Test code for libsvm-python.

libsvm_test.py

#!/usr/bin/env python                                                                                     
#-*- coding:utf-8 -*-                                                                                     

from svm import *
from svmutil import *

# For learning                                                                                            
t_label = [1,-1,1,-1]
t_data = [
    [1.0, 2.0, 3.0],
    [3.0, 1.5, 1.0],
    [2.0, 3.0, 4.0],
    [0.5, 1.0, 1.5]
    ]
problem = svm_problem(t_label, t_data)
parameter = svm_parameter('-s 0 -t 0')
t = svm_train(problem, parameter)

# For predict                                                                                             
p_label = [1, 1, -1, -1]
p_data = [
    [0.3, 0.9, 1.2],
    [2.0, 3.0, 4.5],
    [3.0, 1.0, 0.3],
    [1.0, 0.5, 0.25]
    ]
result = svm_predict(p_label, p_data , t)

print "[Result]"
for r in result:
    print r