def classify(features, results, test_features, test_results, C, gamma):
cli = "%s@%s" %(C, gamma)
st = time.time()
log.info("Classifier begins")
classifier = SVC(C=C, gamma=gamma, kernel="rbf")
classifier.fit(features, results)
st2 = time.time()
prediction = classifier.predict(test_features)
log.info("id: %s Training time: %s, Prediction time: %s" %(cli, st2-st, time.time()-st2) )
error = 0
for index, value in enumerate(prediction):
if test_results[index] != value:
error += 1
return (error/float(len(test_results))) * 100
def classify(train_file, test_file):
"""
Train a model and test
train_file: file that the model is trained on
test_file: file that is used to test the model
"""
X_train, y_train = load_svmlight_file(train_file)
X_test, y_test = load_svmlight_file(test_file, X_train.shape[1])
# X_train = X_train.todense()
# X_test = X_test.todense()
clf = SparseSVC(kernel="linear", C=0.2)
# clf = LogisticRegression(C=1.0, penalty='l1', tol=1e-6)
clf.fit(X_train, y_train)
y_predict = clf.predict(X_test)
print sklearn.metrics.classification_report(y_test, y_predict)
print sklearn.metrics.confusion_matrix(y_test, y_predict)
# Test for 10 rounds using the results from 10 fold cross validations
for i, (train_index, test_index) in enumerate(kf):
print "run %d" % (i+1)
X_train, X_test = X[train_index], X[test_index]
y_train, y_test = y[train_index], y[test_index]
X_den_train, X_den_test = X_den[train_index], X_den[test_index]
# feed models
clf_mNB.fit(X_train, y_train)
clf_ridge.fit(X_train, y_train)
clf_SGD.fit(X_train, y_train)
clf_lSVC.fit(X_train, y_train)
clf_SVC.fit(X_train, y_train)
# get prediction for this fold run
prob_mNB = clf_mNB.predict_proba(X_test)
prob_ridge = clf_ridge.decision_function(X_test)
prob_SGD = clf_SGD.decision_function(X_test)
prob_lSVC = clf_lSVC.decision_function(X_test)
prob_SVC = clf_SVC.predict_proba(X_test)
# add prob functions into the z 2d-array
z_temp = (prob_mNB + prob_ridge + prob_SGD + prob_lSVC + prob_SVC)
z = np.append(z, z_temp, axis=0)
# remove the first sub-1d-array of z, due to the creation with 0s
z = np.delete(z, 0, 0)
for i, (train_index, test_index) in enumerate(kf):
# print "run %d" % (i+1)
X_train, X_test = X[train_index], X[test_index]
y_train, y_test = y[train_index], y[test_index]
X_den_train, X_den_test = X_den[train_index], X_den[test_index]
# feed models
clf_mNB.fit(X_train, y_train)
# clf_bNB.fit(X_train, y_train)
clf_ridge.fit(X_train, y_train)
# clf_kNN.fit(X_train, y_train)
clf_lSVC.fit(X_train, y_train)
# clf_SVC.fit(X_train, y_train)
clf_SVC.fit(X_den_train, y_train)
# get prediction for this fold run
prob_mNB = clf_mNB.predict_proba(X_test)
# prob_bNB = clf_bNB.predict_proba(X_test)
prob_ridge = clf_ridge.decision_function(X_test)
# prob_kNN = clf_kNN.decision_function(X_test)
prob_lSVC = clf_lSVC.decision_function(X_test)
# prob_SVC = clf_SVC.predict_proba(X_test)
prob_SVC = clf_SVC.predict_proba(X_den_test)
# add prob functions into the z 2d-array
# z_temp = (prob_mNB + prob_ridge + prob_SGD + prob_lSVC + prob_SVC)
# z_temp = (prob_mNB + prob_ridge + prob_bNB + prob_lSVC + prob_SVC)
# z_temp = (prob_mNB + 2*prob_ridge + 2*prob_lSVC + prob_SVC)
z_temp = (prob_mNB + prob_ridge + prob_lSVC + prob_SVC)
# Initialize variables for couting the average
f1_all = []
f5_all = []
acc_all = []
pre_all = []
rec_all = []
# Test for 10 rounds using the results from 10 fold cross validations
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]
# fit and predict
clf.fit(X_train, y_train)
pred = clf.predict(X_test)
# print y_test
# print pred
# print type(pred)
# output tree into graph
# out = StringIO()
# out = export_graphviz(clf, out_file=out)
# metrics
# # Original
f1_score = metrics.f1_score(y_test, pred)
f5_score = metrics.fbeta_score(y_test, pred, beta=0.5)
clf_rdg = RidgeClassifier(tol=1e-1) clf_sgd = SGDClassifier(alpha=.0001, n_iter=50, penalty="l2") # Logistic regression requires OneVsRestClassifier which hides # its methods such as decision_function # It will require extra implementation efforts to use it as a candidate # for multilabel classification # clf_lgr = OneVsRestClassifier(LogisticRegression(C=1000,penalty='l1')) # kNN does not have decision function due to its nature # clf_knn = KNeighborsClassifier(n_neighbors=13) # train clf_nb.fit(X, y) clf_lsvc.fit(X, y) clf_rdg.fit(X, y) clf_svc.fit(X, y) clf_sgd.fit(X, y) print "Train time: %0.3fs" % (time() - t0) print # # predict by simply apply the classifier # # this will not use the multi-label threshold # predicted = clf_rdg.predict(X_new) # for doc, category in zip(docs_new, predicted): # print '%r => %s' % (doc, data_train.target_names[int(category)]) # print ####################################
