def test_nx_page_rank(self):
# Because graph based recommendation needs to have all items to predict in the ratings dataframe
recs_number = 1
graph = NXFullGraph(ratings)
alg = NXPageRank()
rs = GraphBasedRS(alg, graph)
# Test prediction and ranking with the Page Rank algorithm, prediction will raise exception
# since it's not a PredictionAlgorithm
with self.assertRaises(NotPredictionAlg):
rs.fit_predict('A000')
result_rank = rs.fit_rank('A000')
self.assertEqual(len(result_rank), 3)
# Test prediction and ranking with the Page Rank algorithm on specified items, prediction will raise exception
# since it's not a PredictionAlgorithm
with self.assertRaises(NotPredictionAlg):
rs.fit_predict('A000', filter_list=self.filter_list)
result_rank_filtered = rs.fit_rank('A000',
filter_list=self.filter_list)
self.assertEqual(len(result_rank_filtered), 2)
# Test top-n ranking with the Page Rank algorithm
result_rank_numbered = rs.fit_rank('A000', recs_number=recs_number)
self.assertEqual(len(result_rank_numbered), recs_number)
def test_fit_graph_w_testrating_methodology(self):
graph = NXFullGraph(ratings)
rs = GraphBasedRS(NXPageRank(), graph)
em = EvalModel(rs, KFoldPartitioning(), metric_list=[Precision()])
sys_result, users_result = em.fit()
self.assertIsInstance(sys_result, pd.DataFrame)
self.assertIsInstance(users_result, pd.DataFrame)
def test_multiple(self):
recs_number = 3
user_id_list = ['A000', 'A001']
# Test prediction and ranking with the Classifier Recommender algorithm
graph = NXFullGraph(ratings)
alg = NXPageRank()
rs = GraphBasedRS(alg, graph)
# Prediction
with self.assertRaises(NotPredictionAlg):
rs.multiple_fit_predict(user_id_list, filter_list=self.filter_list)
# Test ranking with the Classifier Recommender algorithm on specified items
result_rank_filtered = rs.multiple_fit_rank(
user_id_list, filter_list=self.filter_list)
self.assertEqual(set(user_id_list),
set(result_rank_filtered['from_id']))
for user in user_id_list:
self.assertEqual(
len(result_rank_filtered.query('from_id == @user')),
len(self.filter_list))
# Test top-n ranking with the Classifier Recommender algorithm
result_rank_numbered = rs.multiple_fit_rank(user_id_list,
recs_number=recs_number)
self.assertEqual(set(user_id_list),
set(result_rank_numbered['from_id']))
for user in user_id_list:
self.assertEqual(
len(result_rank_numbered.query('from_id == @user')),
recs_number)
def test_graph(self):
catalog = set(ratings.to_id)
users_dir = os.path.join(dir_test_files, 'complex_contents',
'users_codified/')
graph = NXFullGraph(
ratings,
user_contents_dir=users_dir,
item_contents_dir=items_dir,
item_exo_representation="dbpedia",
user_exo_representation='local',
item_exo_properties=['starring'],
user_exo_properties=['1'
] # It's the column in the users .DAT which
# identifies the gender
)
graph_rs = GraphBasedRS(NXPageRank(), graph)
em = EvalModel(graph_rs,
KFoldPartitioning(),
metric_list=[
Precision(relevant_threshold=3),
Recall(),
FMeasure(beta=1),
FMeasure(beta=2, sys_average='micro'),
MRR(),
Correlation('pearson'),
GiniIndex(),
DeltaGap({
'popular': 0.5,
'niche': 0.5
}),
PredictionCoverage(catalog),
PopProfileVsRecs(user_groups={
'popular': 0.5,
'niche': 0.5
},
out_dir='plots/'),
LongTailDistr('plots/', format='svg'),
PopRecsCorrelation('plots/')
],
verbose_predictions=True,
methodology=TestItemsMethodology())
em.fit()
def test_fit_graph_w_allitems_methodology(self):
graph = NXFullGraph(ratings)
rs = GraphBasedRS(NXPageRank(), graph)
items = set([
os.path.splitext(f)[0] for f in os.listdir(items_dir)
if os.path.isfile(os.path.join(items_dir, f)) and f.endswith('xz')
])
em = EvalModel(rs,
KFoldPartitioning(),
metric_list=[Precision()],
methodology=AllItemsMethodology(items))
sys_result, users_result = em.fit()
self.assertIsInstance(sys_result, pd.DataFrame)
self.assertIsInstance(users_result, pd.DataFrame)
def test_calc_rank_graph_based(self):
graph = NXFullGraph(self.ratings_original)
recsys = GraphBasedRS(NXPageRank(), graph)
# 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)