# CPU-Only Requirements (No GPU/CUDA needed) # Core Configuration python-dotenv>=1.0.0 pydantic>=2.5.0 pydantic-settings>=2.1.0 # Web Scraping & Data Collection beautifulsoup4>=4.12.0 selenium>=4.16.0 playwright>=1.40.0 httpx>=0.25.0 aiohttp>=3.9.0 youtube-transcript-api>=0.6.2 # AI & LLM (CPU versions) openai>=1.6.0 anthropic>=0.8.0 langchain>=0.1.0 langchain-openai>=0.0.2 langchain-community>=0.0.10 langgraph>=0.0.20 langchain-core>=0.1.0 tiktoken>=0.5.0 # Databricks Agent Framework (Mosaic AI) databricks-sdk>=0.18.0 mlflow>=2.10.0 databricks-agents>=0.1.0 databricks-vectorsearch>=0.22.0 delta-spark>=3.0.0 # PySpark is large but needed for Delta Lake - consider removing if not using Databricks pyspark>=3.5.0 pandas>=2.1.0 polars>=0.20.0 pyarrow>=14.0.0 # NLP - CPU-only versions (much lighter) # Note: Omitting spacy, transformers, sentence-transformers, torch to save ~2GB # You can add them back if needed: torch --index-url https://download.pytorch.org/whl/cpu nltk>=3.8.0 # Database & Storage sqlalchemy>=2.0.0 chromadb>=0.4.0 qdrant-client>=1.7.0 # API Framework fastapi>=0.109.0 uvicorn[standard]>=0.25.0 # Removed fastapi-cache2 - not strictly required python-multipart>=0.0.6 # Visualization plotly>=5.18.0 folium>=0.15.0 matplotlib>=3.8.0 seaborn>=0.13.0 # Utilities python-dateutil>=2.8.0 pytz>=2023.3 tenacity>=8.2.0 loguru>=0.7.0 click>=8.1.0 tqdm>=4.66.0 # Testing pytest>=7.4.0 pytest-asyncio>=0.21.0 pytest-cov>=4.1.0 pytest-mock>=3.12.0 pytest-httpx>=0.27.0 # Development black>=23.0.0 ruff>=0.1.0 mypy>=1.8.0 pre-commit>=3.6.0