def test_lsa(self):
# Assert LSA properties.
k = 100
lsa = vector.LSA(self.model, k)
self.assertEqual(lsa.model, self.model)
self.assertEqual(lsa.vectors, lsa.u)
self.assertEqual(set(lsa.terms), set(self.model.vector.keys()))
self.assertTrue(isinstance(lsa.u, dict))
self.assertTrue(isinstance(lsa.sigma, list))
self.assertTrue(isinstance(lsa.vt, list))
self.assertTrue(len(lsa.u), len(self.model))
self.assertTrue(len(lsa.sigma), len(self.model) - k)
self.assertTrue(len(lsa.vt), len(self.model) - k)
for document in self.model:
v = lsa.vectors[document.id]
self.assertTrue(isinstance(v, vector.Vector))
self.assertTrue(len(v) <= k)
print("pattern.vector.LSA")
class TestLSA(unittest.TestCase):
corpus = None
def setUp(self):
# Test spam corpus for reduction.
if self.__class__.corpus is None:
self.__class__.corpus = corpus(top=250)
self.corpus = self.__class__.corpus
random.seed(0)
def tearDown(self):
random.seed()
def test_lsa(self):
try:
import numpy
except ImportError, e:
print e
return
# Assert LSA properties.
k = 100
lsa = vector.LSA(self.corpus, k)
self.assertEqual(lsa.corpus, self.corpus)
self.assertEqual(lsa.vectors, lsa.u)
self.assertEqual(set(lsa.terms), set(self.corpus.vector.keys()))
self.assertTrue(isinstance(lsa.u, dict))
self.assertTrue(isinstance(lsa.sigma, list))
self.assertTrue(isinstance(lsa.vt, list))
self.assertTrue(len(lsa.u), len(self.corpus))
self.assertTrue(len(lsa.sigma), len(self.corpus) - k)
self.assertTrue(len(lsa.vt), len(self.corpus) - k)
for document in self.corpus:
v = lsa.vectors[document.id]
self.assertTrue(isinstance(v, vector.Vector))
self.assertTrue(len(v) == k)
print "pattern.vector.LSA"
class TestLSA(unittest.TestCase):
model = None
def setUp(self):
# Test spam model for reduction.
if self.__class__.model is None:
self.__class__.model = model(top=250)
self.model = self.__class__.model
random.seed(0)
def tearDown(self):
random.seed()
def test_lsa(self):
try:
import numpy
except ImportError, e:
print(e)
return
# Assert LSA properties.
k = 100
lsa = vector.LSA(self.model, k)
self.assertEqual(lsa.model, self.model)
self.assertEqual(lsa.vectors, lsa.u)
self.assertEqual(set(lsa.terms), set(self.model.vector.keys()))
self.assertTrue(isinstance(lsa.u, dict))
self.assertTrue(isinstance(lsa.sigma, list))
self.assertTrue(isinstance(lsa.vt, list))
self.assertTrue(len(lsa.u), len(self.model))
self.assertTrue(len(lsa.sigma), len(self.model)-k)
self.assertTrue(len(lsa.vt), len(self.model)-k)
for document in self.model:
v = lsa.vectors[document.id]
self.assertTrue(isinstance(v, vector.Vector))
self.assertTrue(len(v) <= k)
print("pattern.vector.LSA")
