python_data_stack

What problem does it solve?

This Skill provides a comprehensive toolkit for efficient and robust data analysis, modeling, and pipeline construction using Python's most powerful libraries, enabling rigorous statistical inference and scalable data processing.

Core Features & Use Cases

• Statistical Modeling: Fit OLS, GLM, ARIMA, and other advanced statistical models with detailed diagnostics.
• Data Processing: Leverage Polars for high-performance data manipulation on large datasets and DuckDB for in-process SQL analytics.
• Visualization & Profiling: Create informative visualizations with Altair/Plotnine and generate data quality reports with ydata-profiling.
• Pipeline Orchestration: Build reliable ETL pipelines using Prefect or Dagster.
• Use Case: Analyze a multi-gigabyte CSV dataset by loading it into Polars, performing complex SQL queries with DuckDB, generating statistical models using Statsmodels, visualizing results with Altair, and orchestrating the entire process with Prefect.

Quick Start

Use the python_data_stack skill to analyze the 'sales_data.parquet' file using Polars and DuckDB, then generate a statistical summary of the 'revenue' column.