# dmlc/minpy

NumPy interface with mixed backend execution

Repository: https://github.com/dmlc/minpy
Canonical: https://ross.abutalabs.com/products/minpy
Homepage: https://minpy.readthedocs.io/en/latest/
Language: Python
License: NOASSERTION
License Family: other
Archived: true
Last push: 2018-02-19T22:22:39+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3954, "days_push": 3117, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1094, forks 104 (observed 2026-08-28T04:03:34.031590+00:00)

## What it is
MinPy is a Python library providing a pure NumPy interface backed by MXNet, automatically dispatching operators to GPU when available and falling back to NumPy on CPU. It includes automatic gradient generation via Autograd and seamless MXNet symbol integration.

## Use cases
- run numpy code on gpu transparently
- prototype deep learning algorithms with numpy syntax
- get automatic gradients for numpy-like code
- mix mxnet symbolic ops with imperative numpy code
- graceful cpu fallback for missing gpu operators

## When to choose
- you are maintaining legacy code already written against minpy
- you want a historical example of mixed imperative/symbolic execution design

## When to avoid
- starting any new deep learning project - use MXNet Gluon, PyTorch, or JAX instead
- you need maintained software with recent releases
- you need production-grade performance or support

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, gpu-computing, math
- domain: deep-learning, machine-learning, gpu-computing
- platform: python
- tags: numpy, mxnet, autograd, deprecated, mixed-execution, gpu

## Member repositories
- dmlc/minpy (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:34.031590+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 2026-08-30T06:47:36.202234+00:00, confidence not recorded.
  - readme: https://github.com/dmlc/minpy (fetched 2026-08-28T04:03:34.031590+00:00, sha 1fd6e47a2677)
  - registry_pypi: https://pypi.org/pypi/minpy/json (fetched 2026-08-29T12:51:08.015573+00:00, sha e50af739e9f9)
- Data as of 2026-08-30T08:39:29.467469+00:00.