def test_most_similar(self):
sent_ind = IndexedList(SENTENCES)
sentences = IndexedLineDocument(CORPUS)
m = Average(W2V)
m.train(sentences)
o = m.sv.most_similar(positive=0)
self.assertEqual(45, o[0][0])
self.assertEqual(35, o[1][0])
o = m.sv.most_similar(positive=0, indexable=sentences)
self.assertEqual("Looks good and fits snug", o[0][0])
o = m.sv.most_similar(positive=0, indexable=sent_ind)
self.assertEqual("Looks good and fits snug".split(), o[0][0][0])
def _train(self):
self.sens = IndexedList(self.concepts)
print('training SIF...')
self.se = SIF(self.w2v_model)
self.se.train(self.sens)
def test__str(self):
target = "[\'the dog is good\', \"it's nice and comfy\"]"
self.assertEqual(target, str(IndexedList(self.list_a)))
def test_getitem_presplitted(self):
l = IndexedList(self.list_c, pre_splitted=True)
self.assertEqual(["the", "dog", "is", "good"],
self.il.__getitem__(0).words)
def test_init_multiple_splits(self):
with self.assertRaises(RuntimeError):
IndexedList(self.list_a, split_func=self.list_a, pre_splitted=True)
def test__len(self):
l = IndexedList(self.list_a)
self.assertEqual(2, len(l))
def test_init_dict(self):
tmp = {0: "hello there"}
with self.assertRaises(TypeError):
IndexedList(tmp)
def test_init_multiple_args(self):
with self.assertRaises(RuntimeError):
IndexedList(self.list_a, split=True, split_func=self.list_a)
def test_init_multiple_list(self):
l = IndexedList(self.list_a, self.list_b)
self.assertEqual(4, len(l))
def test_init_set(self):
l = IndexedList(self.set_a)
def test_init_list(self):
l = IndexedList(self.list_a)
def setUp(self):
self.list_a = ["the dog is good", "it's nice and comfy"]
self.list_b = ["lorem ipsum dolor", "si amet"]
self.list_c = [s.split() for s in self.list_a]
self.set_a = set(["hello there", "its a set"])
self.il = IndexedList(self.list_a, self.list_b, self.set_a, split=True)
class TestIndexedListFuncs(unittest.TestCase):
def setUp(self):
self.list_a = ["the dog is good", "it's nice and comfy"]
self.list_b = ["lorem ipsum dolor", "si amet"]
self.list_c = [s.split() for s in self.list_a]
self.set_a = set(["hello there", "its a set"])
self.il = IndexedList(self.list_a, self.list_b, self.set_a, split=True)
def test_init_list(self):
l = IndexedList(self.list_a)
def test_init_multiple_list(self):
l = IndexedList(self.list_a, self.list_b)
self.assertEqual(4, len(l))
def test_init_set(self):
l = IndexedList(self.set_a)
def test_init_dict(self):
tmp = {0: "hello there"}
with self.assertRaises(TypeError):
IndexedList(tmp)
def test_init_multiple_args(self):
with self.assertRaises(RuntimeError):
IndexedList(self.list_a, split=True, split_func=self.list_a)
def test_init_multiple_splits(self):
with self.assertRaises(RuntimeError):
IndexedList(self.list_a, split_func=self.list_a, pre_splitted=True)
def test__len(self):
l = IndexedList(self.list_a)
self.assertEqual(2, len(l))
def test__str(self):
target = "[\'the dog is good\', \"it's nice and comfy\"]"
self.assertEqual(target, str(IndexedList(self.list_a)))
def test_getitem(self):
self.assertEqual(["the", "dog", "is", "good"],
self.il.__getitem__(0).words)
self.assertEqual(0, self.il.__getitem__(0).index)
def test_getitem_presplitted(self):
l = IndexedList(self.list_c, pre_splitted=True)
self.assertEqual(["the", "dog", "is", "good"],
self.il.__getitem__(0).words)
def test_delitem(self):
self.il.__delitem__(0)
self.assertEqual(5, len(self.il))
def test_setitem(self):
self.il.__setitem__(0, "is it me?")
self.assertEqual(["is", "it", "me?"], self.il[0].words)
def test_setitem_wrong_dtype(self):
with self.assertRaises(TypeError):
self.il.__setitem__(0, ["is it me?"])
def test_append(self):
self.il.append("is it me?")
self.assertEqual(["is", "it", "me?"], self.il[-1].words)
def test_extend(self):
self.il.extend(self.list_a, self.list_b)
self.assertEqual(10, len(self.il))
def test_init(self):
_ = IndexedList(self.list_a)
def setUp(self):
self.list_a = ["the dog is good", "it's nice and comfy"]
self.list_b = [s.split() for s in self.list_a]
self.il = IndexedList(self.list_a, self.list_b)