# aangelopoulos/conformal-prediction

Lightweight, useful implementation of conformal prediction on real data.

Repository: https://github.com/aangelopoulos/conformal-prediction
Canonical: https://ross.abutalabs.com/products/conformal-prediction
Homepage: http://people.eecs.berkeley.edu/~angelopoulos/blog/posts/gentle-intro/
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: computer-vision, conformal, conformal-prediction, distribution-shift, natural-language-processing, time-series, time-series-prediction, uncertainty, uncertainty-estimation, uncertainty-quantification, conformal-inference
Last push: 2025-11-14T09:33:36+00:00

## Health v2 (maintenance only)
Score: 56/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 52, release rhythm 35, longevity 100
- inputs: {"age_days": 1713, "days_push": 292, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1081, forks 123 (observed 2026-08-28T04:03:30.817225+00:00)

## What it is
A collection of Jupyter notebooks demonstrating conformal prediction (conformal inference) on real-world machine learning tasks like image classification, regression, and time series. It serves as both a practical tutorial and a template sandbox for developing uncertainty quantification methods without needing to run the underlying models.

## Use cases
- learn conformal prediction with real examples
- build prediction sets with coverage guarantees
- quantify uncertainty in machine learning models
- apply conformalized quantile regression for prediction intervals
- experiment with uncertainty estimation under distribution shift
- prototype new conformal prediction methods

## When to choose
- you want to learn or teach conformal prediction hands-on
- you need statistically guaranteed prediction sets or intervals
- you want notebook templates for uncertainty quantification research

## When to avoid
- you need a production-ready conformal prediction library with an API
- you want a maintained pip-installable package rather than notebooks
- your task is unrelated to statistical uncertainty estimation

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, data-science, nlp, computer-vision
- domain: machine-learning, data-science, tutorials
- platform: python
- tags: conformal-prediction, uncertainty-quantification, jupyter-notebooks, prediction-sets, distribution-shift, time-series-forecasting, statistics

## Member repositories
- aangelopoulos/conformal-prediction (main) score 56

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:30.817225+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 2026-08-30T06:51:19.229201+00:00, confidence not recorded.
  - readme: https://github.com/aangelopoulos/conformal-prediction (fetched 2026-08-28T04:03:30.817225+00:00, sha 4778ea8e89a4)
- Data as of 2026-08-30T08:39:29.467469+00:00.