Description
1. Use Matlab’s quadprog() function or Python’s CVXOPT package to implement the kernel
SVM and test its functionality with the data set generated as shown below. Use a Gaussian
kernel with 𝜎𝜎 = 1.75. For 𝐶𝐶 = 10 and 𝐶𝐶 = 100, plot the samples, margin hyperplanes, and
the decision boundary. Also, on the plot, identify and give the count of the support vectors
and the misclassified samples.
rng(100);
class1=mvnrnd([1 3],[1 0; 0 1],60);
class2=mvnrnd([4 1],[2 0; 0 2],40);
For those programming in Python, the dataset is provided in the attached Excel file.
Archive your m-file or py-file together with a report containing the generated plots (2 in
total) in a zip file named Lastname_ExtraCredit.zip and upload it to Blackboard prior to
the deadline.
