numpy を Homebrew でインストール

プロジェクトリンク

• ホームページ
• リポジトリ

コマンドとエイリアス

• f2py
• numpy-config

プロジェクトの歴史

The official NumPy about page says NumPy was created in 2005, building on Numeric and Numarray. The old SciPy history records the project as a reunion: Travis Oliphant wanted to bring the split Numeric/numarray community back to one array package, refactored Numeric to absorb numarray's features, and after naming discussion the multidimensional array project became NumPy.

採用の歴史

NumPy became the array substrate for the Python scientific stack. The 2020 Nature paper 'Array programming with NumPy' describes NumPy as a 2005 unification of Numarray's features with Numeric's small-array performance and C API, and says that by 2020 it underpinned almost every Python library doing scientific or numerical computation, including SciPy, Matplotlib, pandas, scikit-learn, and scikit-image.

使われ方

Users install NumPy directly for dense arrays, vectorized math, broadcasting, random sampling, FFTs, statistics, basic linear algebra, and data interchange with other Python libraries. Package maintainers also care about the compiled-code boundary: `f2py` is distributed with NumPy as both `numpy.f2py` and a standalone command-line tool, making old Fortran routines callable from Python, while the C API supports extensions and downstream projects.

パッケージ好きにとっての重要性

For package ecosystems, NumPy is not just a library; it is an ABI, build, and compatibility gravity well. Python packages with native extensions often pin or test against NumPy versions, scientific Linux/macOS distributions ship optimized BLAS/LAPACK stacks underneath it, and Homebrew's formula page showed 333,990 installs over 365 days on July 1, 2026 despite Python users often installing it through pip or conda instead.

タイムライン

• 1990s-2000s: Numeric and then Numarray serve overlapping array-computing communities.
• 2005: NumPy is created by combining Numeric and Numarray ideas into one project.
• 2020: Nature publishes 'Array programming with NumPy', documenting its role across scientific Python.

Related projects

• Numeric
• Numarray
• SciPy
• Matplotlib
• pandas
• scikit-learn
• scikit-image

ソース

• https://formulae.brew.sh/formula/numpy
• https://numpy.org/about/
• https://numpy.org/doc/stable/f2py/
• https://numpy.org/doc/stable/user/whatisnumpy.html
• https://scipy.github.io/old-wiki/pages/History_of_SciPy
• https://www.nature.com/articles/s41586-020-2649-2