collapse_correlated_features: Collapse correlated features

In mikropml: User-Friendly R Package for Supervised Machine Learning Pipelines

Collapse correlated features

Description

Collapse correlated features

Usage

collapse_correlated_features(
  features,
  group_neg_corr = TRUE,
  corr_method = "spearman",
  corr_thresh = 1,
  progbar = NULL
)

Arguments

features	dataframe of features for machine learning
group_neg_corr	Whether to group negatively correlated features together (e.g. c(0,1) and c(1,0)).
corr_method	Correlation method. Options are the same as those supported by stats::cor: spearman, pearson, kendall. (default: spearman)
corr_thresh	group correlations above or equal to corr_thresh (range 0 to 1; default: 1).
progbar	optional progress bar (default: NULL)

Value

features where perfectly correlated ones are collapsed

Author(s)

Zena Lapp, [e-mail]

Examples

## Not run: 
collapse_correlated_features(mikropml::otu_small[, 2:ncol(otu_small)])

## End(Not run)

mikropml documentation built on Dec. 1, 2025, 9:08 a.m.