Change tests to unittest.TestCase, improve test coverage (#63)

* Fix indentations

* Change _add_special_tokens into staticmethod

* Change all tests of data module to unittest.TestCase

* Remove unused code

* Update README.md

* Add test for read_ui function

* Change tests of all datasets to unittest.TestCase

* Do not repeatedly showing split message in RatioSplit

* Change tests of eval_methods to unittest.TestCase

* Change tests of experiment to unittest.TestCase

* Change tests of metrics to unittest.TestCase

* Change tests of utils to unittest.TestCase

* Improve test coverage

* Fix indentations in test_trainset.py

cornac/eval_methods/ratio_split.py

@@ -93,8 +93,6 @@ def validate_size(val_size, test_size, num_ratings):
 
     def split(self):
         if self._split_ran:
-            if self.verbose:
-                print('Data is already split!')
             return
 
         if self.verbose:

tests/cornac/data/test_reader.py

@@ -10,6 +10,14 @@
 
 class TestReader(unittest.TestCase):
 
+    def test_read_ui(self):
+        data_file = './tests/data.txt'
+        triplets = reader.read_ui(data_file, value=2.0)
+
+        self.assertEqual(len(triplets), 30)
+        self.assertEqual(triplets[0][1], '93')
+        self.assertEqual(triplets[1][2], 2.0)
+
     def test_read_uir(self):
         data_file = './tests/data.txt'
         triplet_data = reader.read_uir(data_file)

tests/cornac/data/test_text.py

@@ -86,7 +86,6 @@ def setUp(self):
         self.module.build(self.id_map)
         self.token_ids = (self.module.vocab.tok2idx[tok] for tok in self.tokens)
 
-
     def test_init(self):
         self.assertCountEqual(self.module.vocab.idx2tok,
                               SPECIAL_TOKENS + self.tokens)
@@ -114,5 +113,12 @@ def test_batch_seq(self):
                                       np.asarray([[a, b, c, 0],
                                                   [c, b, e, c]]))
 
+        self.module.sequences = None
+        try:
+            self.module.batch_seq([0])
+        except ValueError:
+            assert True
+
+
 if __name__ == '__main__':
     unittest.main()

tests/cornac/data/test_trainset.py

@@ -5,15 +5,14 @@
 """
 
 import unittest
+import numpy as np
+import numpy.testing as npt
+from collections import OrderedDict
 
 from cornac.data import reader
 from cornac.data import TrainSet
 from cornac.data import MatrixTrainSet
 
-from collections import OrderedDict
-import numpy as np
-import numpy.testing as npt
-
 
 class TestTrainSet(unittest.TestCase):
 
@@ -132,6 +131,13 @@ def test_uir_tuple(self):
                                             global_ui_set=None,
                                             verbose=True)
 
+        self.assertEqual(len(train_set.uir_tuple), 3)
+        self.assertEqual(len(train_set.uir_tuple[0]), 10)
+
+        train_set.uir_tuple = None
+        self.assertEqual(len(train_set.uir_tuple[1]), 10)
+        self.assertEqual(len(train_set.uir_tuple[2]), 10)
+
         try:
             train_set.uir_tuple = ([], [])
         except ValueError:

tests/cornac/datasets/test_movielens.py

@@ -4,21 +4,28 @@
 @author: Quoc-Tuan Truong <tuantq.vnu@gmail.com>
 """
 
-import random, time
+import unittest
+import random
+import time
 from cornac.datasets import movielens
 
 
-def test_movielens_100k():
-    # only run data download tests 20% of the time to speed up frequent testing
-    random.seed(time.time())
-    if random.random() > 0.8:
-        ml_100k = movielens.load_100k()
-        assert len(ml_100k) == 100000
+class TestMovieLens(unittest.TestCase):
 
+    def test_movielens_100k(self):
+        # only run data download tests 20% of the time to speed up frequent testing
+        random.seed(time.time())
+        if random.random() > 0.8:
+            ml_100k = movielens.load_100k()
+            self.assertEqual(len(ml_100k), 100000)
 
-def test_movielens_1m():
-    # only run data download tests 20% of the time to speed up frequent testing
-    random.seed(time.time())
-    if random.random() > 0.8:
-        ml_1m = movielens.load_1m()
-        assert len(ml_1m) == 1000209
+    def test_movielens_1m(self):
+        # only run data download tests 20% of the time to speed up frequent testing
+        random.seed(time.time())
+        if random.random() > 0.8:
+            ml_1m = movielens.load_1m()
+            self.assertEqual(len(ml_1m), 1000209)
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/cornac/datasets/test_netflix.py

@@ -4,13 +4,21 @@
 @author: Quoc-Tuan Truong <tuantq.vnu@gmail.com>
 """
 
-import random, time
+import unittest
+import random
+import time
 from cornac.datasets import netflix
 
 
-def test_load_data_small():
-    # only run data download tests 20% of the time to speed up frequent testing
-    random.seed(time.time())
-    if random.random() > 0.8:
-        data = netflix.load_data_small()
-        assert len(data) == 607803
+class TestNetflix(unittest.TestCase):
+
+    def test_load_data_small(self):
+        # only run data download tests 20% of the time to speed up frequent testing
+        random.seed(time.time())
+        if random.random() > 0.8:
+            data = netflix.load_data_small()
+            self.assertEqual(len(data), 607803)
+
+
+if __name__ == '__main__':
+    unittest.main()

tests/cornac/datasets/test_tradesy.py

@@ -4,13 +4,21 @@
 @author: Quoc-Tuan Truong <tuantq.vnu@gmail.com>
 """
 
-import random, time
+import unittest
+import random
+import time
 from cornac.datasets import tradesy
 
 
-def test_load_data():
-    # only run data download tests 20% of the time to speed up frequent testing
-    random.seed(time.time())
-    if random.random() > 0.8:
-        data = tradesy.load_data()
-        assert len(data) == 394421
+class TestTradesy(unittest.TestCase):
+
+    def test_load_data(self):
+        # only run data download tests 20% of the time to speed up frequent testing
+        random.seed(time.time())
+        if random.random() > 0.8:
+            data = tradesy.load_data()
+            self.assertEqual(len(data), 394421)
+
+
+if __name__ == '__main__':
+    unittest.main()