DataScienceTraits.jl

October 18, 2024 ยท View on GitHub

Build Status Coverage

This package provides an alternative implementation to ScientificTypes.jl that is lightweight, with a more sensible set of defaults for data science. See https://github.com/JuliaAI/ScientificTypes.jl/issues/180.

Usage

This package is intended for developers of statistical packages that need to dispatch different algorithms depending on scientific types:

using DataScienceTraits

reduction(::Continuous) = sum
reduction(::Categorical) = first

Extensions are provided for third-party types such as CoDa.jl and CategoricalArrays.jl in order to facilitate the integration with existing software stacks.