def test_raise_errors(self):
# Only positive available
self.ratings = pd.DataFrame.from_records(
[("A000", "tt0112281", 1, "54654675")],
columns=["from_id", "to_id", "score", "timestamp"])
alg = ClassifierRecommender({'Plot': 'tfidf'}, SkKNN(), 0)
user_ratings = self.ratings.query('from_id == "A000"')
with self.assertRaises(OnlyPositiveItems):
alg.process_rated(user_ratings, self.movies_dir)
# Only negative available
self.ratings = pd.DataFrame.from_records(
[("A000", "tt0112281", -1, "54654675")],
columns=["from_id", "to_id", "score", "timestamp"])
alg = ClassifierRecommender({'Plot': 'tfidf'}, SkKNN(), 0)
user_ratings = self.ratings.query('from_id == "A000"')
with self.assertRaises(OnlyNegativeItems):
alg.process_rated(user_ratings, self.movies_dir)
# No Item avilable locally
self.ratings = pd.DataFrame.from_records(
[("A000", "non existent", 0.5, "54654675")],
columns=["from_id", "to_id", "score", "timestamp"])
alg = ClassifierRecommender({'Plot': 'tfidf'}, SkKNN(), 0)
user_ratings = self.ratings.query('from_id == "A000"')
with self.assertRaises(NoRatedItems):
alg.process_rated(user_ratings, self.movies_dir)
def setUp(self) -> None:
self.ratings = pd.DataFrame.from_records(
[("A000", "tt0112281", 0.5, "54654675"),
("A000", "tt0112302", 0.5, "54654675"),
("A001", "tt0114576", 0.8, "54654675"),
("A001", "tt0112896", -0.4, "54654675"),
("A000", "tt0112346", -0.5, "54654675"),
("A000", "tt0112453", -0.5, "54654675"),
("A002", "tt0112453", -0.2, "54654675"),
("A002", "tt0113497", 0.5, "54654675"),
("A003", "tt0112453", -0.8, "54654675")],
columns=["from_id", "to_id", "score", "timestamp"])
# tt0112281 is rated for A000 but let's suppose we want to know its rank
self.filter_list = ['tt0112281', 'tt0112760', 'tt0112896']
self.movies_dir = os.path.join(dir_test_files, 'complex_contents',
'movies_codified/')
# IMPORTANT! If a new classifier is added, just add it to this list to test it
self.classifiers_list = [
SkSVC(),
SkKNN(),
SkRandomForest(),
SkLogisticRegression(),
SkDecisionTree(),
SkGaussianProcess()
]
def test_pop_invalid_metric(self):
recsys = ContentBasedRS(
ClassifierRecommender({'Plot': 'tfidf'}, SkKNN(), threshold=3),
self.ratings_original, movies_dir)
# Tries to calc score predictions with a pure ranking algorithm
metric_list = [MAE()]
valid_metric = PredictionCalculator(self.split_list,
recsys).calc_predictions(
self.test_items_list,
metric_list)
score_truth = ScoresNeededMetric.score_truth_list
rank_truth = RankingNeededMetric.rank_truth_list
# The metric is excluded from the valid ones and nothing is calculated since
# there aren't any others
self.assertEqual(len(valid_metric), 0)
self.assertEqual(len(score_truth), 0)
self.assertEqual(len(rank_truth), 0)
# Tries to calc score predictions with a pure ranking algorithm but there are also
# other type of metrics
metric_ranking = NDCG()
metric_score = MAE()
metric_list = [metric_score, metric_ranking]
valid_metric = PredictionCalculator(self.split_list,
recsys).calc_predictions(
self.test_items_list,
metric_list)
score_truth = ScoresNeededMetric.score_truth_list
rank_truth = RankingNeededMetric.rank_truth_list
# The metric MAE is excluded from the valid ones but NDCG is valid so predictions
# for that metric (RankingNeededMetric) are calculated
self.assertIn(metric_ranking, valid_metric)
self.assertNotIn(metric_score, valid_metric)
self.assertEqual(len(score_truth), 0)
self.assertGreater(len(rank_truth), 0)
def test_calc_rank_content_based(self):
recsys = ContentBasedRS(
ClassifierRecommender({'Plot': 'tfidf'}, SkKNN(), threshold=3),
self.ratings_original, movies_dir)
# We just need a Metric of the RankingNeededMetric class to test
metric_list = [NDCG()]
valid_metric = PredictionCalculator(self.split_list,
recsys).calc_predictions(
self.test_items_list,
metric_list)
rank_truth = RankingNeededMetric.rank_truth_list
# We expect this to be empty, since there are no ScoresNeededMetric in the metric list
score_truth = ScoresNeededMetric.score_truth_list
self.assertEqual(valid_metric, metric_list)
self.assertGreater(len(rank_truth), 0)
self.assertEqual(len(score_truth), 0)