def max_intercluster_distance(cls, clustering, matrix):
"""
Calculates d_max, the maximum inter clusters.
@param clustering: The clustering being checked.
@param matrix: The condensed matrix containing all distances.
@return: d_max' value
"""
distances = []
for i in range(len(clustering.clusters) - 1):
for j in range(i + 1, len(clustering.clusters)):
distances.extend(get_inter_cluster_distances(i, j, clustering.clusters, matrix))
return numpy.max(distances)
def max_intercluster_distance(cls, clustering, matrix):
"""
Calculates d_max, the maximum inter clusters.
@param clustering: The clustering being checked.
@param matrix: The condensed matrix containing all distances.
@return: d_max' value
"""
distances = []
for i in range(len(clustering.clusters) - 1):
for j in range(i + 1, len(clustering.clusters)):
distances.extend(
get_inter_cluster_distances(i, j, clustering.clusters,
matrix))
return numpy.max(distances)
def test_get_inter_cluster_distances(self):
matrix = CondensedMatrix(squared_CH_table1)
clusters = [Cluster(None, [1,2]),Cluster(None, [3,4]), Cluster(None, [5])]
numpy.testing.assert_equal(get_inter_cluster_distances(0, 1, clusters, matrix), [6.0, 13.0, 6.0, 17.0])
numpy.testing.assert_equal(get_inter_cluster_distances(1, 2, clusters, matrix), [15.0, 6.0])
numpy.testing.assert_equal(get_inter_cluster_distances(0, 2, clusters, matrix), [15.0, 21.0])