# SciSharp/NumSharp

High Performance Computation for N-D Tensors in .NET, similar API to NumPy.

Repository: https://github.com/SciSharp/NumSharp
Canonical: https://ross.abutalabs.com/products/numsharp
Homepage: https://scisharp.github.io/NumSharp/
Language: C#
License: Apache-2.0
License Family: permissive
Topics: numpy, machine-learning, numsharp, pandas, ai
Last push: 2026-08-24T19:31:47+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 78, longevity 100
- inputs: {"age_days": 2885, "days_push": 9, "days_rel": 66, "gap_med": 37, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1477, forks 205 (observed 2026-08-28T04:04:50.300134+00:00)

## What it is
NumSharp is a .NET library providing NumPy-shaped N-dimensional arrays with broadcasting, slicing views, dtype-aware math, and runtime-generated SIMD kernels. It lets C# and F# code use a NumPy-like programming model without embedding CPython.

## Use cases
- do numpy-style array math in C#
- n-dimensional tensor operations in .NET
- numerical computing without Python
- broadcasting and slicing arrays in F#
- build machine learning infrastructure in .NET
- replace Python numpy dependency in a C# pipeline

## When to choose
- you want NumPy API parity in pure .NET code
- you need high-performance unmanaged array storage with SIMD fast paths
- your project targets NumPy 2.x behavior as source of truth

## When to avoid
- you need the full scientific Python ecosystem (SciPy, pandas) rather than arrays
- you need GPU acceleration out of the box
- you require complete NumPy coverage - some areas are still being tightened

## Facets
- artifact type: library
- maturity: active
- function: math, machine-learning, data-science
- domain: data-science, machine-learning, developer-tools
- platform: dotnet, cross-platform
- tags: numpy, ndarray, numerical-computing, scientific-computing, simd, csharp, fsharp

## Member repositories
- SciSharp/NumSharp (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:50.300134+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 2026-08-30T04:34:22.958420+00:00, confidence not recorded.
  - readme: https://github.com/SciSharp/NumSharp (fetched 2026-08-28T04:04:50.300134+00:00, sha e2266accca30)
  - homepage: https://scisharp.github.io/NumSharp/ (fetched 2026-08-29T11:41:22.972621+00:00, sha ca27508e1006)
- Data as of 2026-08-30T08:39:29.467469+00:00.