Constructor for MELD parameters

Usage

params_meld(
  beta = 60,
  offset = 0,
  order = 1,
  filter = c("heat", "laplacian"),
  chebyshev_order = 50L,
  lap_type = c("combinatorial", "normalised"),
  normalise_indicators = TRUE,
  knn = list()
)

Arguments

beta: Numeric. Smoothing strength; larger values produce smoother densities. Must be strictly positive. Defaults to 60.0.
offset: Numeric. Shift of the filter centre in the rescaled spectrum. Must be in [0, 1]. Defaults to 0.0.
order: Numeric. Filter falloff sharpness; larger values approach a square low-pass. Must be strictly positive. Defaults to 1.0.
filter: String. Filter family to use. One of c("heat", "laplacian"). Defaults to "heat".
chebyshev_order: Integer. Number of Chebyshev coefficients (polynomial terms). Must be >= 2. Defaults to 50L.
lap_type: String. Type of Laplacian to use for spectral filtering. One of c("combinatorial", "normalised"). Defaults to "combinatorial".
normalise_indicators: Boolean. If TRUE, each column of the indicator matrix is divided by its column sum before filtering, making cross-condition densities comparable regardless of cells-per-condition. Defaults to TRUE.
knn: List. Optional overrides for kNN parameters. See params_knn_defaults() for available parameters: k, knn_method, ann_dist, search_budget, n_trees, delta, diversify_prob, ef_budget, m, ef_construction, ef_search, n_list and n_probe. See params_knn_defaults() for the available elements. Defaults to list().

Value

A named list with the following elements:

The elements of params_knn_defaults(), overridden by knn, spliced in at this position.
beta - Numeric. Smoothing strength; larger values produce smoother densities. Must be strictly positive. Defaults to 60.0.
offset - Numeric. Shift of the filter centre in the rescaled spectrum. Must be in [0, 1]. Defaults to 0.0.
order - Numeric. Filter falloff sharpness; larger values approach a square low-pass. Must be strictly positive. Defaults to 1.0.
filter - String. Filter family to use. One of c("heat", "laplacian"). Defaults to "heat".
chebyshev_order - Integer. Number of Chebyshev coefficients (polynomial terms). Must be >= 2. Defaults to 50L.
lap_type - String. Type of Laplacian to use for spectral filtering. One of c("combinatorial", "normalised"). Defaults to "combinatorial".
normalise_indicators - Boolean. If TRUE, each column of the indicator matrix is divided by its column sum before filtering, making cross-condition densities comparable regardless of cells-per-condition. Defaults to TRUE.