LearnDataFrontEnds.jl is a package for developers writing LearnAPI.jl interfaces for algorithms in machine learning or statistics.
A LearnAPI.jl implementation including a data front end from this package gives users the flexibility to supply data in a variety of ways (matrices, tables, R-style formulas, etc) without the need to write the relevant boiler plate to coerce that data into array representations ultimately needed by the core algorithm. Moreover, the design of LearnAPI.jl means these conversions are not unnecessarily repeated when applying meta-algorithms, such as cross-validation.
Documentation is here.