# raphaelvallat/pingouin

Statistical package in Python based on Pandas

Repository: https://github.com/raphaelvallat/pingouin
Canonical: https://ross.abutalabs.com/products/pingouin
Homepage: https://pingouin-stats.org/
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: pandas, statistics, anova, effect-size, multiple-comparisons, cohens-d, bayesian-statistics, ttest, correlations, circular-statistics, statistical-tests, statistical-methods
Last push: 2026-04-05T08:28:47+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 75, release rhythm 45, longevity 100
- inputs: {"age_days": 3077, "days_push": 150, "days_rel": 158, "gap_med": 285.0, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1928, forks 167 (observed 2026-08-28T04:05:55.788315+00:00)

## What it is
Pingouin is an open-source Python 3 statistical package built on Pandas and NumPy that provides simple yet exhaustive statistical tests. It returns rich outputs (effect sizes, confidence intervals, Bayes Factors) from common tests like t-tests and ANOVAs, which low-level libraries like SciPy omit.

## Use cases
- run t-tests and anovas in python
- compute effect sizes like cohen's d
- calculate bayes factors for statistical tests
- perform repeated measures anova on a dataframe
- compute partial and robust correlations
- do power analysis for an experiment
- run pairwise post-hoc tests with p-value correction
- compute intraclass correlation for reliability

## When to choose
- you want detailed statistical test output (effect size, CI, power, Bayes Factor) from a single function call
- you work with Pandas DataFrames and want stats functions that integrate with them
- you need ANOVAs, post-hoc tests, or repeated-measures correlations in Python
- you find scipy.stats too low-level and statsmodels too complex

## When to avoid
- you need advanced modeling like GLMs, time-series analysis, or R-style formula syntax (use statsmodels)
- you only need raw T- and p-values with minimal dependencies (use scipy.stats)
- your project requires a permissive license (Pingouin is GPL-3.0)

## Facets
- artifact type: library
- maturity: active
- function: data-science, math, data-visualization
- domain: data-science, analytics
- platform: python, cross-platform
- tags: statistics, anova, t-test, effect-size, bayesian-statistics, pandas, correlation, power-analysis, circular-statistics

## Member repositories
- raphaelvallat/pingouin (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:55.788315+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:09:08.103477+00:00, confidence not recorded.
  - readme: https://github.com/raphaelvallat/pingouin (fetched 2026-08-28T04:05:55.788315+00:00, sha 11db71d02532)
  - homepage: https://pingouin-stats.org/ (fetched 2026-08-29T10:48:38.076110+00:00, sha 22908b117163)
  - registry_pypi: https://pypi.org/pypi/pingouin/json (fetched 2026-08-29T10:48:38.086767+00:00, sha 34a73970b9f9)
  - site_page: https://pingouin-stats.org/faq.html (fetched 2026-08-29T10:48:38.078854+00:00, sha 5a5738e86770)
  - site_page: https://pingouin-stats.org/changelog.html (fetched 2026-08-29T10:48:38.081045+00:00, sha ddc09e6b0f80)
  - site_page: https://pingouin-stats.org/api.html (fetched 2026-08-29T10:48:38.084925+00:00, sha f6b2071ebd80)
- Data as of 2026-08-30T08:39:29.467469+00:00.