Build the diffusion kernel from sparse adjacency components

Takes the CSC slots directly from R's dgCMatrix (igraph output).

Usage

rs_diffusion_kernel(
  i,
  p,
  x,
  n,
  kernel,
  kernel_params,
  normalised,
  strategy,
  k,
  node_names,
  verbose
)

Arguments

i

Integers. Row indices (0-based, from dgCMatrix@i)

p

Integers. Column pointers (from dgCMatrix@p)

x

Numeric. Values (from dgCMatrix@x)

n

Integer. Matrix dimension

kernel

String. Kernel type string. One of c("regularised_laplacian", "commute_time", "inverse_cosine", "pstep").

kernel_params

Named list. Contains the kernel-specific parameters.

normalised

Boolean. Use normalised Laplacian

strategy

String. One of "full" or "truncated". Shall the full Eigendecomposition of a truncated Eigendecomposition be applied on the data. The latter is useful for large graphs to reduce memory pressure.

k

Integer. Number of eigenvalues for truncated

node_names

String. Character vector of node names.

verbose

Boolean. Verbosity of the function.

Value

External pointer to DiffusionKernel