libpysal.weights.WSP¶
-
class
libpysal.weights.
WSP
(sparse, id_order=None)[source]¶ Thin W class for spreg.
- Parameters
- sparsesparse_matrix
NxN object from scipy.sparse
- id_orderlist
An ordered list of ids, assumed to match the ordering in 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
-
__init__
(self, sparse, id_order=None)[source]¶ Initialize self. See help(type(self)) for accurate signature.
Methods
__init__
(self, sparse[, id_order])Initialize self.
from_W
(W)Constructs a WSP object from the W’s sparse matrix
to_W
(self[, silence_warnings])Convert a pysal WSP object (thin weights matrix) to a pysal W object.
Attributes
Diagonal of \(W^{'}W + WW\).
s0 is defined as:
Trace of \(W^{'}W + WW\).
-
property
diagWtW_WW
¶ Diagonal of \(W^{'}W + WW\).
-
classmethod
from_W
(W)[source]¶ 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
s0
¶ s0 is defined as:
\[s0=\sum_i \sum_j w_{i,j}\]
-
to_W
(self, silence_warnings=False)[source]¶ Convert a pysal WSP object (thin weights matrix) to a pysal W object.
- Parameters
- selfWSP
PySAL sparse weights object
- silence_warningsboolean
Switch to turn off (default on) print statements for every observation with islands
- 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 \(W^{'}W + WW\).