The idea is simple: remove EDA boilerplate and make data inspection something you can run in one command, directly from the terminal or CI.

fasteda data.csv --fun

Or generate a standalone interactive HTML report:

fasteda data.csv --html report.html

What it includes

Automatic statistics + detailed quantiles

Multiple correlation methods (Pearson, Spearman, Kendall)

Outlier detection (IQR + Z-score)

Automated data quality alerts (duplicates, multicollinearity, class imbalance, high cardinality, missing values, mixed types)

Interactive Plotly-based HTML reports

CSV / Excel / JSON / Parquet support

Python API + plugin system

Why I built it

Existing tools are powerful, but often heavy, slow on medium datasets, or not CLI-first. SpeedyEDA is designed for quick inspection, scripts, and daily dev workflows.

Links

PyPI: https://pypi.org/project/speedyeda/

GitHub:https://github.com/Dawaman43/fasteda

Would love feedback—especially on performance, missing checks, and real-world edge cases.

Thanks!