fdth

Frequency distribution tables, histograms, and polygons for numerical and categorical data in Python.

A feature-complete port of the fdth R package (CRAN).

pip install fdth

Until the package is published on PyPI, install from GitHub (source):

pip install git+https://github.com/jcfaria/fdth-python-fork_2026.1.git

Or install the built wheel from this repository’s dist/ folder (project exception — artifacts are versioned on purpose):

pip install https://github.com/jcfaria/fdth-python-fork_2026.1/raw/main/dist/fdth-1.0.0-py3-none-any.whl

Release notes: NEWS.md.

Features

• Frequency distribution tables (FDT) from lists, NumPy arrays, and pandas objects
• Numerical and categorical variables, plus multi-column / grouped tables
• Histograms and frequency polygons (absolute, relative, cumulative)
• Summary measures from grouped data (mean, median, mode, quantiles, variance, …)
• Grouping with multiple keys via by= (beyond the original R single-key limit)

Requires Python 3.10+. Depends on pandas, numpy, and matplotlib.

Quick start

from fdth import fdt

data = [1, 5, 8, 3, 12, 7, 4, 9]
table = fdt(data)
print(table)
table.plot()

fdt() detects the data type and returns NumericalFDT, CategoricalFDT, or MultipleFDT as appropriate.

Grouped example

Build FDTs for every column, split by several grouping variables, and plot:

import pandas as pd
from fdth import fdt

df = pd.DataFrame({
    "Height": [170, 175, 180, 165, 172, 168, 175, 170, 160, 155],
    "Weight": [70, 75, 80, 65, 72, 68, 75, 70, 60, 55],
    "Age": [30, 20, 30, 18, 30, 27, 24, 50, 60, 9],
    "Sex": ["M", "I", "M", "I", "M", "F", "I", "F", "F", "F"],
    "Region": [
        "Northeast", "Southeast", "Northeast", "Southeast", "North",
        "North", "Southeast", "Northeast", "North", "Southeast",
    ],
    "MaritalStatus": [
        "Married", "Single", "Single", "Married", "Single",
        "Single", "Married", "Married", "Single", "Married",
    ],
})

mfdt = fdt(df, by=["Sex", "MaritalStatus", "Region"])
mfdt.plot(numeric_type="fh", categorical_type="fb")

Page 1

Page 2

More notebooks live under examples/python.

Development

git clone https://github.com/jcfaria/fdth-python-fork_2026.1.git
cd fdth-python-fork_2026.1

python -m venv venv
# Linux / macOS:  source venv/bin/activate
# Windows CMD:    venv\Scripts\activate.bat
# Windows PS:     venv\Scripts\Activate.ps1

pip install -e ".[dev]"

A short Git tutorial in Portuguese is available in HelpGit.md.

Background

This Python port grew through collective work in the Probability and Statistics course (Computer Science, UESC) across semesters 2025.1, 2025.2, and 2026.1. Earlier ports existed before that; the object-oriented design solidified in 2025. The package aims for full coverage of the original R fdth functionality.

Credits

Original R package — jcfaria/fdth

• José Cláudio Faria
• Ivan Bezerra Allaman
• Jakson Alves de Aquino

Initial Python port — yuriccosta/fdth-python

• Emyle Silva
• Lucas Gabriel Ferreira
• Yuri Coutinho Costa
• Maria Clara

2025.1 — Gabriel Galdino, Luciene Mª Torquato C. Batista, Stella Ribas, Thainá Guimarães, Yohanan Santana

2025.2 — Alex Amaral dos Santos, Isaque Silva Passos Ribeiro, Kaiala de Jesus Santos, Olinoedson Silva Sena

2026.1 — Ariel Mariano Vieira, Eduardo Ferreira Diniz da Silva, Pedro Lucas do Nascimento de Oliveira

Author / Maintainer

Faria, J. C.
Universidade Estadual de Santa Cruz — UESC
Departamento de Ciências Exatas — DCEX
Ilhéus — Bahia — Brazil

• Email: joseclaudio.faria@gmail.com
• GitHub: jcfaria
• R package: GitHub · CRAN

License

This package is free software under the GNU General Public License, version 2 (GPL-2.0).

See LICENSE for the full text.