def test_naive_case_2(self):
# 2
# |
# 0 - 4 3
# |
# 1
points = [(0,0), (0,-1), (6,1), (6,0), (1,0)]
matrix = CondensedMatrix(pdist(points))
s_algo = SpectralClusteringAlgorithm(matrix, max_clusters = 2, sigma_sq = 0.5, use_k_medoids = False, verbose = True)
clusters = s_algo.perform_clustering({"k":2}).clusters
for c in clusters:
self.assertIn(c.all_elements, [[0,1,4],[2,3]])
def test_naive_case_2(self):
# 2
# |
# 0 - 4 3
# |
# 1
points = [(0,0), (0,-1), (6,1), (6,0), (1,0)]
matrix = CondensedMatrix(pdist(points))
s_algo = SpectralClusteringAlgorithm(matrix, max_clusters = 2, sigma_sq = 0.5, use_k_medoids = False, verbose = True)
clusters = s_algo.perform_clustering({"k":2}).clusters
for c in clusters:
self.assertIn(c.all_elements, [[0,1,4],[2,3]])
def test_naive_case_1(self):
# 1 5 8
# | | |
# 0 - 3 4 6 - 7
# | |
# 2 9
points = [(0, 0), (0, 1), (0, -1), (1, 0), (3, 0), (3, 1), (6, 0), (7, 0), (7, 1), (7, -1)]
matrix = CondensedMatrix(pdist(points))
s_algo = SpectralClusteringAlgorithm(matrix, sigma=20, max_clusters=3, use_k_medoids=False, verbose=True)
clusters = s_algo.perform_clustering({"k": 3}).clusters
# sometimes works, sometimes not (due to kmeans/medoids unstabilities)
for c in clusters:
self.assertIn(c.all_elements, [[0, 1, 2, 3], [6, 7, 8, 9], [4, 5]])
def test_naive_case_1(self):
# 1 5 8
# | | |
# 0 - 3 4 6 - 7
# | |
# 2 9
points = [(0,0),(0,1),(0,-1),(1,0),
(3,0),(3,1),
(6,0),(7,0),(7,1),(7,-1)]
matrix = CondensedMatrix(pdist(points))
s_algo = SpectralClusteringAlgorithm(matrix, sigma = 20, max_clusters = 3, use_k_medoids = False, verbose = True)
clusters = s_algo.perform_clustering({"k":3}).clusters
# sometimes works, sometimes not (due to kmeans/medoids unstabilities)
for c in clusters:
self.assertIn(c.all_elements, [[0, 1, 2, 3],[6, 7, 8, 9],[4, 5]])
Created on 14/01/2013
@author: victor
'''
import numpy
import time
from pyRMSD.condensedMatrix import CondensedMatrix
from pyproct.algorithms.spectral.spectralClusteringAlgorithm import SpectralClusteringAlgorithm
if __name__ == '__main__':
MAX_ELEMENTS = 2000
DATA_LEN = (MAX_ELEMENTS * (MAX_ELEMENTS-1))/2
numpy_matrix_data = numpy.abs(numpy.random.rand(DATA_LEN))
matrix = CondensedMatrix(numpy_matrix_data)
W = SpectralClusteringAlgorithm.calculate_adjacency_matrix(matrix,1)
W_tmp = numpy.array(W)
t1 = time.time()
L_PYTHON = SpectralClusteringAlgorithm.calculate_laplacian(W_tmp, matrix, "PYTHON")[0]
t2 = time.time()
print 'Calculations took %0.3f s' % (t2-t1)
W_tmp = numpy.array(W)
t1 = time.time()
L_NUMPY = SpectralClusteringAlgorithm.calculate_laplacian(W_tmp, matrix, "NUMPY")[0]
t2 = time.time()
print 'Calculations took %0.3f s' % (t2-t1)
W_tmp = numpy.array(W)
t1 = time.time()
Created on 14/01/2013
@author: victor
"""
import numpy
import time
from pyRMSD.condensedMatrix import CondensedMatrix
from pyproct.algorithms.spectral.spectralClusteringAlgorithm import SpectralClusteringAlgorithm
if __name__ == '__main__':
MAX_ELEMENTS = 2000
DATA_LEN = (MAX_ELEMENTS * (MAX_ELEMENTS-1))/2
numpy_matrix_data = numpy.abs(numpy.random.rand(DATA_LEN))
matrix = CondensedMatrix(numpy_matrix_data)
W = SpectralClusteringAlgorithm.calculate_adjacency_matrix(matrix,1)
W_tmp = numpy.array(W)
t1 = time.time()
L_PYTHON = SpectralClusteringAlgorithm.calculate_laplacian(W_tmp, matrix, "PYTHON")[0]
t2 = time.time()
print 'Calculations took %0.3f s' % (t2-t1)
W_tmp = numpy.array(W)
t1 = time.time()
L_NUMPY = SpectralClusteringAlgorithm.calculate_laplacian(W_tmp, matrix, "NUMPY")[0]
t2 = time.time()
print 'Calculations took %0.3f s' % (t2-t1)
W_tmp = numpy.array(W)
t1 = time.time()
