Hi!!!! I am using 2017a matlab. I am not able to use libsvm on it. The image shows the error. Should I install 32bit matlab??

Hi!!!! I am using 2017a matlab. I am not able to use libsvm on it. The image shows the error. Should I install 32bit matlab??

[Walter Roberson]

There is no error shown in the image. It is telling you that a compiler is already set up, and it is giving you a warning that code that was written a decade or more ago will probably need to be changed and recompiled. You have not attempted to do anything with libsvm in what you posted.

I recommend that you use the Statistics Toolbox svm instead of libsvm

[SHWETA KHARA]

Sir. I have data of size 1*6000 (all are decimal values). That is, a vector. My required output will contain 4 classes. If am using "apps -> svm" in matlab then it requires predictors and response. But I am not able to give "response". help me if you can.

[SHWETA KHARA]

I will multi-class SVM.

[Walter Roberson]

libsvm is not suitable for multiclass work itself, but there are some multiclass extensions available for it. But the svm in the statistics toolbox supports some multiclass work.

If you do not have target information for training with then you should probably not be using svm and should probably look one of the clustering techniques such as kmeans.

[SHWETA KHARA]

Sir.. Target in this means "required output" right? My required output will be "0/1/2/3". But somehow, I am not able to give it in response.

This is what I am getting.

[Walter Roberson]

The tool you are using expects that one of the columns contains the output that is appropriate for that that input row -- expects that one of the columns will contain 0, 1, 2, or 3 for that row. You would mark that one row as Response and leave the other rows as Predictor.