Dimensionality Reduction with PCA (SVD)

In [1]:

    

%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd

from sklearn import decomposition, datasets
from matplotlib.colors import ListedColormap


    

In [2]:

    

iris = datasets.load_iris()
pca = decomposition.PCA(n_components=2)
new_dim = pca.fit_transform(iris.data)


    

In [3]:

    

df = pd.DataFrame(new_dim, columns=['X', 'Y'])
df['label'] = iris.target
df.head()


    

    
Out[3]:






  

    

      

      
X
      
Y
      
label
    
  
  

    

      
0
      
-2.684207
      
0.326607
      
0
    
    

      
1
      
-2.715391
      
-0.169557
      
0
    
    

      
2
      
-2.889820
      
-0.137346
      
0
    
    

      
3
      
-2.746437
      
-0.311124
      
0
    
    

      
4
      
-2.728593
      
0.333925
      
0
    
  

In [4]:

    

fig = plt.figure()
fig.suptitle('PCA (SVD)', fontsize=14, fontweight='bold')
ax = fig.add_subplot(111)

plt.scatter(df[df.label == 0].X, df[df.label == 0].Y, color='red', label=iris.target_names[0])
plt.scatter(df[df.label == 1].X, df[df.label == 1].Y, color='blue', label=iris.target_names[1])
plt.scatter(df[df.label == 2].X, df[df.label == 2].Y, color='green', label=iris.target_names[2])

plt.legend(bbox_to_anchor=(1.25, 1))


    

    
Out[4]:





<matplotlib.legend.Legend at 0x10b807e48>