Skip to content

initialize_quadtree

initialize_quadtree(P, max_depth=8, min_depth=1, graded=False, vmin=None, vmax=None)

Builds quadtree or octree from given point cloud.

Builds an adaptatively refined (optionally graded) quadtree for prototyping on adaptative grids. Keeps track of all parenthood and adjacency information so that traversals and differential quantities are easy to compute.

Parameters:

Name Type Description Default
P numpy double array

Matrix of point cloud coordinates

 required
max_depth int, optional (default 8)

max tree depth (min edge length will be bounding_box_length*2^(-max_depth))

 8
min_depth int, optional (default 1)

minimum tree depth (depth one is a single box)

 1
graded

Whether to ensure that adjacent quads only differ by one in depth or not (this is useful for numerical applications, not so much for others like position queries).

 False

Returns:

Name Type Description
C numpy double array

Matrix of cell centers
W numpy double array

Vector of cell half widths
CH numpy int array

Matrix of child indeces (-1 if leaf node)
PAR numpy int array

Vector of immediate parent indeces (to traverse upwards)
D numpy int array

Vector of tree depths
A scipy sparse.csr_matrix

Sparse node adjacency matrix, where a value of a in the (i,j) entry means that node j is to the a-th direction of i (a=1: left; a=2: right; a=3: bottom; a=4: top).
See Also

quadtree_children, in_quadtree.

Notes

This code is purposefully not optimized beyond asymptotics for simplicity in understanding its functionality and translating it to other programming languages beyond prototyping.

Examples:

# Random points
P = np.random.rand(100,2)
# Get quadtree for random points
C,W,CH,PAR,D,A = gpytoolbox.initialize_quadtree(P,graded=True,max_depth=8)
Source code in src/gpytoolbox/initialize_quadtree.py 
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
def initialize_quadtree(P,max_depth=8,min_depth=1,graded=False,vmin=None,vmax=None):
    """Builds quadtree or octree from given point cloud.

    Builds an adaptatively refined (optionally graded) quadtree for prototyping on adaptative grids. Keeps track of all parenthood and adjacency information so that traversals and differential quantities are easy to compute. 

    Parameters
    ----------
    P : numpy double array
        Matrix of point cloud coordinates
    max_depth : int, optional (default 8)
        max tree depth (min edge length will be bounding_box_length*2^(-max_depth))
    min_depth : int, optional (default 1)
        minimum tree depth (depth one is a single box)
    graded bool 
        Whether to ensure that adjacent quads only differ by one in depth or not (this is useful for numerical applications, not so much for others like position queries).

    Returns
    -------
    C : numpy double array 
        Matrix of cell centers
    W : numpy double array 
        Vector of cell half widths
    CH : numpy int array
        Matrix of child indeces (-1 if leaf node)
    PAR : numpy int array 
        Vector of immediate parent indeces (to traverse upwards)
    D : numpy int array
        Vector of tree depths
    A : scipy sparse.csr_matrix
        Sparse node adjacency matrix, where a value of a in the (i,j) entry means that node j is to the a-th direction of i (a=1: left;  a=2: right;  a=3: bottom;  a=4: top).

    See Also
    --------
    quadtree_children, in_quadtree.

    Notes
    -----
    This code is *purposefully* not optimized beyond asymptotics for simplicity in understanding its functionality and translating it to other programming languages beyond prototyping.

    Examples
    --------
    ```python
    # Random points
    P = np.random.rand(100,2)
    # Get quadtree for random points
    C,W,CH,PAR,D,A = gpytoolbox.initialize_quadtree(P,graded=True,max_depth=8)
    ```
    """

    # We start with a bounding box
    dim = P.shape[1]
    if (vmin is None):
        vmin = np.amin(P,axis=0)
    if (vmax is None):
        vmax = np.amax(P,axis=0)
    C = (vmin + vmax)/2.0
    C = C[None,:]
    #print(C)
    W = np.array([np.amax(vmax-vmin)])
    CH = np.tile(np.array([[-1]],dtype=int),(1,2**dim)) # for now it's leaf node
    D = np.array([1],dtype=int)
    A = csr_matrix((1,1))
    PAR = np.array([-1],dtype=int) # supreme Neanderthal ancestral node


    # Now, we will loop
    quad_ind = -1
    while True:
        quad_ind = quad_ind + 1
        if quad_ind>=C.shape[0]:
            break
        is_child = (CH[quad_ind,1]==-1)
        # Does this quad contain any point? (Or is it below our min depth)
        if ((D[quad_ind]<min_depth or np.any(is_in_quad(P,C[quad_ind,:],W[quad_ind]))) and D[quad_ind]<max_depth and is_child):
            # If it does, subdivide it
            C,W,CH,PAR,D,A = subdivide_quad(quad_ind,C,W,CH,PAR,D,A,graded)
    return C,W,CH,PAR,D,A