MaximalInformationOptions

Options & Configuration · AiDotNet.Models.Options

The MINE grid parameters of the maximal information coefficient (Reshef et al. 2011, “Detecting Novel Associations in Large Data Sets”, Science 334), as minepy’s mine_parameter.

For Beginners

MIC tries many grids over a scatter plot of a feature against the target. Alpha sets how many grid cells it may use, and C how finely each column may first be split before columns are merged. The defaults are the paper’s.

Constructors

Constructor Summary
MaximalInformationOptions()

Properties

Property Summary
double Alpha { get; set; } The grid budget exponent: B(n) = max(n^alpha, 4) for alpha in (0, 1], or B = min(alpha, n) for alpha of at least 4 (paper and minepy default 0.6).
double C { get; set; } The clump factor: the x axis is split into at most c times as many superclumps as columns (default 15).