fg-data-profiling

Why we included this project

For anyone doing exploratory analysis in Jupyter, fg-data-profiling (the renamed ydata-profiling) turns a raw DataFrame into a readable report with a single call. It produces per-column statistics, type inference, missing-value and duplicate checks, correlation matrices, and alerts for common quality problems such as skewness or constant values. The report exports as HTML for stakeholders or JSON for downstream tooling, so the same output serves people and machines. Because it handles Pandas and Spark DataFrames as well as time-series and text, it is a fast way to sanity-check incoming data before modeling starts.