In [4]:

    

import matplotlib.pyplot as plt
from sklearn import datasets,svm

# Import the Iris dataset and keep the first two features
iris = datasets.load_iris()
x_train = iris.data[:,:2]
y_train = iris.target

# Instantiate the model and fit to data
clf = svm.SVC(C=1,kernel='rbf',probability=True)
clf.fit(iris.data[:,:2],iris.target)

# Predict (maximum) probabilities for plot
x = arange(4,9,0.05)
y = arange(1,6,0.05)
xx,yy = meshgrid(x,y)
zz = clf.predict_proba(np.vstack((xx.ravel(),yy.ravel())).T).max(axis=1).reshape(xx.shape)

# Create plot
f = plt.figure(figsize=(14,8))
ax = f.add_subplot(111,aspect='equal')
for i,c,label in zip(arange(3),['r','g','b'],iris.target_names):
    plt.scatter(x_train[y_train==i][:,0],x_train[y_train==i][:,1],s=50,color=c,edgecolor='k',label=label)
pcol = pcolormesh(xx,yy,zz,alpha=0.3,zorder=0)

plt.xlim(4,8)
plt.ylim(1.5,4.5)
plt.xlabel('Sepal Length',fontsize=14)
plt.ylabel('Sepal Width',fontsize=14)
plt.title('Python',fontsize=14)
plt.tick_params(axis='both',which='major',labelsize=14)
plt.legend()
plt.colorbar()
plt.savefig('pyplot.png')