You are overwriting the svm with the first loop.
Try to change the name of the classifier, for instance:
for kernel in ('linear','poly', 'rbf'):
classifier_svm = svm.SVC(kernel=kernel, C=1)
classifier_svm.fit(trainingdata_without_labels, trainingdata_labels)
predicted_labels = classifier_svm.predict(testdata_without_labels)
print("testing success ratio with "+ kernel + "kernel :" + str(accuracy_score(testdata_labels, predicted_labels)))
Moreover, I think that what you try to do, finding the optimal kernel, its easier to solve using GridSearchCV:
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import classification_report
from sklearn.svm import SVC
tuned_parameters = [{'kernel': ['linear', 'poly', 'rbf'],
'C': [1]}
]
clf = GridSearchCV(SVC(), tuned_parameters, scoring='accuracy')
clf.fit(trainingdata_without_labels, trainingdata_labels)
print("Best parameters set found on development set:\n")
print(clf.best_params_)
print("\nGrid scores on development set:\n")
means = clf.cv_results_['mean_test_score']
stds = clf.cv_results_['std_test_score']
for mean, std, params in zip(means, stds, clf.cv_results_['params']):
print("%0.3f (+/-%0.03f) for %r"
% (mean, std * 2, params))
print("\nDetailed classification report:\n")
print("The model is trained on the full development set.")
print("The scores are computed on the full evaluation set.")
y_true, y_pred = testdata_labels, clf.predict(testdata_without_labels)
print(classification_report(y_true, y_pred))
With this code snippet you are going to train the model with the 3 kernels and also make a 5 fold cross validation. And finally compute the classification report (precission, recall, f1-score) of the testing variables. The final report should look like this (each row will be a class to predict in your data):
precision recall f1-score support
0 1.00 1.00 1.00 27
1 0.95 1.00 0.97 35
2 1.00 1.00 1.00 36
3 1.00 1.00 1.00 29
4 1.00 1.00 1.00 30
5 0.97 0.97 0.97 40
6 1.00 0.98 0.99 44
7 1.00 1.00 1.00 39
8 1.00 0.97 0.99 39
9 0.98 0.98 0.98 41
accuracy 0.99 360
macro avg 0.99 0.99 0.99 360
weighted avg 0.99 0.99 0.99 360
