geolime.geostats.math.neighborhood

Data:

Callable	Callable type; Callable[[int], str] is a function of (int) -> str.
List	The central part of internal API.

Classes:

MinMaxPointsNeighborhood(dim, angles, …)
Neighborhood()

Functions:

anisotropic_distance([a])	Anisotropic distance between two points according to the anisotropy matrix a.
arg_sort_near_points(p, x[, distance])	Function to compute the indexes indicating the order in which the rows of X are close to p according to the distance distance (from small distance to large distance)
euclidian_distance(x, y)	Compute the Euclidean distance between two points.
rot_zyx(angles[, deg, clockwise])	Example of a convention (RGeoStats)
rotational_angles(angles)	Compute the angles in radians necessary for matrix rotation operation from geographic azimuth, dip and pitch angles defined in degrees
which_in_ball(x[, p, r, distance])	Function to return the indexes of the points which are in a (closed) ball in the Euclidean space.
which_in_parallelepiped(x[, p, r, distance])	Function to return the indexes of the points which are in a (closed) parallelepiped in the Euclidean space.
which_nearest_in_ball(x[, p, r, distance, …])	Function to obtain the N closest points of a set of points to the # center of a ball which are on that ball.

class geolime.geostats.math.neighborhood.MinMaxPointsNeighborhood(dim: int, angles: List[float], scales: List[float], min_n: int, max_n: int, convention: Callable = <function rot_zyx>)

Bases: geolime.geostats.math.neighborhood.Neighborhood

Methods:

select(x, p)

Neighborhood selection function.

select(x: numpy.ndarray, p: numpy.ndarray)

Neighborhood selection function.

Parameters

• x (np.ndarray) – A numpy matrix where each row is a vector in the Euclidean Space

• p (np.ndarray) – A numpy vector indicating the position (center of the ball). By default is a vector of zeros in dimension 3.

Returns

This function returns the indixes of the N rows of X which are the closest to the vector p according to the distance

Return type

np.ndarray

class geolime.geostats.math.neighborhood.Neighborhood

Bases: object

Methods:

select(x, p)

select(x, p)

geolime.geostats.math.neighborhood.arg_sort_near_points(p: numpy.ndarray, x: numpy.ndarray, distance: Callable = <function euclidian_distance>)

Function to compute the indexes indicating the order in which the rows of X are close to p according to the distance distance (from small distance to large distance)

Parameters

• p (np.ndarray) – Vector with respect to which calculate the distances.

• x (np.ndarray) – A numoy matrix with the vectors to be analyzed in its rows.

• distance (Callable) – A Distance function.

Returns

Returns the indexes of the rows of X according to which are closer to p according to the distance Distance. Note: this function could be easily vectorized in p using the Distance Matrix

Return type

np.ndarray

geolime.geostats.math.neighborhood.which_in_ball(x: numpy.ndarray, p: numpy.ndarray = array([0., 0., 0.]), r: float = 1.0, distance: Callable = <function euclidian_distance>)

Function to return the indexes of the points which are in a (closed) ball in the Euclidean space. THIS FUNCTION IS ONLY ADAPTED (AND OPTIMIZED) FOR DISTANCES INDUCED BY A NORM.

Parameters

• x (np.ndarray) – A numpy matrix whose rows represent vectors in the Euclidean space.

• p (np.ndarray) – Position vector indicating the center of the ball, by default a # vector of zeros in dimension 3.

• r (float) – Radius of the ball, by default r = 1.

• distance (Callable) – A distance function, by default euclidian_distance.

Returns

Return type

A numpy arrray

geolime.geostats.math.neighborhood.which_in_parallelepiped(x: numpy.ndarray, p: numpy.ndarray = array([0., 0., 0.]), r: float = 1.0, distance: Callable = <function euclidian_distance>)

Function to return the indexes of the points which are in a (closed) parallelepiped in the Euclidean space. THIS FUNCTION IS ONLY ADAPTED (AND OPTIMIZED) FOR DISTANCES INDUCED BY A NORM.

Parameters

• x (np.ndarray) – A numpy matrix whose rows represent vectors in the Euclidean space.

• p (np.ndarray) – Position vector indicating the center of the ball, by default a # vector of zeros in dimension 3.

• r (float) – Radius of the parallelepiped in each dimension, by default r = 1 and defines a cube.

• distance (Callable) – A distance function, by default euclidian_distance.

Returns

Return type

A numpy arrray

geolime.geostats.math.neighborhood.which_nearest_in_ball(x: numpy.ndarray, p: numpy.ndarray = array([0., 0., 0.]), r: float = 1.0, distance: Callable = <function euclidian_distance>, min_n: int = 1, max_n: int = 100)

Function to obtain the N closest points of a set of points to the # center of a ball which are on that ball. THIS FUNCTION IS ONLY ADAPTED (AND OPTIMIZED) FOR DISTANCES INDUCED BY A NORM.

Parameters

• x (np.ndarray) – A numpy matrix where each row is a vector in the Euclidean Space

• p (np.ndarray) – A numpy vector indicating the position (center of the ball). By default is a vector of zeros in dimension 3.

• r (float) – Radius of the ball. By default r=1.

• distance (Callable) – Distance function defining the ball. By default euclidian_distance.

• min_n (int) – Minimum number of points in the ball to be considered otherwise returns no points.

• max_n (int) – Maximum number of points in the ball to be considered (the N closest).

Returns

• A numpy array. This function returns the indixes of the N rows of X

• which are the closest to the vector p according to the distance

• Distance and which are in the ball of center p and radius r (according

• to that distance). If there is less points in the ball than N, the

• indexes of all those points are returned. There is no particular order

• in which the indexes are delivered.