def test_redefine_cluster_with_map(self):
initial_cluster = Cluster(None,[1,3,4,7,8])
final_cluster_1 = Cluster(None,[0,1,4]) #-> elements [1,3,8] of initial cluster
final_cluster_2 = Cluster(None,[2,3]) #-> elements [4,7] of initial cluster
self.assertItemsEqual( [1,3,8],Refiner.redefine_cluster_with_map(initial_cluster, final_cluster_1).all_elements)
self.assertItemsEqual( [4,7],Refiner.redefine_cluster_with_map(initial_cluster, final_cluster_2).all_elements)
def __kmedoids_compression(self, clustering, matrix_handler):
"""
"""
representatives = []
for cluster in clustering.clusters:
# Guess 'correct' number of elements for this cluster
cluster_size = cluster.get_size()
expected_cluster_elements = cluster_size * (float(self.parameters["final_number_of_frames"]) / clustering.total_number_of_elements)
expected_cluster_elements = int(math.ceil(expected_cluster_elements))
remapped_matrix = get_submatrix(matrix_handler.distance_matrix, cluster.all_elements)
# Prepare and run kmedoids algorithm
kmedoids = KMedoidsAlgorithm(remapped_matrix)
# print "KMEDOIDS:: EXPECTED", expected_cluster_elements, cluster_size, clustering.total_number_of_elements, self.parameters["final_number_of_frames"]
new_clustering = kmedoids.perform_clustering({
"k": expected_cluster_elements,
"seeding_type": "EQUIDISTANT"
})
# print "NEW CLUSTERING SIZE clusters: %d elements: %d"%(len(new_clustering.clusters), new_clustering.total_number_of_elements)
# reverse the remapping and add it to representatives
remapped_representatives = new_clustering.get_medoids(remapped_matrix)
fake_cluster = Cluster(None, remapped_representatives)
representatives.extend(Refiner.redefine_cluster_with_map(cluster, fake_cluster).all_elements)
return representatives
def __kmedoids_compression(self, clustering, matrix_handler):
"""
"""
representatives = []
for cluster in clustering.clusters:
# Guess 'correct' number of elements for this cluster
cluster_size = cluster.get_size()
expected_cluster_elements = cluster_size * (
float(self.parameters["final_number_of_frames"]) /
clustering.total_number_of_elements)
expected_cluster_elements = int(
math.ceil(expected_cluster_elements))
remapped_matrix = get_submatrix(matrix_handler.distance_matrix,
cluster.all_elements)
# Prepare and run kmedoids algorithm
kmedoids = KMedoidsAlgorithm(remapped_matrix)
# print "KMEDOIDS:: EXPECTED", expected_cluster_elements, cluster_size, clustering.total_number_of_elements, self.parameters["final_number_of_frames"]
new_clustering = kmedoids.perform_clustering({
"k":
expected_cluster_elements,
"seeding_type":
"EQUIDISTANT"
})
# print "NEW CLUSTERING SIZE clusters: %d elements: %d"%(len(new_clustering.clusters), new_clustering.total_number_of_elements)
# reverse the remapping and add it to representatives
remapped_representatives = new_clustering.get_medoids(
remapped_matrix)
fake_cluster = Cluster(None, remapped_representatives)
representatives.extend(
Refiner.redefine_cluster_with_map(cluster,
fake_cluster).all_elements)
return representatives
def test_repartition_with_kmedoids(self):
Refiner.KMedoidsAlgorithmClass = KMedoidsAlgorithmStub
clustering = Refiner.repartition_with_kmedoids(Cluster(None,[1,3,4,7,8]), 0, None)
# We'll suppose that the order is always the same
self.assertItemsEqual( [1,3,8],clustering.clusters[0])
self.assertItemsEqual( [4,7],clustering.clusters[1])
