Confusion Matrix of SVM

5 次查看（过去 30 天）
Lana Aydin
Lana Aydin 2020-11-4
评论： Lana Aydin 2020-11-10
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';

请先登录，再回答此问题。

回答（1 个）

Pranav Verma
Pranav Verma 2020-11-10
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
  1 个评论
Lana Aydin
Lana Aydin 2020-11-10
Ok thank you Pranav, I will check it out. Thanks again.

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Statistics and Machine Learning Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by