SVM using cvxopt solvers - Printable Version

SVM using cvxopt solvers - Printable Version

+- Python Forum (https://python-forum.io)
+-- Forum: Python Coding (https://python-forum.io/forum-7.html)
+--- Forum: Data Science (https://python-forum.io/forum-44.html)
+--- Thread: SVM using cvxopt solvers (/thread-9537.html)

SVM using cvxopt solvers - cfc0712 - Apr-15-2018

I am using python(x,y) python 2.7.10.0

I want to build a SVM using cvxopt solvers but

"sol=solvers.qp(Q,q,A,b)
^
SyntaxError: invalid syntax
>>> Traceback (most recent call last):"

is shown when the solvers is called. Please kindly help solving the problem. Many thanks.

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.colors
from cvxopt import solvers
from cvxopt import matrix
from cvxopt.solvers import qp

X1=np.array([[3,4],[2,3]])
X2=np.array([[10,10], [11,11]])
X=np.concatenate((X1,X2), axis=0)

Y1=np.concatenate((np.ones((2,1)),X1), axis=1)
Y2=np.concatenate((np.ones((2,1))*-1,-X2), axis=1)

A=matrix(np.concatenate((Y1,Y2), axis=0),tc='d')

b=matrix(-1*np.ones((4,1)),tc='d')

Q=matrix(2*np.eye(3),tc='d')
q=matrix(np.zeros((3,1)),tc='d'

sol=solvers.qp(Q,q,A,b)
a_con=sol['x']
x=np.arange(0,100,15)
y=-(a_con[0]+a_con[1]*x)/a_con[2]
axes = plt.gca()
axes.set_xlim([0,20])
axes.set_ylim([0,20])
plt.scatter(X[[0,1],0], X[[0,1],1], color='k')
plt.scatter(X[[2,3],0], X[[2,3],1], color='r')
plt.plot(x,y)
plt.show()

RE: SVM using cvxopt solvers - scidam - Apr-15-2018

you omitted parentheses in line 20.