dba_clust_generate: dba_clust_generate

In manuclaeys/TimeSeriesBandits:

Description

Generate a clustering of freatures with dba and tsclust functions. Apply a global averaging method for time series under DTW (Petitjean, Ketterlin and Gancarski 2011) of numerical times series features.

Usage

dba_clust_generate(dt, type = "partitional",
  list_cluster_k = list_cluster_k, learn_size = (nrow(dt)),
  explanatory_variable = colnames(dt), print_time = FALSE,
  arm_for_learn = "all", print_graph = "all")

Arguments

dt	A matrix, json or data frame where each row is a time series, or a list where each element is a time series
type	What type of clustering method to use: "partitional", "hierarchical", "tadpole" or "fuzzy" (optional),
list_cluster_k	Number of desired clusters. It can be a numeric vector with different values.
learn_size	number of items dedicated to the learnset (step 1) ,
explanatory_variable	list of covariates (optional),
print_time	computational time
arm_for_learn	arm dedicated to the learnset (step 1) (optional),
print_graph	print clusters and controid c("all", "centroid","cluster","none")

Value

An object of class TSClusters-class

Examples

dba_clust_generate(dt = data.train[1:1000,listInteger],  list_cluster_k = list_cluster_k , print_graph = "none")

manuclaeys/TimeSeriesBandits documentation built on March 16, 2020, 3:22 p.m.