Confusion Matrix of SVM

Hello

I would like to ask about how to find the valure of Linear, Gaussian, Poly = 2, Poly = 3 for dataset of iris

I used the code bellow it runs without problem but I don't know how to calculate the value of them

t = templateSVM('Standardize',true,'BoxConstraint',100,'KernelFunction','linear','KernelScale','auto');

Mdl = fitcecoc(dataTrain(:,2:4),dataTrain.Species,'Learners',t);

Predictions_SVM_Linear = predict(Mdl,dataTest(:,2:4));

figure;

C_SVM_Linear = confusionmat(dataTest.Species,Predictions_SVM_Linear);

cm_SVM_Linear = confusionchart(C_SVM_Linear,{'Iris-setosa','Iris-versicolor','Iris-virginia'});

cm_SVM_Linear.Title = 'Iris Classification Using Linear SVM';

cm_SVM_Linear.RowSummary = 'row-normalized';

cm_SVM_Linear.ColumnSummary = 'column-normalized';

Answer 1

[Pranav Verma]

Hi Baraah,

From your question I understand that you want to use Linear, Gaussian and Polynomial Kernel functions in templateSVM function. You can use the following name, value pairs for these:

• 'KernelFunction','gaussian'
• 'KernelFunction','linear'
• 'KernelFunction','polynomial','PolynomialOrder',2 (for polynomial of order '2')

For further information on the Kernel Functions, please refer to the below documentation:

https://www.mathworks.com/help/stats/templatesvm.html?s_tid=srchtitle#namevaluepairarguments

Thanks

[Lana Aydin]

Ok thank you Pranav, I will check it out. Thanks again.