ggplot_cv: Visualizing cross validation output

In MarianSchoen/DTD: Digital Tissue Deconvolution

Visualizing cross validation output

Description

Plot the regularization parameter lambda vs the resulting loss function value. The lambda with the minimal mean loss in cross validation is marked with a red dot. For an example see "Cross validation" in the package vignette 'browseVignettes("DTD")'

Usage

ggplot_cv(
  DTD.model,
  title = "",
  LAMBDA.TRANS.FUN = log2,
  upper.x.axis.info = "non-zero",
  x.lab = expression("log2(" * lambda * ")"),
  y.lab = "Loss function"
)

Arguments

DTD.model	: list as returned by train_deconvolution_model DTD_cv_lambda_R or DTD_cv_lambda_cxx
title	string, additionally title
LAMBDA.TRANS.FUN	function, will be applied to the lambda sequence, e.g. to get equidistance x ticks
upper.x.axis.info	string, either "non-zero" or "geometric-mean". Sets the information that will be printed on the upper x axis.
x.lab	string, used as x label on the plot
y.lab	string, used as y label on the plot

Value

a ggplot plot object

MarianSchoen/DTD documentation built on April 29, 2022, 1:59 p.m.