Different accuracy for LibSVM and scikit-learn

Question

For the same dataset and parameters, I get different accuracy for implementing LibSVM and scikit-learn SVM, although scikit-learn also uses LibSVM internally .

What did I miss?

LibSVM command line version:

 me@my-compyter :~/Libraries/libsvm-3.16$ ./svm-train -c 1 -g 0.07 heart_scale heart_scale.model optimization finished, #iter = 134 nu = 0.433785 obj = -101.855060, rho = 0.426412 nSV = 130, nBSV = 107 Total nSV = 130 me@my-compyter :~/Libraries/libsvm-3.16$ ./svm-predict heart_scale heart_scale.model heart_scale.result Accuracy = 86.6667% (234/270) (classification) 

Scikit-learn NuSVC Version:

 In [1]: from sklearn.datasets import load_svmlight_file In [2]: X_train, y_train = load_svmlight_file('heart_scale') In [3]: from sklearn import svm In [4]: clf = svm.NuSVC(gamma=0.07,verbose=True) In [5]: clf.fit(X_train,y_train) [LibSVM]* optimization finished, #iter = 118 C = 0.479830 obj = 9.722436, rho = -0.224096 nSV = 145, nBSV = 125 Total nSV = 145 Out[5]: NuSVC(cache_size=200, coef0=0.0, degree=3, gamma=0.07, kernel='rbf', max_iter=-1, nu=0.5, probability=False, shrinking=True, tol=0.001, verbose=True) In [6]: pred = clf.predict(X_train) In [7]: from sklearn.metrics import accuracy_score In [8]: accuracy_score(y_train, pred) Out[8]: 0.8481481481481481 

Scikit-learn SVC Version:

 In [1]: from sklearn.datasets import load_svmlight_file In [2]: X_train, y_train = load_svmlight_file('heart_scale') In [3]: from sklearn import svm In [4]: clf = svm.SVC(gamma=0.07,C=1, verbose=True) In [5]: clf.fit(X_train,y_train) [LibSVM]* optimization finished, #iter = 153 obj = -101.855059, rho = -0.426465 nSV = 130, nBSV = 107 Total nSV = 130 Out[5]: SVC(C=1, cache_size=200, class_weight=None, coef0=0.0, degree=3, gamma=0.07, kernel='rbf', max_iter=-1, probability=False, shrinking=True, tol=0.001, verbose=True) In [6]: pred = clf.predict(X_train) In [7]: from sklearn.metrics import accuracy_score In [8]: accuracy_score(y_train, pred) Out[8]: 0.8666666666666667 

Update

Update1: updated scikit-learn example from SVR to NuSVC, see ogrisel answer

Update2: added output for verbose=True

Update3: added version of SVC scikit-learn

So it looks like my problem is resolved. If I use SVC with C=1 and not NuSVC, I get the same results as libsvm, but can someone explain why NuSVC and SVC (C = 1) give different results, although they should do the same (see ogrisel answer)?

Answer 1

SVR is a regression model, not a classification model. svm-train -c 1 is the Nu-SVC model, available as the sklearn.svm.NuSVC class.