def with_svm(): training_sentences, training_target_tags = return_tags_training_set() instance = SVMWithGridSearch(TAGS_TRAINING_SENTENCES, TAG_TARGETS) classifier = instance.classifier() print "Accuracy with 200 samples with SVM grid search %.3f" % ( classifier.score(TEST_SENTENCES, TEST_TARGET)) classifier = GridSearchCV(pipeline, tuned_parameters, verbose=1) classifier.fit(TAGS_TRAINING_SENTENCES, TAG_TARGETS) print "Accuracy with 200 samples with LDA %.3f" % (classifier.score( TEST_SENTENCES, TEST_TARGET)) print "Best score: %0.3f" % classifier.best_score_ print "Best parameters set:" best_parameters = classifier.best_estimator_.get_params() for param_name in sorted(parameters.keys()): print "\t%s: %r" % (param_name, best_parameters[param_name])
def with_svm(): training_sentences, training_target_tags = return_tags_training_set() instance = SVMWithGridSearch(TAGS_TRAINING_SENTENCES, TAG_TARGETS) classifier = instance.classifier() print "Accuracy with 200 samples with SVM grid search %.3f"%(classifier.score(TEST_SENTENCES, TEST_TARGET)) classifier= GridSearchCV(pipeline, tuned_parameters, verbose=1) classifier.fit(TAGS_TRAINING_SENTENCES, TAG_TARGETS) print "Accuracy with 200 samples with LDA %.3f"%(classifier.score(TEST_SENTENCES, TEST_TARGET)) print "Best score: %0.3f" % classifier.best_score_ print "Best parameters set:" best_parameters = classifier.best_estimator_.get_params() for param_name in sorted(parameters.keys()): print "\t%s: %r" % (param_name, best_parameters[param_name])
def svm_grid_search_classifier(self): print "\n Running {0} \n".format(inspect.stack()[0][3]) instance = SVMWithGridSearch(self.training_sentences, self.training_target_tags) classifier = instance.classifier() return classifier