select.top.features: Function to select top features

In pranithavangala/SeqGL_pv: Group lasso for Dnase/ChIP-seq data

Description

The differences are determined by weighted means. The number of examples in which a particular feature is defined is used to estimate the means in both classes.

Usage

select.top.features(features, labels, feature.count = 10000,
  min.example.fraction = 0.01)

Arguments

features	Feature matrix
labels	Feature labels (Should be +1/-1)
feature.count	Number of features to select. Default is 10000.
min.example.fraction	Minimal fraction of examples that a feature should defined in. Default is 1%.

Value

Vector of selected feature indexes.

pranithavangala/SeqGL_pv documentation built on May 25, 2019, 11:25 a.m.