hclust_performance_plot: Hierarchical clustering performance plot

In lisaschneider0509/lipidomeComparisonR:

Description

'hclust_performance_plot' takes a data frame and calculates which distance function and hclust function combination has the best performance.

Usage

hclust_performance_plot(
  input_df,
  dist_methods = c("euclidean", "manhattan"),
  hclust_methods = c("average", "single", "complete"),
  out_path = "none"
)

Arguments

input_df	data frame.
dist_methods	vector. Options: c("euclidean", "maximum", "manhattan", "canberra", "binary", "minkowski"). Default: c("euclidean", "maximum", "manhattan", "canberra", "minkowski")
hclust_methods	vector. Options: c("ward.D", "ward.D2", "single", "complete", "average", "mcquitty", "median", "centroid"). Default: c("average", "single", "complete")
out_path	string. Path to save the text output to. If specified, the output is saved in a .txt-file. If not specified, the output is printed th the device.

Details

This function calculated the performance for all combinations of the following distance and hclust functions distance functions = c("euclidean", "manhattan"), hclust functions = c("average", "single", "complete")). The results are printed in a table. The combination with the highest performance value will produce the best clustering.

lisaschneider0509/lipidomeComparisonR documentation built on Aug. 12, 2020, 12:52 a.m.