In [1]:

    

# Load libraries
from sklearn.svm import SVC
from sklearn import datasets
from sklearn.preprocessing import StandardScaler
import numpy as np


    

In [2]:

    

#Load data with only two classes
iris = datasets.load_iris()
X = iris.data[:100,:]
y = iris.target[:100]


    

In [3]:

    

# Standarize features
scaler = StandardScaler()
X_std = scaler.fit_transform(X)


    

In [4]:

    

# Create support vector classifier object
svc = SVC(kernel='linear', random_state=0)

# Train classifier
model = svc.fit(X_std, y)


    

In [5]:

    

# View support vectors
model.support_vectors_


    

    
Out[5]:





array([[-0.5810659 ,  0.43490123, -0.80621461, -0.50581312],
       [-1.52079513, -1.67626978, -1.08374115, -0.8607697 ],
       [-0.89430898, -1.46515268,  0.30389157,  0.38157832],
       [-0.5810659 , -1.25403558,  0.09574666,  0.55905661]])

In [6]:

    

# View indices of support vectors
model.support_


    

    
Out[6]:





array([23, 41, 57, 98], dtype=int32)

In [7]:

    

# View number of support vectors for each class
model.n_support_


    

    
Out[7]:





array([2, 2], dtype=int32)