def test_minmax_condensed(self):
condensed_matrix = CondensedDistanceMatrix([ 1.,
4.5, 8.5,
7.2, 4.5, 7.8,
6.7, 3.6,2.2, 2.0])
expected = (1,8.5)
self.assertEqual(condensed_matrix.get_minimum_and_maximum(),expected)
def test_normalize_condensed_matrix(self):
condensed = CondensedDistanceMatrix(
[1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6, 2.2, 2.])
expected = CondensedDistanceMatrix(
[0.0, 0.47, 1.0, 0.83, 0.47, 0.91, 0.76, 0.35, 0.16, 0.13])
minmax = condensed.get_minimum_and_maximum()
condensed.normalize(minmax[0], minmax[1])
for i in range(len(condensed.get_data())):
self.assertAlmostEqual(condensed.get_data()[i],
expected.get_data()[i], 2)
def test_save_condensed_matrix(self):
# with final spaces!
expected_matrix_string = """1.0 4.5 7.2 6.7
8.5 4.5 3.6
7.8 2.2
2.0
"""
condensed_matrix = CondensedDistanceMatrix(
[1.0, 4.5, 7.2, 6.7, 8.5, 4.5, 3.6, 7.8, 2.2, 2.0])
output = cStringIO.StringIO()
condensed_matrix.save(output)
self.assertEqual(expected_matrix_string, output.getvalue())
def test_load_condensed_matrix(self):
matrix_string = """1.0
4.5 8.5
7.2 4.5 7.8
6.7 3.6 2.2 2.0
"""
expected_matrix = CondensedDistanceMatrix([ 1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6,2.2, 2.])
input = cStringIO.StringIO(matrix_string)
loaded_matrix = load_condensed_matrix(input)
for i in range(len(expected_matrix.get_data())):
self.assertAlmostEqual(expected_matrix.get_data()[i],\
loaded_matrix.get_data()[i],3)
def test_load_condensed_matrix(self):
matrix_string = """1.0
4.5 8.5
7.2 4.5 7.8
6.7 3.6 2.2 2.0
"""
expected_matrix = CondensedDistanceMatrix(
[1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6, 2.2, 2.])
input = cStringIO.StringIO(matrix_string)
loaded_matrix = load_condensed_matrix(input)
for i in range(len(expected_matrix.get_data())):
self.assertAlmostEqual(expected_matrix.get_data()[i],\
loaded_matrix.get_data()[i],3)
def test_compare_condensed_matrixes(self):
cm1 = CondensedDistanceMatrix([1,2,3,4,5,6,7,8,9,0,1,2,3,4,5,6,7,8,9])
cm2 = CondensedDistanceMatrix([6,7,8,9,0,1,2,3,4,5,6,7,8,9,0,1,2,3,4])
cm3 = CondensedDistanceMatrix([1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1])
cm4 = CondensedDistanceMatrix([0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5])
result_1 = cm1.compare_with(cm2)
result_2 = cm1.compare_with(cm3)
result_3 = cm3.compare_with(cm4,1.,2.)
result_4 = cm3.compare_with(cm4,1.,1.)
self.assertEqual(result_1, (5.0, 0.0))
self.assertEqual(result_2, (3.8421052631578947, 2.6008734948643863))
self.assertEqual(result_3, (0., 0.))
self.assertEqual(result_4, (0.5, 0.))
def test_normalize_condensed_matrix(self):
condensed = CondensedDistanceMatrix([ 1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6,2.2, 2.])
expected = CondensedDistanceMatrix([0.0, 0.47, 1.0, 0.83, 0.47, 0.91, 0.76, 0.35, 0.16, 0.13])
minmax = condensed.get_minimum_and_maximum()
condensed.normalize(minmax[0], minmax[1])
for i in range(len(condensed.get_data())):
self.assertAlmostEqual(condensed.get_data()[i],expected.get_data()[i],2)
def test_save_condensed_matrix(self):
# with final spaces!
expected_matrix_string = """1.0 4.5 7.2 6.7
8.5 4.5 3.6
7.8 2.2
2.0
"""
condensed_matrix = CondensedDistanceMatrix([1.0, 4.5, 7.2, 6.7,
8.5, 4.5, 3.6,
7.8, 2.2,
2.0])
output = cStringIO.StringIO()
condensed_matrix.save(output)
self.assertEqual(expected_matrix_string,output.getvalue())
def test_equal(self):
cm1 = CondensedDistanceMatrix(
[1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
cm2 = CondensedDistanceMatrix(
[1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
cm3 = CondensedDistanceMatrix(
[6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4])
cm4 = CondensedDistanceMatrix([6, 7, 8, 9, 0, 1, 2, 3])
self.assertEqual(cm1 == cm2, True)
self.assertEqual(cm1 == cm3, False)
self.assertEqual(cm1 == cm4, False)
self.assertEqual(cm2 == cm3, False)
self.assertEqual(cm2 == cm4, False)
self.assertEqual(cm3 == cm4, False)
def test_gen_condensed_matrix(self):
obs = [(1, 1), (2, 1), (4, 5), (7, 7), (5, 7)]
## distance matrix
distance_matrix = CompleteDistanceMatrix(distance.cdist(obs, obs))
## lower distance matrix (wo diagonal)
expected_distance_condensed = CondensedDistanceMatrix(
distance.pdist(obs))
distance_condensed = complete_to_condensed(distance_matrix)
self.assertEqual(True,
distance_condensed == expected_distance_condensed)
def test_everything_for_rand_distrib(self):
distances = CondensedDistanceMatrix([0]*int((100*(100-1))/2))
distribution = [60,33,7]
rand_alg = FakeDistributionRandomClusteringAlgorithm(distances)
clusterization = rand_alg.perform_clustering(kwargs = {"distribution":distribution})
class_map = {0:0.,1:0.,2:0.}
for i in range(len(clusterization.clusters)):
class_map[i] += clusterization.clusters[i].get_size()
for i in range(len(clusterization.clusters)):
self.assertTrue(class_map[i] >distribution[i]-1 and class_map[i]<distribution[i]+1)
def test_data_sharing(self):
mylist = [ 1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6,2.2, 2.]
myarray = np.array([ 1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6,2.2, 2.])
mylistaarray = np.array(mylist)
condensed1 = CondensedDistanceMatrix(mylist)
condensed2 = CondensedDistanceMatrix(myarray)
condensed3 = CondensedDistanceMatrix(mylistaarray)
mylist[5] = 0.
self.assertEqual(False, mylist[5] == condensed1.get_data()[5])
myarray[5] = 0.
self.assertEqual(False, myarray[5] == condensed2.get_data()[5])
mylistaarray[5] = 0.
self.assertEqual(False, mylistaarray[5] == condensed3.get_data()[5])
mycontents = condensed3.get_data()
mycontents[5] = 0.
self.assertEqual(True, mycontents[5] == condensed3.get_data()[5] and\
condensed3.get_data()[5] == 0.)
def test_item_get(self):
condensed_matrix_1 = CondensedDistanceMatrix([1.0, 4.5,7.2,
8.5, 4.5,
7.8])
condensed_matrix_2 = CythonCondensedMatrix([.0]*6)
complete_matrix = [[0.0, 1.0, 4.5, 7.2],
[1.0, 0.0, 8.5, 4.5],
[4.5, 8.5, 0.0, 7.8],
[7.2, 4.5, 7.8, 0.0]]
row_len = condensed_matrix_1.row_length
for i in range(row_len):
for j in range(row_len):
condensed_matrix_2[i,j] = complete_matrix[i][j]
## The access for a complete and a condensed matrix is exactly the same
for i in range(row_len):
for j in range(row_len):
self.assertEquals(condensed_matrix_1[i,j],complete_matrix[i][j])
## And we can build a condensed matrix as a complete matrix
self.assertItemsEqual(condensed_matrix_1.get_data(), condensed_matrix_2.get_data())
def test_everything(self):
distances = CondensedDistanceMatrix([ 12.36931688, 5.83095189, 9.43398113, 12.52996409, 15.65247584,
17.4642492, 9.21954446, 4.47213595, 3.16227766, 4.47213595,
5.65685425, 5., 8.06225775, 11.18033989, 13.15294644,
3.16227766, 6.32455532, 8.24621125, 3.16227766, 5.09901951, 2. ])
rand_alg = RandomClusteringAlgorithm(distances)
for i in range(100): #@UnusedVariable
clusterization = rand_alg.perform_clustering(kwargs = {"max_num_of_clusters":5})
self.assertLess(len(clusterization.clusters),6)
absolutely_all_elements = []
for c in clusterization.clusters:
absolutely_all_elements.extend(c.all_elements)
absolutely_all_elements.sort()
self.assertItemsEqual(absolutely_all_elements, range(distances.row_length))
def test_validate_dimensions(self):
condensed_matrix_1 = CondensedDistanceMatrix(
[1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6, 2.2, 2.])
self.assertEqual(
True,
condensed_matrix_1._CondensedDistanceMatrix__validate_dimensions())
condensed_matrix_2 = CondensedDistanceMatrix(
[1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6])
self.assertEqual(
False,
condensed_matrix_2._CondensedDistanceMatrix__validate_dimensions())
def test_compare_condensed_matrixes(self):
cm1 = CondensedDistanceMatrix(
[1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
cm2 = CondensedDistanceMatrix(
[6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4])
cm3 = CondensedDistanceMatrix(
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1])
cm4 = CondensedDistanceMatrix([
0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5,
0.5, 0.5, 0.5, 0.5, 0.5, 0.5
])
result_1 = cm1.compare_with(cm2)
result_2 = cm1.compare_with(cm3)
result_3 = cm3.compare_with(cm4, 1., 2.)
result_4 = cm3.compare_with(cm4, 1., 1.)
self.assertEqual(result_1, (5.0, 0.0))
self.assertEqual(result_2, (3.8421052631578947, 2.6008734948643863))
self.assertEqual(result_3, (0., 0.))
self.assertEqual(result_4, (0.5, 0.))
def test_data_sharing(self):
mylist = [1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6, 2.2, 2.]
myarray = np.array([1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6, 2.2, 2.])
mylistaarray = np.array(mylist)
condensed1 = CondensedDistanceMatrix(mylist)
condensed2 = CondensedDistanceMatrix(myarray)
condensed3 = CondensedDistanceMatrix(mylistaarray)
mylist[5] = 0.
self.assertEqual(False, mylist[5] == condensed1.get_data()[5])
myarray[5] = 0.
self.assertEqual(False, myarray[5] == condensed2.get_data()[5])
mylistaarray[5] = 0.
self.assertEqual(False, mylistaarray[5] == condensed3.get_data()[5])
mycontents = condensed3.get_data()
mycontents[5] = 0.
self.assertEqual(True, mycontents[5] == condensed3.get_data()[5] and\
condensed3.get_data()[5] == 0.)
def test_validate_dimensions(self):
condensed_matrix_1 = CondensedDistanceMatrix([ 1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6,2.2, 2.])
self.assertEqual(True,condensed_matrix_1._CondensedDistanceMatrix__validate_dimensions())
condensed_matrix_2 = CondensedDistanceMatrix([ 1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6])
self.assertEqual(False,condensed_matrix_2._CondensedDistanceMatrix__validate_dimensions())
def test_minmax_condensed(self):
condensed_matrix = CondensedDistanceMatrix(
[1., 4.5, 8.5, 7.2, 4.5, 7.8, 6.7, 3.6, 2.2, 2.0])
expected = (1, 8.5)
self.assertEqual(condensed_matrix.get_minimum_and_maximum(), expected)