---
id: arraykit
version: "1.12.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# arraykit — Array utilities for StaticFrame

License: permissive · Maintenance: active · Downloads: 116.2K/mo

## What it is and what it does

ArrayKit is a C-extension library that implements array utilities for NumPy, designed to accelerate performance-critical operations. It was created to support StaticFrame's internal needs but is published as a standalone package. The library provides functions for array creation, transformation, grouping, mapping, and specialized operations like factorization and block indexing.

The package depends only on NumPy and requires Python 3.10 or later. It is actively maintained with recent releases adding new functions like map_object() and prepare_iter_for_array(). Wheels are pre-built for multiple platforms and Python versions, including free-threaded variants for Python 3.13+, reducing installation friction on supported systems.

Use it for:

- Accelerate array grouping and reduction operations in data processing pipelines that require better performance than pure NumPy.
- Use TriMap for optimized join operations when working with multiple arrays that need to be aligned or merged.
- Parse delimited text files into typed NumPy arrays with automatic type inference via delimited_to_arrays().
- Perform factorization on array data to identify unique values and their indices efficiently.
- Build block-indexed structures for managing non-contiguous column selections in tabular data.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

ArrayKit provides C-extension implementations of array creation and transformation utilities optimized for NumPy, primarily used internally by StaticFrame for performance-critical operations.

Yes, if you are a StaticFrame user or need the specific array utilities it provides. The package is actively maintained, has no known vulnerabilities, and offers MIT licensing. However, if you are not using StaticFrame and do not need its specialized functions (TriMap, factorize, group_reduce, etc.), standard NumPy may be sufficient. Install friction is moderate due to compiled wheels, but pre-built binaries are available for common platforms.

## Install

pip install arraykit
uv add arraykit
poetry add arraykit

## Installing arraykit

Before you install: Medium install friction due to compiled wheels for multiple platforms and Python versions. Active maintenance with a release 8 days ago; the package is actively developed and supports current Python versions including free-threaded builds.

License in practice: MIT license permits unrestricted use, modification, and distribution in both open-source and commercial projects with minimal restrictions.

Quickstart:

import numpy as np
import arraykit as ak

arr = np.array([1, 2, 3])
result = ak.nonzero_1d(arr)

Requires Python>=3.10 and numpy>=1.24.3; compiled wheels available for macOS (x86_64, arm64), Linux (x86_64, musllinux), and Windows (32/64-bit).

Verify before relying:

- Whether arraykit is intended for direct use or primarily as a StaticFrame internal dependency
- Performance improvements over pure NumPy equivalents for specific operations
- API stability guarantees or deprecation policy for exported functions

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 116.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

numpy array utilities, array transformation performance, numpy c extensions, array creation helpers, staticframe dependencies, numpy-extension, performance-oriented

[View on SkillFed](https://skillfed.io/packages/arraykit) · [View on PyPI](https://pypi.org/project/arraykit/)