Overview
- If the pairwise distances between a set of points is available the distances can be obtained as a square matrix using the SciPy function squareform().
- Given a set of points to get only the upper or lower triangular portion the pdist() function can be used.
Example
The bird dataset used in the hierarchical clustering example is used here.
# Example Python program that forms the pairwise
# distances between points as a square matrix
# using the SciPy function squareform()
import scipy.spatial.distance as dist
import numpy
# The data points
brain_weights_birds = numpy.array([[0.5, 0.01365], [0.45, 0.00725],
[0.3, 0.001], [1.8, 0.012],
[2, 0.022], [1, 0.065]
])
# Get the condensed distance matrix
inTriangularForm = dist.pdist(brain_weights_birds)
# Form the square matrix of pairwise distances
inSquareForm = dist.squareform(inTriangularForm)
print("Pairwise distances:")
print(inTriangularForm)
print("Pairwise distances as a square matrix:")
print(inSquareForm)Output
Pairwise distances:
[0.05040794 0.20039966 1.30000105 1.50002324 0.50262991 0.15013015
1.35000836 1.55007018 0.55302356 1.50004033 1.7001297 0.70291963
0.20024984 0.8017537 1.00092407]
Pairwise distances as square form:
[[0. 0.05040794 0.20039966 1.30000105 1.50002324 0.50262991]
[0.05040794 0. 0.15013015 1.35000836 1.55007018 0.55302356]
[0.20039966 0.15013015 0. 1.50004033 1.7001297 0.70291963]
[1.30000105 1.35000836 1.50004033 0. 0.20024984 0.8017537 ]
[1.50002324 1.55007018 1.7001297 0.20024984 0. 1.00092407]
[0.50262991 0.55302356 0.70291963 0.8017537 1.00092407 0. ]]