corelay.processor.affinity

Affinity (similarity) processors.

Classes

Affinity	Base class for Affinity (Similarity) classes.
RadialBasisFunction	Radial Basis Function affinity
SparseKNN	Sparse K-Nearest-Neighbors affinity

class corelay.processor.affinity.Affinity(*args, **kwargs)

Bases: Processor

Base class for Affinity (Similarity) classes.

Each subclass implements a __call__ function to compute its corresponding affinity matrix of some data.

class corelay.processor.affinity.RadialBasisFunction(*args, **kwargs)

Bases: Affinity

Radial Basis Function affinity

Parameters:

sigma (float) – RBF scale

function(data)

Compute Radial Basis Function affinity matrix.

Parameters:

data (numpy.ndarray) – Distance matrix used to compute affinity matrix.

Returns:

Dense RBF affinity matrix

Return type:

np.ndarray

class corelay.processor.affinity.SparseKNN(*args, **kwargs)

Bases: Affinity

Sparse K-Nearest-Neighbors affinity

Parameters:

• n_neighbors (int) – Number of neighbors to consider.

• symmetrix (bool) – If True, Affinity matrix is set to the mean of itself and itself transposed.

function(data)

Compute Sparse K-Nearest-Neighbors affinity matrix.

Parameters:

data (numpy.ndarray) – Distance matrix used to compute affinity matrix.

Returns:

Sparse CSR representation of KNN affinity matrix

Return type:

sp.csr_matrix