def test_exception_add_score_column(self):
# Test exception score column can't be converted into float
ri = RatingsImporter(source=self.raw_source,
from_id_column='user_id',
to_id_column='item_id',
score_column='stars')
with self.assertRaises(ValueError):
ri.add_score_column('review_title', 'text')
def test_add_score_column_w_score_processor(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4)
ratings = ri.import_ratings()
expected_columns = ['from_id', 'to_id', 'score']
result_columns = list(ratings.columns)
self.assertEqual(expected_columns, result_columns)
ratings_w_added_column = ri.add_score_column(
'review_title',
column_name='text_polarity',
score_processor=TextBlobSentimentAnalysis())
expected_columns = ['from_id', 'to_id', 'score', 'text_polarity']
result_columns = list(ratings_w_added_column.columns)
self.assertEqual(expected_columns, result_columns)
score_column_added = list(ratings['text_polarity'])
self.assertTrue(-1 <= score <= 1 for score in score_column_added)
def test_add_score_column(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4)
ratings = ri.import_ratings()
expected_columns = ['from_id', 'to_id', 'score']
result_columns = list(ratings.columns)
self.assertEqual(expected_columns, result_columns)
ratings_w_added_column = ri.add_score_column(
'stars', column_name='score_duplicate')
expected_columns = ['from_id', 'to_id', 'score', 'score_duplicate']
result_columns = list(ratings_w_added_column.columns)
self.assertEqual(expected_columns, result_columns)
score_column_added = list(ratings['score_duplicate'])
expected = [float(row['stars']) for row in self.raw_source_content]
self.assertEqual(expected, score_column_added)
