Increase epsilon for comparing in KL tests (piskvorky#1211)

gensim/test/test_similarity_metrics.py

@@ -188,7 +188,7 @@ def test_distributions(self):
         vec_2 = [(1, 0.1), (3, 0.8), (4, 0.1)]
         result = matutils.kullback_leibler(vec_2, vec_1, 8)
         expected = 0.55451775
-        self.assertAlmostEqual(expected, result)
+        self.assertAlmostEqual(expected, result, places=5)
 
         # KL is not symetric; vec1 compared with vec2 will contain log of zeros and return infinity
         vec_1 = [(2, 0.1), (3, 0.4), (4, 0.1), (5, 0.1), (1, 0.1), (7, 0.2)]
@@ -201,14 +201,14 @@ def test_distributions(self):
         vec_2 = csr_matrix([[1, 0.4], [0, 0.2], [2, 0.2]])
         result = matutils.kullback_leibler(vec_1, vec_2, 3)
         expected = 0.0894502
-        self.assertAlmostEqual(expected, result)
+        self.assertAlmostEqual(expected, result, places=5)
 
         # checking ndarray, list as inputs
         vec_1 = np.array([0.6, 0.1, 0.1, 0.2])
         vec_2 = [0.2, 0.2, 0.1, 0.5]
         result = matutils.kullback_leibler(vec_1, vec_2)
         expected = 0.40659450877
-        self.assertAlmostEqual(expected, result)
+        self.assertAlmostEqual(expected, result, places=5)
 
         # testing LDA distribution vectors
         np.random.seed(0)
@@ -217,7 +217,7 @@ def test_distributions(self):
         lda_vec2 = model[[(2, 2), (1, 3)]]
         result = matutils.kullback_leibler(lda_vec1, lda_vec2)
         expected = 4.283407e-12
-        self.assertAlmostEqual(expected, result)
+        self.assertAlmostEqual(expected, result, places=5)
 
 class TestJaccard(unittest.TestCase):
     def test_inputs(self):