Skip to content

Conversation

@viettr
Copy link

@viettr viettr commented Jun 10, 2022

Hi Jacob,

this pull request extends trac in two ways. First it allows for classification and second the user can add additional non-compositional covariates. Another extension is the implementation of the two-stage procedure proposed in https://onlinelibrary.wiley.com/doi/abs/10.1111/biom.12995. The pull request also provides more vignettes and tests.

There are major changes in the following files:

  • trac.R (extension: classification, additional covariates including normalization scheme and rescaling of betas, input checking, probabilistic output)
  • cv_trac.R (extension: classification, additional covariates)
  • log_contrast.R (extension: classification, additional covariates)
  • cv_log_contrast.R (extension: classification, additional covariates)

Merge remote-tracking branch 'upstream/master'

# Conflicts:
#	DESCRIPTION
#	R/cv_trac.R
#	R/trac.R
#	man/trac.Rd
viettr and others added 26 commits June 8, 2021 06:20
Merge branch 'master' of https://github.com/viettr/trac

# Conflicts:
#	.Rbuildignore
#	.gitignore
#	DESCRIPTION
#	R/cv_trac.R
#	R/predict.R
#	R/trac.R
#	man/cv_trac.Rd
#	man/trac.Rd
#	tests/testthat.R
#	tests/testthat/test-trac.R
Merge remote-tracking branch 'upstream/master'

# Conflicts:
#	R/log_contrast.R
#	R/trac.R
Comment on lines +33 to +34
if (is.null(fit$w_additional_covariates)) {
fit$w_additional_covariates <- NULL
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

if NULL, then set it to NULL?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thats a good point! I tried to make it backwards compatible but this line is unnecessary.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants