In [1]:

    

from pml.api import *


    

In [2]:

    

data = load("../dataset_ext2.csv")
data.fill_missing_with_feature_means()


    

In [3]:

    

reduced = pca(data, 2)


    

In [7]:

    

clustered = kmeans(reduced, k=2, distance=euclidean)


    

In [11]:

    

clustered.calculate_purity()


    

    
Out[11]:





0.5625

In [12]:

    

data.num_features()


    

    
Out[12]:





13

In [13]:

    

data.num_samples()


    

    
Out[13]:





80

In [ ]: