brainspace.null_models.moran.MoranRandomization

class brainspace.null_models.moran.MoranRandomization(procedure='singleton', spectrum='nonzero', joint=False, n_rep=100, n_ring=1, tol=1e-10, random_state=None)

Moran spectral randomization.

Parameters:

• procedure ({'singleton, 'pair'}, optional) – Procedure to generate the random samples. Default is ‘singleton’.
• spectrum ({'all', 'nonzero'}, optional) – Eigenvalues/vectors to select. If ‘all’, recover all eigenvectors except one. Otherwise, select all except non-zero eigenvectors. Default is ‘nonzero’.
• joint (boolean, optional) – If True variables are randomized jointly. Otherwise, each variable is randomized separately. Default is False.
• n_rep (int, optional) – Number of randomizations. Default is 100.
• n_ring (int, optional) – Neighborhood size to build the weight matrix. Only used if user provides a surface mesh. Default is 1.
• tol (float, optional) – Minimum value for an eigenvalue to be considered non-zero. Default is 1e-10.
• random_state (int or None, optional) – Random state. Default is None.

Variables:

• mev (1D ndarray, shape (n_components,)) – Eigenvalues of the weight matrix in descending order.
• mem (2D ndarray, shape (n_vertices, n_components)) – Eigenvectors of the weight matrix in same order.

See also

SpinPermutations

__init__(procedure='singleton', spectrum='nonzero', joint=False, n_rep=100, n_ring=1, tol=1e-10, random_state=None)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__([procedure, spectrum, joint, …])	Initialize self.
fit(w)	Compute Moran eigenvectors map.
get_params([deep])	Get parameters for this estimator.
randomize(x)	Generate random samples from x.
set_params(**params)	Set the parameters of this estimator.

fit(w)

Compute Moran eigenvectors map.

Parameters:w (BSPolyData, ndarray or sparse matrix, shape = (n_verts, n_verts)) – Spatial weight matrix or surface. If surface, the weight matrix is built based on the inverse geodesic distance between each vertex and the vertices in its n_ring. Returns:self (object) – Returns self.

randomize(x)

Generate random samples from x.

Parameters:x (1D or 2D ndarray, shape = (n_verts,) or (n_verts, n_feat)) – Array of variables arranged in columns, where n_feat is the number of variables. Returns:output (ndarray, shape = (n_rep, n_verts, n_feat)) – Random samples. If n_feat == 1, shape = (n_rep, n_verts).