# JustGlowing/minisom

:red_circle: MiniSom is a minimalistic implementation of the Self Organizing Maps

Repository: https://github.com/JustGlowing/minisom
Canonical: https://ross.abutalabs.com/products/minisom
Language: Python
License: MIT
License Family: permissive
Topics: machine-learning, vector-quantization, som, neural-networks, unsupervised-learning, manifold-learning, self-organizing-map, clustering, outlier-detection, dimensionality-reduction, kohonen
Last push: 2026-06-09T10:41:31+00:00

## Health v2 (maintenance only)
Score: 75/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 86, release rhythm 46, longevity 100
- inputs: {"age_days": 4809, "days_push": 85, "days_rel": 203, "gap_med": 175.0, "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 1601, forks 444 (observed 2026-08-28T04:05:09.627280+00:00)

## What it is
MiniSom is a minimalistic, NumPy-based Python implementation of Self Organizing Maps (SOM), a type of neural network for unsupervised learning, with optional Numba JIT acceleration. It is designed to be simple and readable so researchers can extend it and students can quickly understand the algorithm.

## Use cases
- cluster high-dimensional data with a self-organizing map
- reduce dimensionality of my dataset for visualization
- perform vector quantization on numpy data
- detect outliers in unlabeled data
- learn how SOMs work by reading a simple implementation
- visualize nonlinear relationships in high-dimensional data on a 2D map

## When to choose
- you need a lightweight, dependency-minimal SOM implementation in Python
- you want readable source code to study or extend the SOM algorithm
- you work with NumPy arrays and want easy integration into a scientific Python stack
- you want optional JIT speedups via Numba

## When to avoid
- you need a full deep learning framework with GPU training and rich model tooling
- you need supervised learning or other neural network architectures
- you need highly optimized, distributed training for very large datasets

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, data-science
- domain: machine-learning, data-science
- platform: python
- tags: self-organizing-map, som, unsupervised-learning, clustering, dimensionality-reduction, vector-quantization, numpy, numba

## Member repositories
- JustGlowing/minisom (main) score 75

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:09.627280+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 2026-08-30T03:52:39.954586+00:00, confidence not recorded.
  - readme: https://github.com/JustGlowing/minisom (fetched 2026-08-28T04:05:09.627280+00:00, sha 0da3c02c3422)
  - registry_pypi: https://pypi.org/pypi/minisom/json (fetched 2026-08-29T11:24:15.655782+00:00, sha 88a109a4e993)
- Data as of 2026-08-30T08:39:29.467469+00:00.