def templateLengthProcessData(data, start_medians, expected_cluster_length, ccore, **kwargs):
tolerance = kwargs.get('tolerance', 0.01)
metric = kwargs.get('metric', None)
itermax = kwargs.get('itermax', 200)
if isinstance(data, str):
sample = read_sample(data)
else:
sample = data
kmedians_instance = kmedians(sample, start_medians, tolerance, ccore, metric=metric, itermax=itermax)
kmedians_instance.process()
clusters = kmedians_instance.get_clusters()
medians = kmedians_instance.get_medians()
if itermax == 0:
assert clusters == []
assert start_medians == medians
return
obtained_cluster_sizes = [len(cluster) for cluster in clusters]
assert len(sample) == sum(obtained_cluster_sizes)
assert len(medians) == len(clusters)
if expected_cluster_length is not None:
obtained_cluster_sizes.sort()
expected_cluster_length.sort()
if obtained_cluster_sizes != expected_cluster_length:
print(obtained_cluster_sizes)
assert obtained_cluster_sizes == expected_cluster_length
def templateClusterAllocationOneDimensionData(self):
input_data = [ [random()] for i in range(10) ] + [ [random() + 3] for i in range(10) ] + [ [random() + 5] for i in range(10) ] + [ [random() + 8] for i in range(10) ];
kmedians_instance = kmedians(input_data, [ [0.0], [3.0], [5.0], [8.0] ], 0.025);
kmedians_instance.process();
clusters = kmedians_instance.get_clusters();
assert len(clusters) == 4;
for cluster in clusters:
assert len(cluster) == 10;
def template_clustering(start_centers, path, tolerance = 0.25):
sample = read_sample(path);
kmedians_instance = kmedians(sample, start_centers, tolerance);
(ticks, result) = timedcall(kmedians_instance.process);
clusters = kmedians_instance.get_clusters();
print("Sample: ", path, "\t\tExecution time: ", ticks, "\n");
draw_clusters(sample, clusters);
def templateLengthProcessData(self, path_to_file, start_centers, expected_cluster_length):
sample = read_sample(path_to_file);
kmedians_instance = kmedians(sample, start_centers, 0.025);
kmedians_instance.process();
clusters = kmedians_instance.get_clusters();
obtained_cluster_sizes = [len(cluster) for cluster in clusters];
assert len(sample) == sum(obtained_cluster_sizes);
obtained_cluster_sizes.sort();
expected_cluster_length.sort();
assert obtained_cluster_sizes == expected_cluster_length;
def templateClusterAllocationTheSameObjects(self, number_objects, number_clusters, ccore_flag = False):
value = random();
input_data = [ [value] ] * number_objects;
initial_centers = [];
for i in range(number_clusters):
initial_centers.append([ random() ]);
kmedians_instance = kmedians(input_data, initial_centers);
kmedians_instance.process();
clusters = kmedians_instance.get_clusters();
object_mark = [False] * number_objects;
allocated_number_objects = 0;
for cluster in clusters:
for index_object in cluster:
assert (object_mark[index_object] == False); # one object can be in only one cluster.
object_mark[index_object] = True;
allocated_number_objects += 1;
assert (number_objects == allocated_number_objects); # number of allocated objects should be the same.
def templateClusterAllocationTheSameObjects(number_objects, number_clusters, ccore_flag):
value = random()
input_data = [ [value] ] * number_objects
initial_centers = []
for i in range(number_clusters):
initial_centers.append([ random() ])
kmedians_instance = kmedians(input_data, initial_centers, ccore=ccore_flag)
kmedians_instance.process()
clusters = kmedians_instance.get_clusters()
object_mark = [False] * number_objects
allocated_number_objects = 0
for cluster in clusters:
for index_object in cluster:
assert (object_mark[index_object] is False) # one object can be in only one cluster.
object_mark[index_object] = True
allocated_number_objects += 1
assert (number_objects == allocated_number_objects) # number of allocated objects should be the same.
def testDifferentDimensions(self):
kmedians_instance = kmedians([ [0, 1, 5], [0, 2, 3] ], [ [0, 3] ]);
self.assertRaises(NameError, kmedians_instance.process);
def testCoreInterfaceIntInputData(self):
kmedians_instance = kmedians([ [1], [2], [3], [20], [21], [22] ], [ [2], [21] ], ccore=True)
kmedians_instance.process()
assert len(kmedians_instance.get_clusters()) == 2
def process_kmedians(sample):
instance = kmedians(sample, [ [random() + (multiplier * 5), random() + (multiplier + 5)] for multiplier in range(NUMBER_CLUSTERS) ])
(ticks, _) = timedcall(instance.process)
return ticks
def testDifferentDimensions(self):
kmedians_instance = kmedians([ [0, 1, 5], [0, 2, 3] ], [ [0, 3] ], ccore=False)
self.assertRaises(NameError, kmedians_instance.process)
def process_kmedians(sample):
instance = kmedians(sample, [ [random() + (multiplier * 5), random() + (multiplier + 5)] for multiplier in range(NUMBER_CLUSTERS) ])
(ticks, _) = timedcall(instance.process)
return ticks
def testTotalWCESimple4(self):
sample = [[0, 1, 5], [7, 8, 9], [0, 2, 3], [4, 5, 6]]
initial_medians = [[0, 3, 2], [4, 6, 5]]
kmedians_instance = kmedians(sample, initial_medians, ccore=False)
self.assertNotEqual(self, kmedians_instance.get_total_wce(), 16.0)
# Clean up our visual
ax.set(title='Elbow Plot',
xlabel='Number of Clusters',
ylabel='Total distance')
sns.despine(offset=5, trim=True)
# Let's now run a k-medians (instead of mean distance, let's calculate median distance)
# We can do this efficiently with the pyclustering library.
# If you don't have it installed, install it!
#python -m pip install pyclustering
from pyclustering.cluster.kmedians import kmedians
from pyclustering.cluster import cluster_visualizer
# Create instance of K-Medians algorithm.
initial_medians = [[0.0, 0.1], [2.5, 0.7], [3.5, 1.5]]
kmedians_instance = kmedians(xt, initial_medians)
# Run cluster analysis and obtain results.
kmedians_instance.process()
clusters = kmedians_instance.get_clusters()
medians = kmedians_instance.get_medians()
# How well did it do?
# Sum of metric errors is calculated using distance between point and its center
print(kmedians_instance.get_total_wce())
# Visualize clustering results.
visualizer = cluster_visualizer()
visualizer.append_clusters(clusters, xt)
#visualizer.append_cluster(initial_medians, marker='*', markersize=10)
visualizer.append_cluster(medians, marker='*', markersize=10, color='k')
visualizer.show()