`scipy.spatial.distance._distance:canberra`

Signature

def canberra(u, v, w = None)

Compute the Canberra distance between two 1-D arrays.

The Canberra distance is defined as

d (u, v) = i \sum \frac{∣ u _{i} - v _{i} ∣}{∣ u _{i} ∣ + ∣ v _{i} ∣} .

u : (N,) array_like: Input array.
v : (N,) array_like: Input array.
w : (N,) array_like, optional: The weights for each value in u and v. Default is None, which gives each value a weight of 1.0

When u[i] and v[i] are 0 for given i, then the fraction 0/0 = 0 is used in the calculation.

from scipy.spatial import distance

✓

distance.canberra([1, 0, 0], [0, 1, 0])
distance.canberra([1, 1, 0], [0, 1, 0])

✗

np.float64(1.0)