--- a
+++ b/Semantic Features/GetXwsSampleConcat.m
@@ -0,0 +1,19 @@
+function [XwsSampleConcat, YSampleConcat] = GetXwsSampleConcat(X, Y, numSegments)
+%GetXwsSampleConcat Creates a proper X and Y matrix suitable for the
+%ensemble learning processes we had designed. (2 levels). 
+%  Rearranges the raw Xws data so the segments are each listed one after
+%  the other sample wise down the rows. To match it, an Y matrix is made
+%  with the same values copied so the rows for each segmenetation match the
+%  row it belongs to for labels. Probably should have made this into an
+%  array of X matrixes (3d) but this worked. 
+
+newWidth = size(X,2) / numSegments;
+
+XwsSampleConcat = X(:,1:newWidth);
+YSampleConcat = Y;
+for i = 2:numSegments
+    XwsSampleConcat = vertcat(XwsSampleConcat, X(:,(i-1)*newWidth+1:i*newWidth));
+    YSampleConcat = vertcat(YSampleConcat, Y);
+end
+
+end
\ No newline at end of file