asbisen / mlbase.jl

A set of functions to support the development of machine learning algorithms

MLBase.jl

Swiss knife for machine learning.

This package does not implement specific machine learning algorithms. Instead, it provides a collection of useful tools to support machine learning programs, including:

• Data manipulation & preprocessing
• Score-based classification
• Performance evaluation (e.g. evaluating ROC)
• Cross validation
• Model tuning (i.e. search best settings of parameters)

Notes: This package depends on StatsBase and reexports all names therefrom.

Resources

• Documentation: http://mlbasejl.readthedocs.org/en/latest/
• Release Notes: https://github.com/JuliaStats/MLBase.jl/blob/master/NEWS.md