neighbors.DistanceMetric

DistanceMetric class

This class provides a uniform interface to fast distance metric functions. The various metrics can be accessed via the get_metric class method and the metric string identifier (see below). For example, to use the Euclidean distance:

>>> dist = DistanceMetric.get_metric('euclidean')
>>> X = [[0, 1, 2],
         [3, 4, 5]])
>>> dist.pairwise(X)
array([[ 0.        ,  5.19615242],
       [ 5.19615242,  0.        ]])

Available Metrics The following lists the string metric identifiers and the associated distance metric classes:

Metrics intended for real-valued vector spaces:

Metrics intended for two-dimensional vector spaces: Note that the haversine distance metric requires data in the form of [latitude, longitude] and both inputs and outputs are in units of radians.

Metrics intended for integer-valued vector spaces: Though intended for integer-valued vectors, these are also valid metrics in the case of real-valued vectors.

Metrics intended for boolean-valued vector spaces: Any nonzero entry is evaluated to ?True?. In the listings below, the following abbreviations are used:

• N : number of dimensions
• NTT : number of dims in which both values are True
• NTF : number of dims in which the first value is True, second is False
• NFT : number of dims in which the first value is False, second is True
• NFF : number of dims in which both values are False
• NNEQ : number of non-equal dimensions, NNEQ = NTF + NFT
• NNZ : number of nonzero dimensions, NNZ = NTF + NFT + NTT

User-defined distance:

Here func is a function which takes two one-dimensional numpy arrays, and returns a distance. Note that in order to be used within the BallTree, the distance must be a true metric: i.e. it must satisfy the following properties

1. Non-negativity: d(x, y) >= 0
2. Identity: d(x, y) = 0 if and only if x == y
3. Symmetry: d(x, y) = d(y, x)
4. Triangle Inequality: d(x, y) + d(y, z) >= d(x, z)

Because of the Python object overhead involved in calling the python function, this will be fairly slow, but it will have the same scaling as other distances.

Methods

__init__()

x.__init__(...) initializes x; see help(type(x)) for signature

dist_to_rdist()

Convert the true distance to the reduced distance.

The reduced distance, defined for some metrics, is a computationally more efficent measure which preserves the rank of the true distance. For example, in the Euclidean distance metric, the reduced distance is the squared-euclidean distance.

get_metric()

Get the given distance metric from the string identifier.

See the docstring of DistanceMetric for a list of available metrics.

Parameters:	metric : string or class name The distance metric to use **kwargs : additional arguments will be passed to the requested metric

pairwise()

Compute the pairwise distances between X and Y

This is a convenience routine for the sake of testing. For many metrics, the utilities in scipy.spatial.distance.cdist and scipy.spatial.distance.pdist will be faster.

Parameters:	X : array_like Array of shape (Nx, D), representing Nx points in D dimensions. Y : array_like (optional) Array of shape (Ny, D), representing Ny points in D dimensions. If not specified, then Y=X. Returns : ??- : dist : ndarray The shape (Nx, Ny) array of pairwise distances between points in X and Y.

rdist_to_dist()

Convert the Reduced distance to the true distance.

The reduced distance, defined for some metrics, is a computationally more efficent measure which preserves the rank of the true distance. For example, in the Euclidean distance metric, the reduced distance is the squared-euclidean distance.