MLCore.jl is a Julia package providing some basic API for machine learning tasks. It is meant to be used and extended by other packages such as MLUtils.jl.

Defines the following methods:

• numobs(x)
• getobs(data, idx), getobs(data, idxs), and getobs(data)
• getobs!(buffer, data, idx)

It also provides implementations for Base types such as arrays, tuples, named tuples, and dictionaries Also provides implementations for Tables.jl tables.

Read the documentation for more details.

Related Packages

• MLUtils.jl: Contains a broad set of utilities for machine learning tasks. Methods in MLCore.jl used to be part of MLUtils.jl but were moved to MLCore.jl to reduce dependencies.
• DataAPI.jl: Defines a common API for working with data in Julia. Mainly targeting tabular data.
• StatsAPI.jl: Defines a common API for statistical operations in Julia. Some methods in MLCore.jl are inspired by StatsAPI.jl but have different semantics (see this issue).
• LearnAPI.jl: A broad API for machine learning models in Julia. MLCore.jl is meant to be complementary to LearnAPI.jl (see this issue).