clusterData: Cluster time series data.

In nlhuong/vistimeseq: Analysis for short time-series data

Description

Perform cluster assignment using hierarchical clustering and static or dynamic branch cutting. If 'subset' is specified the clsutering is performed on the subset of the data, and the rest of the rows are assigned based on the distances to the centroids of computed clusters.

Usage

clusterData(
  X,
  dist = "euclidean",
  dynamic = FALSE,
  hclust_params = list(),
  static_cut_params = list(h = 0.5),
  dynamic_cut_params = list(max_height = 0.9)
)

Arguments

X	a data matrix or data.frame where rows are time series.
dist	the distance metric for the dissimilarity used for clustering.
dynamic	whether dynamic branch cutting should be done for cluster assignment.
hclust_params	parameters for hclust function.
static_cut_params	parameters for assignClusterStatic.
dynamic_cut_params	parameters for assignClusterDynamic.

Value

a list with the hclust object, as well as clust_map and clust_centroids data.frames.

Examples

data("endoderm_small")
endoderm_small <- normalizeData(endoderm_small)
X <- SummarizedExperiment::assays(endoderm_small)$norm
clust_res <- clusterData(X)
head(clust_res$clust_centroids)
head(clust_res$clust_map)

nlhuong/vistimeseq documentation built on Sept. 4, 2021, 2:41 a.m.