cross_validation.py

from sklearn.datasets import load_svmlight_file
from sklearn.naive_bayes import MultinomialNB
from sklearn.svm.sparse import LinearSVC
from sklearn.cross_validation import StratifiedKFold
from sklearn import metrics
import numpy as np

X, y = load_svmlight_file("fr.vec")
y[y == -1] = 0
kf = StratifiedKFold(y, k = 10, indices=True)
#clf = MultinomialNB()
clf = LinearSVC()
mean_li = []
for train_index, test_index in kf:
    X_train, X_test = X[train_index], X[test_index]
    y_train, y_test = y[train_index], y[test_index]
    clf.fit(X_train, y_train)
    y_predicted = clf.predict(X_test)
    print metrics.confusion_matrix(y_test, y_predicted)
    print metrics.classification_report(y_test, y_predicted)
    mean_li.append(sum(y_predicted == y_test) / float(len(y_test)))

print np.mean(mean_li)

train_test_svm.py

#!/usr/bin/python

import sys
from numpy import loadtxt
import numpy as np
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.pipeline import Pipeline
from sklearn.svm.sparse import LinearSVC

my_data = loadtxt(sys.argv[1], delimiter='\t', dtype='S')
my_test_data = loadtxt(sys.argv[2], delimiter='\t', dtype='S')

text_clf = Pipeline([
	    ('vect', CountVectorizer()),
	    ('tfidf', TfidfTransformer()),
	    ('clf', LinearSVC()),
	])
print("Training...")
my_clf = text_clf.fit(my_data[:,4], my_data[:,3])

print("Done! \nClassifying test set...")
predicted = my_clf.predict(my_test_data[:,4])


print(np.mean(predicted == my_test_data[:,3]))