Pandas Profiling

Create an Exploratory Data Analysis Report with Minimal Effort ( One Line of Code )

Install Pandas Profiling

pip install pandas-profiling

Report Structure

The Pandas Profiling Report :

Overview :

• Basic Information about Data.

• Number of Columns.

• Number of Rows.

• Data Size.

• Percentage of Missing Values.

• Data Type.

Reproduction :

• Information about Report Creation.

Warnings :

• Warnings while Creating | Producing the Report.

Variables :

• Detailed Analysis of Each Variables.

• A Histogram for Continuous Variable.

• A Bar Chart for Categorical Variable.

Interactions :

• Bivariate Relationship between Numerical Variables.

Correlations :

• The Different Types of Correlations.

• Numerical Variables : Pearson’s Correlation | Spearman’s Correlation | Kendall’s Correlation | Phik Correlation.

• Categorical Variables : Cramer’s V Correlation.

Missing Values :

• Missing Values in the Data Set.

Samples :

• First and Last 10 Rows of the Data Set.

Duplicate Rows :

• If there is are any Duplicate Rows in the Data Set ?

Disadvantage :

• If the Data Set is Very Large, It will take a Long Time to Create a Report (Could be Hours.)

• We have a Basic Exploratory Data Analysis using a Profiling Package

• Definitely not a Complete Exploration.

Heart.html gets Generated when we Create a Profile Report