Hello everyone, welcome to the course of machine learning with Python. In this video, we should learn how to implement support vector machine classifier using the escalon library of Python. So first we should import necessary libraries. So let's go ahead and run this. Note that from SK learn dot CSV we have imported is VC is VC stands for Support Vector classifier. Now, we shall load the data set.

So again, we will be classifying Iris data set so we have imported it as data set over here. So in x and y, we are loading the ITC data set. Now we should split the data set into create and test so from SK learn dot model underscore selection we have imported create underscore test underscore split here we have specified this size to be equals to 0.25. That means 25% of the original data points will be assigned to the test data and risk 75 will be as cleaning data okay? So, let's go ahead and run this. So, as we can see, there are total 112 training data points and 38 test data points.

Now, we should normalize the data or standardize the data. So, from a skill and pre processing, we are importing standard scaler class is C is the instance adoption of the standard scaler class now, we shall fit transform our x train and x test with this standard scalar object okay. So, after doing this all the columns in extreme will have unit credits as zero mean and Same goes with the column of x test. Now, we should define our support vector machine classifier with linear and polynomial kernel. So, this linear Support Vector classifier is equals to SVC we are specifying kernel to be equals to linear and this is polynomial Support Vector classifier where we have specified kernel to be close to polynomial and Deeply three, we can specify any degree over here, but usually the v2 or v3 is preferable, because it is computationally less expensive as compared some higher degree Fine.

Let's go ahead and on this particular cell, now, we should fit the training data set on our support vector classifier. So, we are fitting both linear Support Vector classifier as well as the polynomial Support Vector classifier with our training data set. Now, we shall make the prediction using Support Vector classifier. So one thing I must mention over here, so decision functions shape is equals to over here that means one versus rest, this is because it is a multi class classification. so here also decision function shape is equals to over here five. Now we shall make the predictions using Support Vector classifier.

So while creating one, these particular variables towards all the prediction made by the linear classifier Linear Support Vector classifier and why credit to stores all the prediction made by polynomial support vector machine classifier. Now, we should evaluate the model. So, from SK learn dot matrix we are importing confusion matrix no confusion matrix for linear classifier we have called this confusion matrix function and in the place of one credit we have passed why credit one and in the place of while true we have passed white test which was our original levels of test dataset or this is basically the ground truth. So, we will pin the confusion matrix for linear classifier, okay. Similarly, we go ahead and do the same for polynomial classifier, okay. Now as you can see, the accuracy for the linear classifier can be obtained by the trees of the confusion matrix boosted by the linear classifier divided by the total number of points.

And for the point of this classifier the same would be the confusion matrix of the country. matrix produced by the polymer Support Vector classifier divided by the total number of test data points. Okay, so let's go ahead and compute this matrix accuracy for linear support vector machine classifier and accuracy for polymer support vector machine classifier. So let's go ahead and run this particular set. So at least for linear kernel support vector machine classifier is 97.37%. And for polynomial of degree three, it is hundred percent Okay, so we have seen how to implement support vector machine classifier in this video.

In the next video, we shall introduce another classifier known as Naive Bayes classifier. So see you in the next lecture. Thank you.