# iskandr/fancyimpute

Multivariate imputation and matrix completion algorithms implemented in Python

Repository: https://github.com/iskandr/fancyimpute
Canonical: https://ross.abutalabs.com/products/fancyimpute
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2023-10-25T17:26:07+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3954, "days_push": 1043, "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 1281, forks 178 (observed 2026-08-28T04:04:13.847900+00:00)

## What it is
A Python library offering a variety of multivariate imputation and matrix completion algorithms such as KNN, SoftImpute, and NuclearNormMinimization for filling in missing values in data matrices. It is now in bare maintenance mode, with IterativeImputer delegated to scikit-learn.

## Use cases
- fill in missing values in a dataset with NaN entries
- complete a partially observed matrix using low-rank methods
- impute missing features using k-nearest neighbors
- compare imputation methods by mean squared error
- handle missing data before training a machine learning model

## When to choose
- you need a ready-made set of matrix completion and imputation algorithms in Python
- your data has missing numeric entries and you want alternatives to simple mean/median filling
- you want KNN, SoftImpute, or nuclear norm minimization implementations without writing them yourself

## When to avoid
- you need actively developed imputation features or new algorithms
- you only need IterativeImputer, which is better obtained directly from scikit-learn
- you need guaranteed long-term support, since the project is in bare maintenance mode

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, data-science
- domain: data-science, machine-learning
- platform: python
- tags: missing-data, imputation, matrix-completion, numpy, scikit-learn

## Member repositories
- iskandr/fancyimpute (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:13.847900+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 2026-08-30T05:02:32.320437+00:00, confidence not recorded.
  - readme: https://github.com/iskandr/fancyimpute (fetched 2026-08-28T04:04:13.847900+00:00, sha 973f79f40fa2)
  - registry_pypi: https://pypi.org/pypi/fancyimpute/json (fetched 2026-08-29T12:13:05.994754+00:00, sha 99e92ad904d9)
- Data as of 2026-08-30T08:39:29.467469+00:00.