# uncertainty-toolbox/uncertainty-toolbox

Uncertainty Toolbox: a Python toolbox for predictive uncertainty quantification, calibration, metrics, and visualization

Repository: https://github.com/uncertainty-toolbox/uncertainty-toolbox
Canonical: https://ross.abutalabs.com/products/uncertainty-toolbox
Homepage: https://uncertainty-toolbox.github.io
Language: Python
License: MIT
License Family: permissive
Topics: uncertainty, uncertainty-quantification, uncertainty-estimation, uncertainty-metrics, calibration, sharpness, uncertainty-toolbox, metrics, visualizations, visualization, recalibration, predictive-uncertainty, scoring-rules, uncertainty-calibration, toolbox, bayesian-neural-networks, bayesian-deep-learning
Last push: 2025-03-05T00:46:24+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 9, release rhythm 8, longevity 100
- inputs: {"age_days": 2188, "days_push": 547, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2012, forks 152 (observed 2026-08-28T04:06:05.179198+00:00)

## What it is
A Python library for predictive uncertainty quantification, providing metrics, visualizations, and recalibration procedures for regression models. It also includes a glossary and curated paper list on uncertainty estimation.

## Use cases
- evaluate calibration of model uncertainty estimates
- compare predictive uncertainties across models with standard metrics
- visualize confidence intervals and calibration curves
- recalibrate regression predictions to improve uncertainty quality
- compute scoring rules for probabilistic forecasts
- learn about uncertainty quantification terminology and papers

## When to choose
- you need standardized metrics for regression uncertainty and calibration
- you want quick recalibration of predicted means and standard deviations
- you need plots and benchmarks for probabilistic predictions

## When to avoid
- you need uncertainty quantification for classification tasks (toolbox focuses on regression)
- you want full Bayesian deep learning training methods rather than evaluation
- you need production-scale distributed metric computation

## Facets
- artifact type: library
- maturity: active
- function: data-science, machine-learning, data-visualization, benchmarking
- domain: machine-learning, data-science
- platform: python, cross-platform
- tags: uncertainty-quantification, calibration, predictive-uncertainty, scoring-rules, recalibration, bayesian-deep-learning, regression, algorithms

## Member repositories
- uncertainty-toolbox/uncertainty-toolbox (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:05.179198+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 2026-08-30T03:01:03.337435+00:00, confidence not recorded.
  - readme: https://github.com/uncertainty-toolbox/uncertainty-toolbox (fetched 2026-08-28T04:06:05.179198+00:00, sha fe61b87f82e0)
  - homepage: https://uncertainty-toolbox.github.io (fetched 2026-08-29T10:41:02.788058+00:00, sha 2523b18385c0)
  - registry_pypi: https://pypi.org/pypi/uncertainty-toolbox/json (fetched 2026-08-29T10:41:02.797009+00:00, sha 587521893cfc)
- Data as of 2026-08-30T08:39:29.467469+00:00.