def test_prediction_pipeline(self):
"""Test pipeline prediction."""
train_data, _ = pipelines.extract_labeled_features(
self.test_data,
nvd_attributes=['project', 'description'],
nltk_feed_attributes=['description'])
clf = classifiers.NBClassifier().fit(train_data)
pred_data = [
'Sample project name prediction',
'Sample project name prediction',
'Sample project name prediction',
]
pred_pipeline = pipelines.get_prediction_pipeline(clf, )
n_candidates = 3
predictions = pred_pipeline.fit_predict(pred_data,
classifier__n=n_candidates,
classifier__sample=True)
self.assertIsNotNone(predictions)
self.assertEqual(predictions.shape[1], n_candidates)
self.assertEqual(predictions.shape[-1], 2) # candidate - proba
def test_evaluation(self):
"""Test evaluation of extracted features"""
test_data = _get_test_data()
featuresets, _ = pipelines.extract_labeled_features(
data=test_data,
attributes=['description'],
)
clf = classifiers.NBClassifier().fit(featuresets)
self.assertIsNotNone(clf)
# evaluation == 0.0
zero_labels = [None] * len(featuresets)
score = clf.evaluate(featuresets, zero_labels, sample=True)
self.assertIsNotNone(score)
self.assertEqual(score, 0.0)
score = classifiers.cross_validate(clf,
featuresets,
zero_labels,
sample=True)
self.assertIsNotNone(score)
self.assertEqual(score.mean, 0.0)
def _export_classifier():
"""Set up for unit tests by exporting classifier."""
raw_data = _get_test_data()
data, _ = pipelines.extract_labeled_features(
data=raw_data,
nvd_attributes=['project', 'description'],
nltk_feed_attributes=['description'])
classifier = classifiers.NBClassifier()
classifier = classifier.fit(data)
tmp_dir = tempfile.mkdtemp(prefix='test_export_')
pickle_path = classifier.export(export_dir=tmp_dir)
return pickle_path