Recommendation.jl is a minimal, customizable Julia package for building recommender systems. Pre-built basic functionalities include:
julia> using Pkg; Pkg.add("Recommendation")
This package contains a unified DataAccessor
module and several non-personalized/personalized recommenders, as well as evaluation metrics such as Recall
:
See Getting Started in documentation for the details.
Change the code and test locally:
julia> using Pkg; Pkg.activate(@__DIR__); Pkg.instantiate()
julia> Pkg.test("Recommendation")
Note that unit tests for dataset loaders (e.g., load_movielens_latest()
) are conditionally triggered as follows, so that CI does not make excessive download requests to the external sites:
julia> Pkg.test("Recommendation", test_args=["data", "download"])
Build documentation contents:
$ julia --project=docs -e 'using Pkg; Pkg.develop(PackageSpec(path=pwd())); Pkg.instantiate()'
$ julia --project=docs docs/make.jl
$ open docs/build/index.html
Follow JuliaRegistries/Registrator.jl for releasing.