas==2.2.3  # Data handling for tables
pillow==10.4.0  # Image processing