So are we, I will say I copy this portion and there are two, two here. So these are going to be a classification system that specify the algorithm string algorithm equation. see on this slide, we get a training data can we get a training data, training data, and then we need to get a testing data so, I copy this URL and I pour testing data here. Test the data testing data need to write a class index specify the size that the string I want to index will be integer copy here and I post on my here specify the class index of training data grass index or training data. Copy again specify a crossing does all the testing data a cross is also be training okay this one should be testing that cross a vacation. So, clustering I change to classification.
Okay. So the first one should be training data. The second one should be testing data training in data, testing data and this one should be called algorithm training testing index Testing Testing index then for the key Kenya semicolon, then for option three I will copy the whole thing here and then this is evaluation evaluate model should be the same as classification method. So, I changed this to be evaluate model Okay. Then all this is to remove Okay. Then I run the code Okay, so let's say I choose one.
So please specify that algorithm. So one is our clustering. Should be a K means. So I go into my code I say for cluster ri. k means specify the number of clusters. Maybe I give around AI and training data, Guevara D dry the dry ah maybe I reserve CPU.
So let's see I go to the tea dry here. I go into the iris AI ever. Okay, so as you specify here Iris, ar is a cluster for me. So I have around cluster two, cluster 3456 for cluster a cluster nine. So this is clustering guy. Let's see.
If I want to do classification, I can select to specify a algorithm. So let's say a classification and to use tree decision tree. The training data sets to PD drive. Maybe for classification for prediction, I use CPU for classification I use Iris. Specify the testing set the dry eye race test. So she'll be Iris testing say, race test.ai ba, ba AR and then specify the crossing that all should be four, four.
Okay, so I have all the prediction and classification here. They are, let's say I want to do model evaluation. Random program again. So, I can do something on this tree. The algorithm I will use maybe decision tree or MLP to use tree to specify the training set the drive I raised er testings are also the same industry before for and I get all the classification result here are the accuracy or the evaluation result here