Mercurial > repos > deepakjadmin > feature_selection_test1

changeset 4:5364cf43a8c1 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author deepakjadmin
date Sun, 02 Oct 2016 05:36:30 -0400
parents 91c141c5efa6
children 016c69bfb2a1
files featureselect/feature_selection.R
diffstat 1 files changed, 4 insertions(+), 11 deletions(-) [+]
line wrap: on
line diff
--- a/featureselect/feature_selection.R	Fri Mar 25 06:48:10 2016 -0400
+++ b/featureselect/feature_selection.R	Sun Oct 02 05:36:30 2016 -0400
@@ -22,7 +22,7 @@
 
 colRate <- apply(rawData[, predictorNames, drop = FALSE],
                  2, function(x) mean(is.na(x)))
-colExclude <- colRate > 0.01
+colExclude <- colRate > 0.1
 	if(any(colExclude)){
 				predictorNames <- predictorNames[-which(colExclude)]
 				rawData <- RAWDATA[, c(predictorNames,"outcome")]
@@ -32,15 +32,8 @@
                 		rowRate <- apply(rawData[, predictorNames, drop = FALSE],
                  		1, function(x) mean(is.na(x)))
 			
-rowno <- dim(rawData)[1]
-if (rowno <= 1000){
-cutoff <- rowno / (rowno * 100)
-} else if (rowno > 1000 & rowno <= 5000) { 
-cutoff <- rowno / (rowno * 100 * 0.5 )	
-} else {
-cutoff <- rowno / (rowno * 100 * 0.5 * 0.5)
-}
-rowExclude <- rowRate > cutoff
+
+rowExclude <- rowRate > 0
 	if(any(rowExclude)){
   				rawData <- rawData[!rowExclude, ]
     				##hasMissing <- apply(rawData[, predictorNames, drop = FALSE],
@@ -55,7 +48,7 @@
 
                     		} 
                     
-set.seed(1234)
+set.seed(2)
 
 #print(dim(dataX))
 #print(dim(rawData))