DataFrame Summary Tools in Jupyter Notebook

This is python version of summarytools, which is used to generate standardized and comprehensive summary of dataframe in Jupyter Notebooks.

The idea is originated from the summarytools R package (https://github.com/dcomtois/summarytools).

• Only dfSummary, freq, and ctable() functions are made available for now
• Added two html widgets to avoid displaying lengthy content
  • collapsible summary
  • tabbed summary

Installation

pip install summarytools

Dependencies

1. python 3.6+
2. pandas >= 1.4.0

Quick Start

the quick-start notebook is available in here or

out-of-box dfSummary, freq and ctable functions will generate a HTML based data frame summary.

import pandas as pd
from summarytools import dfSummary
titanic = pd.read_csv('./data/titanic.csv')
dfSummary(titanic)

collapsible summary

import pandas as pd
from summarytools import dfSummary
titanic = pd.read_csv('./data/titanic.csv')
dfSummary(titanic, is_collapsible = True)

tabbed summary

import pandas as pd
from summarytools import dfSummary, tabset
titanic = pd.read_csv('./data/titanic.csv')
vaccine = pd.read_csv('./data/country_vaccinations.csv')
vaccine['date'] = pd.to_datetime(vaccine['date'])

tabset({
    'titanic': dfSummary(titanic).render(),
    'vaccine': dfSummary(vaccine).render()})

Export notebook as HTML

when export jupyter notebook to HTML, make sure Export Embedded HTML extension is installed and enabled.

Using the following bash command to retain the data frame summary in exported HTML.

jupyter nbconvert --to html_embed path/of/your/notebook.ipynb

Stargazers over time