k_center.landmarks
k_center.landmarks(X, k=15, eps=-1.0, seed=0, full_output=False, metric='euclidean')
Computes landmarks points for a point set or set of distance using the ‘maxmin’ method.
This function computes a prefix of the greedy permutation of X
using the ‘maxmin’ method; if k
is set, only the first k
points are found, otherwise if eps
is positive the prefix is expanded until a cover over X
of balls with radius eps
is found.
Parameters
Name | Type | Description | Default |
---|---|---|---|
X |
ArrayLike | (n x d) matrix of n points in d dimensions, a distance matrix, or a set of pairwise distances | required |
k |
Optional[int] | number of landmarks requested. Defaults to 15. | 15 |
eps |
Optional[float] | covering radius to stop finding landmarks at. If negative, uses k instead (default). | -1.0 |
seed |
int | index of the initial point to be the first landmark. Defaults to 0. | 0 |
full_output |
bool | whether to return insertion radii and predecessors. Defaults to False. | False |
metric |
str | metric distance to use. Ignored if X is a set of distances. See details. |
'euclidean' |
Returns
Type | Description |
---|---|
np.ndarray | Indices of the landmark points; if full_output = True , also returns a dictionary containing auxiliary information. |
Notes
- The first radius is always the diameter of the point set, which can be expensive to compute for high dimensions, so by default “inf” is used as the first covering radius
- If
metric = "euclidean"
andX
is a set of coordinates, the landmarks may be computed from the point set directly. For all other metrics, you must compute all pairwise distances first and supply that asX
. - If ‘k’ is specified an ‘eps’ is not (or it’s -1.0), then the procedure stops when ‘k’ landmarks are found. The converse is true if k = 0 and eps > 0. If both are specified, the both are used as stopping criteria for the procedure (whichever becomes true first).
- dictionary containing the insertion radii whose values ‘r’ yield a cover of ‘a’ when balls of radius ‘r’ are places at the landmark points and the predecessors.