augmentDeepInsight2.m

function [XTrainNew,YTrainNew] = augmentDeepInsight2(XTrain,YTrain,num)
% augment non-image samples to make it balance for DeepInsight procedure

class = length(unique(double(YTrain)));
%num=500;

for j=1:class
    max_class(j) = sum(double(YTrain)==j);
end
inx=1:length(double(YTrain));
XTrainNew=[]; YTrainNew=[];
for j=1:class
    if max_class(j) < num % augment
        [XTrainNewClass,YTrainNewClass] = augmentDeepInsightClass(XTrain,YTrain,num,j,inx); 
        XTrainNew = cat(4,XTrainNew,XTrainNewClass);
        YTrainNew = [YTrainNew;YTrainNewClass];
    end
end
XTrainNew=cat(4,XTrain,XTrainNew);
YTrainNew=[YTrain;YTrainNew];

% [row,col]=sort(max_class,'descend');
% MaxClass = col(1);
% MaxClassVal = row(1);
% MinClass = col(end);
% MinClassVal = row(end);
% MinClassVal_samples = MinClassVal*(MinClassVal+1)/2;% + (MinClassVal-1)*(MinClassVal-2)/2;
% num = MinClassVal_samples-MaxClassVal;
% AUG=1000;
% if num>1000
%     num=1000;
%     AUG=100;
% end
% 
% if num>0
%     %perform
%     inx=1:length(double(YTrain));
%     XTrainNew=[];
%     YTrainNew=[];
%     for j=1:class
%         num = MinClassVal_samples - max_class(j);
%         if num > 1000
%             num=1000;
%         end
%         [XTrainNewClass,YTrainNewClass] = augmentDeepInsightClass(XTrain,YTrain,num,j,inx);     
%         XTrainNew = cat(4,XTrainNew,XTrainNewClass);
%         YTrainNew = [YTrainNew;YTrainNewClass];
%     end
%     inx2=1:length(double(YTrainNew));
%     for j=1:class
%         [XTrainNewClass,YTrainNewClass] = augmentDeepInsightClass(XTrainNew,YTrainNew,AUG,j,inx2);
%         XTrainNew = cat(4,XTrainNew,XTrainNewClass);
%         YTrainNew = [YTrainNew;YTrainNewClass];
%     end
% else
%     inx=1:length(double(YTrain));
%     XTrainNew=[];
%     YTrainNew=[];
%     for j=1:class
%         num = MaxClassVal - max_class(j);
%         if num>0
%         [XTrainNewClass,YTrainNewClass] = augmentDeepInsightClass(XTrain,YTrain,num,j,inx);     
%         XTrainNew = cat(4,XTrainNew,XTrainNewClass);
%         YTrainNew = [YTrainNew;YTrainNewClass];
%         else
%             [XTrainNewClass,YTrainNewClass] = augmentDeepInsightClass(XTrain,YTrain,AUG,j,inx);
%             XTrainNew = cat(4,XTrainNew,XTrainNewClass);
%             YTrainNew = [YTrainNew;YTrainNewClass];
%         end
%         
%     end
%     inx2=1:length(double(YTrainNew));
%     for j=1:class
%         if j ~= MaxClass
%             [XTrainNewClass,YTrainNewClass] = augmentDeepInsightClass(XTrainNew,YTrainNew,AUG,j,inx2);
%             XTrainNew = cat(4,XTrainNew,XTrainNewClass);
%             YTrainNew = [YTrainNew;YTrainNewClass];
%         end
%     end
%     disp('samples in lower class are too scarce to make an estimate');
% end
% XTrainNew=cat(4,XTrain,XTrainNew);
% YTrainNew=[YTrain;YTrainNew];