bai20_SoftMargin_SVM_sklearn.py

# -*- coding: utf-8 -*-
"""
Created on Fri May  1 11:36:39 2020

@author: phamk
"""


# To support both python 2 and python 3
from __future__ import division, print_function, unicode_literals

from matplotlib import rc
rc('font',**{'family':'sans-serif','sans-serif':['Helvetica']})
## for Palatino and other serif fonts use:
#rc('font',**{'family':'serif','serif':['Palatino']})
rc('text', usetex=True)

# list of points 
import numpy as np 
import matplotlib.pyplot as plt
from scipy.spatial.distance import cdist
from matplotlib.backends.backend_pdf import PdfPages
np.random.seed(22)

means = [[2, 2], [4, 2]]
cov = [[.7, 0], [0, .7]]
N = 20
X0 = np.random.multivariate_normal(means[0], cov, N) # each row is a data point 
X1 = np.random.multivariate_normal(means[1], cov, N)

with PdfPages('data.pdf') as pdf:
    plt.plot(X0[:, 0], X0[:, 1], 'bs', markersize = 8, alpha = 1)
    plt.plot(X1[:, 0], X1[:, 1], 'ro', markersize = 8, alpha = 1)
    plt.axis('equal')
    plt.ylim(0, 4)
    plt.xlim(0, 5)

    # hide tikcs 
    cur_axes = plt.gca()
    cur_axes.axes.get_xaxis().set_ticks([])
    cur_axes.axes.get_yaxis().set_ticks([])

    plt.xlabel('$x_1$', fontsize = 20)
    plt.ylabel('$x_2$', fontsize = 20)
    pdf.savefig()
    # plt.savefig('logistic_2d.png', bbox_inches='tight', dpi = 300)
    plt.show()
    
X = np.vstack((X0, X1))
y = np.vstack((np.ones((N,1 )), -np.ones((N,1 )))).reshape((2*N,))

from sklearn.svm import SVC
C = 100
clf = SVC(kernel = 'linear', C = C)
clf.fit(X, y) 

w_sklearn = clf.coef_.reshape(-1, 1)
b_sklearn = clf.intercept_[0]
print(w_sklearn.T, b_sklearn)