In [10]:

    

from sklearn import datasets
from sklearn.decomposition import RandomizedPCA
from sklearn.decomposition import PCA
from sklearn import svm

import pylab as pl
import numpy as np


    

In [2]:

    

iris = datasets.load_iris()


    

In [4]:

    

X, y = iris.data, iris.target


    

In [6]:

    

X.shape, y.shape


    

    
Out[6]:





((150, 4), (150,))

In [7]:

    

iris.target_names


    

    
Out[7]:





array(['setosa', 'versicolor', 'virginica'], 
      dtype='|S10')

In [8]:

    

X_rand_pca = RandomizedPCA(n_components=2).fit_transform(X)


    

In [9]:

    

X_rand_pca.shape


    

    
Out[9]:





(150, 2)

In [12]:

    

from itertools import cycle
color = ['b', 'g', 'r']
for i,c in zip(np.unique(y), cycle(color)):
    pl.scatter(X_rand_pca[y==i,0], X_rand_pca[y==i, 1], c=c, label=i, alpha=.8)


    

In [13]:

    

from sklearn.decomposition import PCA
iris_pca_basic = PCA(n_components=2).fit_transform(iris.data)

for i, c in zip(np.unique(iris.target), cycle(color)):
    pl.scatter(iris_pca_basic[iris.target==i,0], iris_pca_basic[iris.target==i,1],
               c=c, alpha=.8)


    

In [30]:

    

from sklearn.decomposition import ProbabilisticPCA, KernelPCA
iris_pca_probalistic = ProbabilisticPCA(n_components=2).fit_transform(iris.data)

X_p= iris_pca_probalistic
for i, c in zip(np.unique(iris.target), cycle(color)):
    pl.scatter(X_p[iris.target==i,0], X_p[iris.target==i,1], c=c, alpha=.7, label=iris.target_names[i])
    pl.legend(loc='upper right')


    

In [29]:

    

from sklearn.decomposition import ProbabilisticPCA, KernelPCA
iris_pca_probalistic = KernelPCA(n_components=2).fit_transform(iris.data)

X_p= iris_pca_probalistic
for i, c, name in zip(np.unique(iris.target), cycle(color), np.unique(iris.target_names)):
    pl.scatter(X_p[iris.target==i,0], X_p[iris.target==i,1], c=c, alpha=.7, label=name)
    pl.legend(loc='lower right')


    

In [32]:

    

clf = svm.SVC(kernel='rbf').fit(X_p, iris.target)


    

In [33]:

    

clf


    

    
Out[33]:





SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0, degree=3, gamma=0.0,
  kernel='rbf', max_iter=-1, probability=False, random_state=None,
  shrinking=True, tol=0.001, verbose=False)

In [34]:

    

from sklearn.cross_validation import cross_val_score
cross_val_score(clf, X_p, iris.target, cv=10)


    

    
Out[34]:





array([ 1.        ,  1.        ,  0.8       ,  1.        ,  0.86666667,
        1.        ,  1.        ,  0.93333333,  0.93333333,  1.        ])

In [35]:

    

cross_val_score(svm.SVC(kernel='rbf'), iris.data, iris.target, cv=10)


    

    
Out[35]:





array([ 1.        ,  1.        ,  0.86666667,  1.        ,  0.93333333,
        1.        ,  1.        ,  1.        ,  1.        ,  1.        ])

In [36]:

    

iris.data.shape, X_p.shape


    

    
Out[36]:





((150, 4), (150, 2))

In [ ]: