libpysal.weights.WSP

class libpysal.weights.WSP(sparse, id_order=None, index=None)

Thin W class for spreg.

Parameters:

sparsescipy.sparse.{matrix-type}

NxN object from scipy.sparse

Examples

From GAL information

>>> import scipy.sparse
>>> from libpysal.weights import WSP
>>> rows = [0, 1, 1, 2, 2, 3]
>>> cols = [1, 0, 2, 1, 3, 3]
>>> weights =  [1, 0.75, 0.25, 0.9, 0.1, 1]
>>> sparse = scipy.sparse.csr_matrix((weights, (rows, cols)), shape=(4,4))
>>> w = WSP(sparse)
>>> w.s0
4.0
>>> w.trcWtW_WW
6.395
>>> w.n
4

Attributes:

nint

description

s0float

s0 is defined as:

trcWtW_WWfloat

Trace of \(W^{'}W + WW\).

__init__(sparse, id_order=None, index=None)

Methods

__init__(sparse[, id_order, index])
from_W(W)	Constructs a WSP object from the W's sparse matrix.
to_W([silence_warnings])	Convert a pysal WSP object (thin weights matrix) to a pysal W object.

Attributes

diagWtW_WW	Diagonal of \(W^{'}W + WW\).
id_order	An ordered list of ids, assumed to match the ordering in sparse.
s0	s0 is defined as:
trcWtW_WW	Trace of \(W^{'}W + WW\).

property diagWtW_WW

Diagonal of \(W^{'}W + WW\).

classmethod from_W(W)

Constructs a WSP object from the W’s sparse matrix.

Parameters:

Wlibpysal.weights.W

A PySAL weights object with a sparse form and ids.

Returns:

A WSP instance.

property id_order

An ordered list of ids, assumed to match the ordering in sparse.

property s0

s0 is defined as:

\[s0=\sum_i \sum_j w_{i,j}\]

to_W(silence_warnings=False)

Convert a pysal WSP object (thin weights matrix) to a pysal W object.

Parameters:

selfWSP

PySAL sparse weights object.

silence_warningsbool

Switch to True to turn off print statements for every observation with islands. Default is False, which does not silence warnings.

Returns:

wW

PySAL weights object.

Examples

>>> from libpysal.weights import lat2SW, WSP, WSP2W

Build a 10x10 scipy.sparse matrix for a rectangular 2x5 region of cells (rook contiguity), then construct a libpysal sparse weights object (self).

>>> sp = lat2SW(2, 5)
>>> self = WSP(sp)
>>> self.n
10
>>> print(self.sparse[0].todense())
[[0 1 0 0 0 1 0 0 0 0]]

Convert this sparse weights object to a standard PySAL weights object.

>>> w = WSP2W(self)
>>> w.n
10
>>> print(w.full()[0][0])
[0. 1. 0. 0. 0. 1. 0. 0. 0. 0.]

property trcWtW_WW

Trace of \(W^{'}W + WW\).