rf_clf.comps.summ: rf_clf.comps.summ

In shihuang047/crossRanger: Cross-Application of Ranger Models using microbiome data

rf_clf.comps.summ

Description

Runs standard random forests with oob estimation for classification of one level VS all other levels of one category in the datasets. The output includes a summary of rf models in the sub datasets, all important statistics for each of features, and plots.

Usage

rf_clf.comps.summ(
  df,
  f,
  comp_group,
  clr_transform = TRUE,
  nfolds = 3,
  verbose = FALSE,
  ntree = 5000,
  p_cutoff = 0.05,
  p.adj.method = "bonferroni",
  q_cutoff = 0.05,
  outdir = NULL
)

Arguments

df	Training data: a data.frame.
f	A factor in the metadata with at least two levels (groups).
comp_group	A string indicates the group in the f
clr_transform	A string indicates one class in the 'c_category' column of metadata.
nfolds	The number of folds in the cross validation.
verbose	A boolean value indicates if showing computation status and estimated runtime.
ntree	The number of trees.
p_cutoff	The cutoff of p values for features, the default value is 0.05.
p.adj.method	The p-value correction method, default is "bonferroni".
q_cutoff	The cutoff of q values for features, the default value is 0.05.
outdir	The outputh directory, default is "./".

Value

A list includes a summary of rf models in the sub datasets, all important statistics for each of features, and plots.

See Also

ranger

Examples

df <- data.frame(t(rmultinom(60, 300,c(.001,.6,.2,.3,.299))))
f=factor(c(rep("A", 15), rep("B", 15), rep("C", 15), rep("D", 15)))
comp_group="A"
rf_clf.comps.summ(df, f, comp_group, verbose=FALSE, ntree=500, p_cutoff=0.05,
                  p.adj.method = "bonferroni", q_cutoff=0.05, outdir=NULL)

shihuang047/crossRanger documentation built on Feb. 7, 2023, 10:03 p.m.