In [70]:

    

import pandas as p
from sklearn import datasets,metrics,svm,cross_validation
import numpy as n
import matplotlib.pyplot as pl
from matplotlib.colors import ListedColormap


    

In [8]:

    

iris1 = datasets.load_iris()


    

In [10]:

    

irisd = iris1.data[:,:2]
irist = iris.target


    

In [ ]:

    




    

In [ ]:

    




    

In [13]:

    

#x_train,x_test,y_train,y_test = cross_validation.train_test_split(iris.data,iris.target,test_size = 0.2)


    

In [264]:

    

e = svm.SVC(kernel = 'rbf',C = 1000000, gamma = 0.1)


    

In [265]:

    

e.fit(irisd,irist)


    

    
Out[265]:





SVC(C=1000000, cache_size=200, class_weight=None, coef0=0.0,
  decision_function_shape=None, degree=3, gamma=0.1, kernel='rbf',
  max_iter=-1, probability=False, random_state=None, shrinking=True,
  tol=0.001, verbose=False)

In [266]:

    

x_min = irisd[:,0].min() - 1


    

In [267]:

    

y_min = irisd[:,1].min() - 1


    

In [268]:

    

x_max = irisd[:,0].max() + 1


    

In [269]:

    

y_max = irisd[:,1].max() + 1


    

In [270]:

    

h = (x_max - x_min)/100
xrange = n.arange(x_min,x_max,h)
h = (y_max - y_min)/100
yrange = n.arange(y_min,y_max,h)


    

In [271]:

    

h1


    

    
Out[271]:





0.056000000000000008

In [272]:

    

h2


    

    
Out[272]:





0.044000000000000004

In [273]:

    

xrange.shape


    

    
Out[273]:





(100,)

In [274]:

    

yrange.shape


    

    
Out[274]:





(100,)

In [275]:

    

xx,yy = n.meshgrid(xrange,yrange)


    

In [276]:

    

xx.ravel()
yy.ravel()


    

    
Out[276]:





array([ 1.   ,  1.   ,  1.   , ...,  5.356,  5.356,  5.356])

In [277]:

    

q = n.c_[xx.ravel(),yy.ravel()]


    

In [278]:

    

pred = e.predict(q)


    

In [279]:

    

%matplotlib inline


    

In [280]:

    

pred = pred.reshape(xx.shape)


    

In [281]:

    

#pl.figure()
cmap_light = ListedColormap(['#FFAAAA', '#AAFFAA', '#AAAAFF'])
cmap_bold = ListedColormap(['#FF0000', '#00FF00', '#0000FF'])


    

In [ ]:

    




    

In [282]:

    

pl.figure()


    

    
Out[282]:





<matplotlib.figure.Figure at 0x263ec8f86a0>






    






<matplotlib.figure.Figure at 0x263ec8f86a0>

In [ ]:

    




    

In [283]:

    

#l.contour(xx,yy,pred)
pl.pcolormesh(xx, yy, pred, cmap=cmap_light, shading = 'gouraud',edgecolor='face') 
pl.scatter(irisd[:,0],irisd[:,1],c = irist,cmap=cmap_bold)


    

    
Out[283]:





<matplotlib.collections.PathCollection at 0x263ec969c50>






    

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    

#PCA - principal component analysis- hidden meanings


    

In [ ]: