search

rpytel1 / log-strategy

Project conducted for Seminar in Machine Learning for Software Engineering. Aim of our research was to explore possible directions of Deep Learning solutions for log detection in a snippet of code.
1 stars 1 forks source link

Classifier #14

Open jan-gerling opened 5 years ago

jan-gerling commented 5 years ago

SVM classifier (160k, 0.2 positive balance): https://drive.google.com/file/d/1FulPT0ytri2ODRG5Oa54-Ci_NmNagmUw/view?usp=sharing svm.SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0, decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf', max_iter=-1, probability=False, random_state=None, shrinking=True, tol=0.001, verbose=False)

-----------------------RESULTS----------------------- 20000 Accuracy: 0.946 Jaccard Index: 0.377 Precision: 0.46747676912080055 Recall: 0.6612740141557129 Average precision: 0.32588023964105517 -----------------------END RESULTS-----------------------

jan-gerling commented 5 years ago

SVM classifier (16k, 0.2 positive balance):

svm.SVC(kernel='rbf', C=1.0, cache_size=500, class_weight=None, coef0=0.0,
                               decision_function_shape='ovr', degree=3, max_iter=-1, probability=False,
                               random_state=None, shrinking=True, tol=0.001, verbose=False)
Best parameter for svm: {'C': 1, 'gamma': 0.01}

-----------------------RESULTS----------------------- 20000svm Accuracy: 0.9123 Jaccard Index: 0.33 Precision: 0.34657039711191334 Recall: 0.8736097067745198 Average precision: 0.30901726299766746 -----------------------END RESULTS-----------------------

jan-gerling commented 5 years ago

All classifier (svm and random forest with varying configs and training data set sizes): https://drive.google.com/file/d/1TReBbGlpujMPRR4vK9Ipr3tFp21H4Lsv/view?usp=sharing