OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Evaluation & Observability

Conformal Prediction

Conformal Prediction is an open-source Jupyter Notebook repository that demonstrates conformal inference on real-world machine-learning problems. Its examples apply uncertainty methods to precomputed model outputs, producing prediction sets, intervals, outlier flags and abstentions across image, text, regression and time-series tasks.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Conformal Prediction — OpenViST project profile in AI Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to examine conformal prediction on concrete machine-learning tasks without downloading or running the original models. Choose the relevant notebook to explore its method and expected output.

Who it is for

  • Machine-learning practitioners exploring uncertainty estimates for predictions
  • Researchers and students studying conformal prediction through task-based notebook examples
  • Notebook users adapting existing examples to experiment with conformal methods

Documented uses

  • Explore prediction sets for image classification and multilabel classification.
  • Study prediction intervals for regression and time-series forecasting.
  • Examine conformal outlier detection, image segmentation and selective classification examples.
  • Use example notebooks as templates for developing or testing conformal prediction methods.
Before you choose

Requirements & limits

The repository is public and released under the MIT license. It documents Google Colab links in the top cell of each notebook and local Conda setup.

  • Notebook examples download sample data and precomputed model outputs automatically.
  • Local setup requires Conda and 1.5 GB of storage for downloaded data.
  • The original machine-learning models are not run by the notebooks.
  • Examples include image classification, medical expenditure regression, multilabel classification, text outlier detection, tumor segmentation, time-series forecasting under distribution shift and selective classification.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the notebooks locally

Prepare a local Conda environment and Jupyter notebook for running the repository's examples.

Before you start

  • The repository and its notebook files
  • Conda
  • 1.5 GB of disk space for automatically downloaded data

Steps

  1. Create the environment with conda env create -f environment.yml.

  2. Open a notebook in Jupyter and activate the conformal environment if a dependency error occurs.

  3. Run all notebook cells; required data and model outputs are downloaded automatically.

Expected outcome

A locally prepared notebook that can run an example using downloaded sample data and precomputed model outputs.

Guide limitations

  • The documented storage requirement is 1.5 GB.
  • The notebooks use precomputed model outputs and do not run the original models.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
aangelopoulos/conformal-prediction · Official homepage title · retained in its source language
How we review sourcesSuggest a correction