Algorithms for Feature Weighting

Iterative RELIEF

class mlpy.IRelief(T=1000, sigma=1.0, theta=0.001)

Iterative RELIEF for feature weighting.

Parameters :
T : integer (> 0)

max loops

sigma : float (> 0.0)

kernel width

theta : float (> 0.0)

convergence parameter

learn(x, y)

Compute the feature weights.

Parameters :
x : 2d array_like object

training data (N, P)

y : 1d array_like object integer (only two classes)

target values (N)

Raises :

SigmaError

loops()

Returns the number of loops.

weights()

Returns the feature weights.

[Sun07]Yijun Sun. Iterative RELIEF for Feature Weighting: Algorithms, Theories, and Applications. IEEE Trans. Pattern Anal. Mach. Intell. 29(6): 1035-1051, 2007.

Table Of Contents

Previous topic

Cluster Analysis

Next topic

Feature Selection

This Page