SciPy

scipy.spatial.distance.wminkowski

scipy.spatial.distance.wminkowski(u, v, p, w)[source]

Compute the weighted Minkowski distance between two 1-D arrays.

The weighted Minkowski distance between u and v, defined as

\[\left(\sum{(|w_i (u_i - v_i)|^p)}\right)^{1/p}.\]
Parameters:
u : (N,) array_like

Input array.

v : (N,) array_like

Input array.

p : int

The order of the norm of the difference \({||u-v||}_p\).

w : (N,) array_like

The weight vector.

Returns:
wminkowski : double

The weighted Minkowski distance between vectors u and v.

Notes

wminkowski is DEPRECATED. It implements a definition where weights are powered. It is recommended to use the weighted version of minkowski instead. This function will be removed in a future version of scipy.

Examples

>>> from scipy.spatial import distance
>>> distance.wminkowski([1, 0, 0], [0, 1, 0], 1, np.ones(3))
2.0
>>> distance.wminkowski([1, 0, 0], [0, 1, 0], 2, np.ones(3))
1.4142135623730951
>>> distance.wminkowski([1, 0, 0], [0, 1, 0], 3, np.ones(3))
1.2599210498948732
>>> distance.wminkowski([1, 1, 0], [0, 1, 0], 1, np.ones(3))
1.0
>>> distance.wminkowski([1, 1, 0], [0, 1, 0], 2, np.ones(3))
1.0
>>> distance.wminkowski([1, 1, 0], [0, 1, 0], 3, np.ones(3))
1.0

Previous topic

scipy.spatial.distance.sqeuclidean

Next topic

scipy.spatial.distance.dice