1 import matplotlib.pyplot as plt
 2 from sklearn.decomposition import PCA
 3 from sklearn.datasets import load_iris
 4  
 5 data = load_iris()
 6 y = data.target
 7 X = data.data
 8 pca = PCA(n_components=2)
 9 reduced_X = pca.fit_transform(X)
10  
11 red_x, red_y = [], []
12 blue_x, blue_y = [], []
13 green_x, green_y = [], []
14  
15 for i in range(len(reduced_X)):
16     if y[i] == 0:
17         red_x.append(reduced_X[i][0])
18         red_y.append(reduced_X[i][1])
19     elif y[i] == 1:
20         blue_x.append(reduced_X[i][0])
21         blue_y.append(reduced_X[i][1])
22     else:
23         green_x.append(reduced_X[i][0])
24         green_y.append(reduced_X[i][1])
25  
26 plt.scatter(red_x, red_y, c=‘r‘, marker=‘x‘)
27 plt.scatter(blue_x, blue_y, c=‘b‘, marker=‘D‘)
28 plt.scatter(green_x, green_y, c=‘g‘, marker=‘.‘)
29 plt.show()

 1 from numpy.random import RandomState
 2 import matplotlib.pyplot as plt
 3 from sklearn.datasets import fetch_olivetti_faces
 4 from sklearn import decomposition
 5  
 6  
 7 n_row, n_col = 2, 3
 8 n_components = n_row * n_col
 9 image_shape = (64, 64)
10  
11  
12 ###############################################################################
13 # Load faces data
14 dataset = fetch_olivetti_faces(shuffle=True, random_state=RandomState(0))
15 faces = dataset.data
16  
17 ###############################################################################
18 def plot_gallery(title, images, n_col=n_col, n_row=n_row):
19     plt.figure(figsize=(2. * n_col, 2.26 * n_row)) 
20     plt.suptitle(title, size=16)
21  
22     for i, comp in enumerate(images):
23         plt.subplot(n_row, n_col, i + 1)
24         vmax = max(comp.max(), -comp.min())
25  
26         plt.imshow(comp.reshape(image_shape), cmap=plt.cm.gray,
27                    interpolation=‘nearest‘, vmin=-vmax, vmax=vmax)
28         plt.xticks(())
29         plt.yticks(())
30     plt.subplots_adjust(0.01, 0.05, 0.99, 0.94, 0.04, 0.)
31  
32      
33 plot_gallery("First centered Olivetti faces", faces[:n_components])
34 ###############################################################################
35  
36 estimators = [
37     (‘Eigenfaces - PCA using randomized SVD‘,
38          decomposition.PCA(n_components=6,whiten=True)),
39  
40     (‘Non-negative components - NMF‘,
41          decomposition.NMF(n_components=6, init=‘nndsvda‘, tol=5e-3))
42 ]
43  
44 ###############################################################################
45  
46 for name, estimator in estimators:
47     print("Extracting the top %d %s..." % (n_components, name))
48     print(faces.shape)
49     estimator.fit(faces)
50     components_ = estimator.components_
51     plot_gallery(name, components_[:n_components])
52  
53 plt.show()