SVM can be used both for classification, which is SVC, and for prediction, or to become regression, which is SVR. sklearn also integrates the SVR class in the svm module. We also use a small example to illustrate how SVR works. X = [[0, 0], [1, 1]] y = [0.5, 1.5] clf = svm.SVR() clf.fit(X, y) result = clf.predict([2, 2]) print result