class testMatrixStatistics(unittest.TestCase):
def setUp(self):
random.seed(12345)
num_elems = 50 * 49 // 2
self.contents = random.sample(xrange(num_elems + 1), num_elems)
self.condensedMatrix = CondensedMatrix(self.contents)
def test_mean(self):
self.assertAlmostEquals(self.condensedMatrix.calculateMean(),
numpy.mean(self.contents)) #, delta = 1)
def test_variance(self):
self.assertAlmostEquals(self.condensedMatrix.calculateVariance(),
numpy.var(self.contents)) #, delta = 1)
def test_skewness(self):
self.assertAlmostEquals(self.condensedMatrix.calculateSkewness(
), -0.000733274016748) #scipy.stats.skew(self.contents))#d, delta = 1)
def test_kurtosis(self):
self.assertAlmostEquals(
self.condensedMatrix.calculateKurtosis(),
-1.20119577613) #scipy.stats.kurtosis(self.contents))#, delta = 1)
def test_max(self):
self.assertAlmostEquals(self.condensedMatrix.calculateMax(),
numpy.max(self.contents)) #, delta = 1)
def test_min(self):
self.assertAlmostEquals(self.condensedMatrix.calculateMin(),
numpy.min(self.contents)) #, delta = 1)
class testMatrixStatistics(unittest.TestCase):
def setUp(self):
random.seed(12345)
num_elems = 50*49/2;
self.contents = random.sample(xrange(num_elems+1),num_elems)
self.condensedMatrix = CondensedMatrix(self.contents)
def test_mean(self):
self.assertAlmostEquals(self.condensedMatrix.calculateMean(), numpy.mean(self.contents))#, delta = 1)
def test_variance(self):
self.assertAlmostEquals(self.condensedMatrix.calculateVariance(), numpy.var(self.contents))#, delta = 1)
def test_skewness(self):
self.assertAlmostEquals(self.condensedMatrix.calculateSkewness(), -0.000733274016748)#scipy.stats.skew(self.contents))#d, delta = 1)
def test_kurtosis(self):
self.assertAlmostEquals(self.condensedMatrix.calculateKurtosis(), -1.20119577613)#scipy.stats.kurtosis(self.contents))#, delta = 1)
def test_max(self):
self.assertAlmostEquals(self.condensedMatrix.calculateMax(), numpy.max(self.contents))#, delta = 1)
def test_min(self):
self.assertAlmostEquals(self.condensedMatrix.calculateMin(), numpy.min(self.contents))#, delta = 1)
def create_matrices(data, verbose=False):
"""
Creates all the matrices for the observations (datasets) and returns both.
"""
condensed_matrices = {}
all_observations = {}
for dataset_name in data.all_datasets:
dataset = data.all_datasets[dataset_name]
# Creating the matrix
observations = dataset_loading_2D(dataset,
data.scale_factor[dataset_name])
all_observations[dataset_name] = observations
condensed_matrix_data = distance.pdist(observations)
condensed_matrix = CondensedMatrix(condensed_matrix_data)
condensed_matrices[dataset_name] = condensed_matrix
if verbose:
print "Matrix for %s:" % dataset_name
print "-----------------------"
print "Max dist. = ", condensed_matrix.calculateMax()
print "Min dist. = ", condensed_matrix.calculateMin()
print "Mean dist. = ", condensed_matrix.calculateMean()
print "Variance = ", condensed_matrix.calculateVariance()
print "-----------------------\n"
return condensed_matrices, all_observations
def test_full_run(self):
condensed_matrix = CondensedMatrix(matrix)
cmax = condensed_matrix.calculateMax()
alg = RandomAlgorithm.RandomClusteringAlgorithm(condensed_matrix)
values = []
calculator = MeanMinimumDistanceCalculator(10)
for i in range(2,20):
clustering = alg.perform_clustering({
"max_num_of_clusters":-1,
"num_clusters":i
})
values.append( calculator.evaluate(clustering, condensed_matrix, 30))
self.assertTrue(max(values) < cmax)
def test_full_run(self):
condensed_matrix = CondensedMatrix(matrix)
cmax = condensed_matrix.calculateMax()
alg = RandomAlgorithm.RandomClusteringAlgorithm(condensed_matrix)
values = []
calculator = MeanMinimumDistanceCalculator(10)
for i in range(2, 20):
clustering = alg.perform_clustering({
"max_num_of_clusters": -1,
"num_clusters": i
})
values.append(calculator.evaluate(clustering, condensed_matrix,
30))
self.assertTrue(max(values) < cmax)
def create_matrices(data, verbose = False):
"""
Creates all the matrices for the observations (datasets) and returns both.
"""
condensed_matrices = {}
all_observations = {}
for dataset_name in data.all_datasets:
dataset = data.all_datasets[dataset_name]
# Creating the matrix
observations = dataset_loading_2D(dataset,data.scale_factor[dataset_name])
all_observations[dataset_name] = observations
condensed_matrix_data = distance.pdist(observations)
condensed_matrix = CondensedMatrix(condensed_matrix_data)
condensed_matrices[dataset_name] = condensed_matrix
if verbose:
print "Matrix for %s:"%dataset_name
print "-----------------------"
print "Max dist. = ",condensed_matrix.calculateMax()
print "Min dist. = ",condensed_matrix.calculateMin()
print "Mean dist. = ",condensed_matrix.calculateMean()
print "Variance = ",condensed_matrix.calculateVariance()
print "-----------------------\n"
return condensed_matrices, all_observations