Wrapper function for Boost parameters

Usage

params_boost(
  boost_rate = 0.25,
  replace = FALSE,
  resolution = 1,
  n_iters = 20L,
  p_thresh = 1e-07,
  voter_thresh = 0.9,
  fast_cluster = FALSE,
  normalisation = list(),
  hvg = list(),
  pca = list(),
  knn = list(k = 0L),
  fast_cluster_params = list()
)

Arguments

boost_rate: Numeric. Boosting rate for the algorithm. Must be between 0 and 1. Defaults to 0.25.
replace: Boolean. Whether to use replacement during boosting. Defaults to FALSE.
resolution: Numeric. Resolution parameter for graph-based clustering. Higher values lead to more clusters. Defaults to 1.0.
n_iters: Integer. Number of iterations to run the algorithm. Defaults to 20L.
p_thresh: Numeric. P-value threshold for significance testing. Defaults to 1e-07.
voter_thresh: Numeric. Voter threshold across iterations. Proportion of iterations a cell must be assigned to a cluster to be considered a member. Must be between 0 and 1. Defaults to 0.9.
fast_cluster: Boolean. Shall fast Louvain clustering be applied, i.e., k-means clustering and use the centroids for kNN graph generation and Louvain clustering with then backpropagating the membership based on centroid proximity. Defaults to FALSE.
normalisation: List. Optional overrides for normalisation parameters. See params_norm_doublets_defaults() for available parameters: log_transform, mean_center, normalise_variance, target_size. See params_norm_doublets_defaults() for the available elements. Defaults to list().
hvg: List. Optional overrides for highly variable gene selection parameters. See params_hvg_defaults() for available parameters: min_gene_var_pctl, hvg_method, loess_span, clip_max. See params_hvg_defaults() for the available elements. Defaults to list().
pca: List. Optional overrides for PCA parameters. See params_pca_defaults() for available parameters: no_pcs, random_svd. See params_pca_defaults() for the available elements. Defaults to list().
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, extract_knn, m, ef_construction, ef_search, n_list and n_probe. Note: this function defaults to k = 0L (automatic neighbour detection). See params_knn_defaults() for the available elements. Defaults to list(k = 0L).
fast_cluster_params: List. Optional overrides for the fast clustering parameters. Only relevant if fast_cluster = TRUE. See params_fast_cluster_default() for available parameters: km_type, n_centroids, kmeans_iters and batch_size. See params_fast_cluster_default() for the available elements. Defaults to list().

Value

A named list with the following elements:

The elements of params_norm_doublets_defaults(), overridden by normalisation, spliced in at this position.
The elements of params_hvg_defaults(), overridden by hvg, spliced in at this position.
The elements of params_pca_defaults(), overridden by pca, spliced in at this position.
The elements of params_knn_defaults(), overridden by knn, spliced in at this position.
The elements of params_fast_cluster_default(), overridden by fast_cluster_params, spliced in at this position.
boost_rate - Numeric. Boosting rate for the algorithm. Must be between 0 and 1. Defaults to 0.25.
replace - Boolean. Whether to use replacement during boosting. Defaults to FALSE.
resolution - Numeric. Resolution parameter for graph-based clustering. Higher values lead to more clusters. Defaults to 1.0.
fast_cluster - Boolean. Shall fast Louvain clustering be applied, i.e., k-means clustering and use the centroids for kNN graph generation and Louvain clustering with then backpropagating the membership based on centroid proximity. Defaults to FALSE.
n_iters - Integer. Number of iterations to run the algorithm. Defaults to 20L.
p_thresh - Numeric. P-value threshold for significance testing. Defaults to 1e-07.
voter_thresh - Numeric. Voter threshold across iterations. Proportion of iterations a cell must be assigned to a cluster to be considered a member. Must be between 0 and 1. Defaults to 0.9.