Data-Centric-AI-Community/fg-data-profiling

What is it?

Key features

• 01One-line generation of a full profiling report from a pandas DataFrame.
• 02Automatic type inference and univariate statistics with distribution plots.
• 03Correlations, missing-data analysis and duplicate row detection.
• 04Time-series analysis including ACF and PACF plots and seasonality.
• 05Alerts list highlighting data quality issues such as skewness and constant values.
• 06Export to HTML, JSON, Jupyter widgets, or a command-line interface.

Use cases

• Quickly understanding an unfamiliar dataset before modeling.
• Checking data quality issues in a data pipeline via JSON output.
• Comparing two versions of a dataset, such as before and after preprocessing.
• Sharing a profiling report as a standalone HTML file with stakeholders.

Quick start

1. 1

   Install the package with pip

   pip install fg-data-profiling
2. 2

   Import pandas and the ProfileReport class, then build a DataFrame

3. 3

   Generate the standard profiling report

   profile = ProfileReport(df, title="Profiling Report")
4. 4

   Export the report to an HTML file

   profile.to_file("your_report.html")

Things to know

• !The old ydata-profiling package is no longer maintained, so existing users must migrate to the renamed package and update their imports.
• !Spark support is present but the README notes it is still a work in progress that needs more contributors.

In collections

Topics

FAQ

What is fg-data-profiling?

Python library that generates a one-line exploratory data analysis and data quality report for pandas and Spark DataFrames, exportable as HTML or JSON. fg-data-profiling is a Python package that produces an extended exploratory data analysis of a DataFrame with a single call, similar in spirit to pandas' describe() but much more detailed. The generated report covers type inference, univariate statistics and distributions, correlations, missing data, duplicate rows, time-series properties such as autocorrelation and seasonality, and text, file and image columns.

Is fg-data-profiling free to use? What license does it have?

fg-data-profiling is open source under the MIT license. Check the license terms before using it in commercial products.

Is fg-data-profiling still maintained?

Actively maintained. The last commit to fg-data-profiling was on Nov 5, 2026.

What can I use fg-data-profiling for?

Quickly understanding an unfamiliar dataset before modeling. Checking data quality issues in a data pipeline via JSON output. Comparing two versions of a dataset, such as before and after preprocessing. Sharing a profiling report as a standalone HTML file with stakeholders.

How do I install fg-data-profiling?

Install the package with pip: pip install fg-data-profiling Import pandas and the ProfileReport class, then build a DataFrame Generate the standard profiling report: profile = ProfileReport(df, title="Profiling Report") Export the report to an HTML file: profile.to_file("your_report.html")

Where can I get fg-data-profiling?

The source code and installation instructions are on GitHub: https://github.com/Data-Centric-AI-Community/fg-data-profiling