A Python implementation of the summarytools R package

This is going to be a slow developping project because 1) I'm a bit rusty in Python and 2) I don't have much free time. So if you'd like to contribute by all means do!

To see what summarytools is all about, see summarytools: An R Package For Descriptive Statistics

How to install

Just copy the summarytools.py file in your Python/Lib/site-packages folder.

Frequency tables with freq

First, a quick setup

import pandas as pd
from summarytools import freq

Import a sample dataset:

link = "https://raw.githubusercontent.com/dcomtois/summarytools-python/master/data/tobacco.csv"
tobacco = pd.read_csv(link)

Bare-bones example

freq(tobacco.age_gr)

         Freq    % Valid    % Valid Cum.    % Total    % Total Cum.
-----  ------  ---------  --------------  ---------  --------------
18-34     258      26.46           26.46      25.80           25.80
35-50     241      24.72           51.18      24.10           49.90
51-70     317      32.51           83.69      31.70           81.60
71 +      159      16.31          100.00      15.90           97.50
NaN        25                                  2.50          100.00
Total    1000     100.00          100.00     100.00          100.00

The returned object is a pandas Dataframe, and it is displayed using the tabulate library.

Markdown formatting

When switching format to 'pipe', we get pretty formatted markdown tables where mardown is supported, such as here on GitHub.

freq(tobacco.gender, format = 'pipe')

|       |   Freq |   % Valid |   % Valid Cum. |   % Total |   % Total Cum. |
|:------|-------:|----------:|---------------:|----------:|---------------:|
| F     |    489 |     50.00 |          50.00 |     48.90 |          48.90 |
| M     |    489 |     50.00 |         100.00 |     48.90 |          97.80 |
| NaN   |     22 |           |                |      2.20 |         100.00 |
| Total |    978 |    100.00 |         100.00 |    100.00 |         100.00 |

Fancier console display

When working in a regular Python console or in IPython, the 'fancy_grid' format from tabular produces good results:

freq(tobacco.gender, format = 'fancy_grid')

╒═══════╤════════╤═══════════╤════════════════╤═══════════╤════════════════╕
│       │   Freq │   % Valid │   % Valid Cum. │   % Total │   % Total Cum. │
╞═══════╪════════╪═══════════╪════════════════╪═══════════╪════════════════╡
│ F     │    489 │     50.00 │          50.00 │     48.90 │          48.90 │
├───────┼────────┼───────────┼────────────────┼───────────┼────────────────┤
│ M     │    489 │     50.00 │         100.00 │     48.90 │          97.80 │
├───────┼────────┼───────────┼────────────────┼───────────┼────────────────┤
│ NaN   │     22 │           │                │      2.20 │         100.00 │
├───────┼────────┼───────────┼────────────────┼───────────┼────────────────┤
│ Total │   1000 │    100.00 │         100.00 │    100.00 │         100.00 │
╘═══════╧════════╧═══════════╧════════════════╧═══════════╧════════════════╛

Parameters

For now, only a few parameters are implemented. We'll store a frequency table to show how we can tweak the display after its creation, using the print() method.

ft = freq(tobacco.gender, format = 'pipe')

Omit the Total row

ft.print(totals=False)

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
F	489	50.00	50.00	48.90	48.90
M	489	50.00	100.00	48.90	97.80
NaN	22			2.20	100.00

Omit missing data reporting

ft.print(nans=False)

	Freq	% Valid	% Valid Cum.
F	489	50.00	50.00
M	489	50.00	100.00
Total	978	100.00	100.00

Control number of decimals to show

ft.print(digits=1)

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
F	489	50.0	50.0	48.9	48.9
M	489	50.0	100.0	48.9	97.8
NaN	22			2.2	100.0
Total	978	100.0	100.0	100.0	100.0

More to come... In the meantime feel free to chip in and contribute!