Refit Metric-Conductance Graph Low-Pass Regression

Reuses a fitted metric graph low-pass eigensystem to smooth new responses.

Usage

# S3 method for class 'metric.graph.lowpass.fit'
refit(
  object,
  y,
  per.column.gcv = FALSE,
  eta.grid = NULL,
  n.candidates = 40L,
  n.cores = 1L,
  block.size = NULL,
  verbose = FALSE,
  ...
)

Arguments

object

A "metric.graph.lowpass.fit" object.

y

Numeric vector or matrix with one row per graph vertex.

per.column.gcv

Logical. If TRUE, select eta independently for each response column using the cached eigenbasis.

eta.grid

Optional positive numeric eta grid for per-column GCV.

n.candidates

Number of eta candidates when eta.grid = NULL.

n.cores

Positive integer retained for compatibility. Response columns are always processed sequentially; values greater than one do not enable parallel execution. Per-column GCV results report n.cores.used = 1L.

block.size

Optional block size for fixed-eta multi-column refits.

verbose

Logical progress flag.

...

Additional arguments are not currently supported.

Value

A list of class "metric.graph.lowpass.refit".

See also

refit for supported fitted objects and migration.

Examples

adj <- list(2L, c(1L, 3L), c(2L, 4L), 3L)
lengths <- list(1, c(1, 1), c(1, 1), 1)
fit <- fit.metric.graph.lowpass(
  adj, lengths, y = 1:4, n.eigenpairs = 4L,
  eta.grid = c(0.1, 1), eigen.solver = "dense"
)
refit(fit, y = 4:1)
#> 
#> Metric-Conductance Graph Low-Pass Refit
#> ======================================
#> 
#> Responses: 1
#> Per-column GCV: no; eta used: 1.0000e-01