how to use the NaiveBayes.fit for classification in matlab?

[poster]

I'm using Naive Bayes classifier:

 for i=1 :100

 load('datas3.mat');

 data=datas3;

 [n,m]=size(data);

 rows=(1:n);

 testcount=floor((0.2)*n);

 acc1=0; SenAr=0;  SpeAr=0; MCCAr=0;

 kk=10;

 for k=1: kk

   test_rows=randsample(rows,testcount);

   train_rows=setdiff(rows,test_rows);

   test=data(test_rows,:);

   train=data(train_rows,:);

   xtest=test(:,1:m-1);

   ytest=test(:,m);

   xtrain=train(:,1:m-1);

   ytrain=train(:,m);

   NB=NaiveBayes.fit(xtrain,ytrain);

   out=NB.predict(xtest); 

these errors were shown:
Warning: NaiveBayes will be removed in a future release. Use fitcnb instead.

In NaiveBayes.fit (line 8)
In accu (line 30)
Error using NaiveBayes.fit>gaussianFit (line 373)
The within-class variance in each feature of TRAINING must be positive. The within-class variance
in feature 1 in class 0
are not positive.
Error in NaiveBayes.fit (line 337)
obj = gaussianFit(obj, training, gindex);
Error in accu (line 30)
NB=NaiveBayes.fit(xtrain,ytrain);

I'll be very grateful to have your opinions. Thanks.

More answer...