Release CRAN version 1.5.0

[alexzwanenburg]

Prior to merge with master

• [x] increment package version
• [x] create release name
• [x] update NEWS.md
• [x] code check: run devtools::check(args=c("--no-examples", "--no-tests"), vignettes=FALSE), alternatively rcmdcheck::rcmdcheck(args=c("--no-examples", "--no-tests"))
• [x] test check: run devtools::test(): set options("testthat.progress.max_fails"=Inf) and options("Ncpus"=10)
• [x] create archives for testing backward compatibility against future versions: run tests/old_experiments/archiveExperiments.R
• [x] test backward compatibility against archived experiments: manually run tests/testthat/test-update_object.R
• [x] pre-compile vignettes: run vignettes/compile.R
• [x] code and vignette check: rcmdcheck::rcmdcheck(args=c("--no-examples", "--as-cran"))
• [x] check reverse dependencies
• [x] create pull request
• [x] check continuous integration tests
• [x] Build source tarball using devtools::build().
• [x] Run devtools::check_built(path=pkg, env_vars = list("_R_CHECK_DEPENDS_ONLY_" = TRUE)), with pkg the path to the tarball.
• [x] update cran_comments.md
• [x] merge dev branch into master

Post-merge

• [x] Check for merging errors.
  • [x] Run a post-merge code check: rcmdcheck::rcmdcheck(args=c("--no-examples", "--as-cran"))
  • [x] Run post-merge test suite, if required.
  • [x] Pre-compile vignettes, if required.
  • [x] Rebuild source tarball using devtools::build(), if required.

CRAN

• [x] Check CRAN-policies on https://cran.r-project.org/web/packages/policies.html
• [x] Upload source tarball to https://cran.r-project.org/submit.html
• [x] Check CRAN checks

Github release

• [x] prepare a GithHub release
  • [x] copy news from NEWS.md
  • [x] add source tarball as data

[alexzwanenburg]

Version name: Whole Whale

[alexzwanenburg]

The only issue directly tied to familiar is the following:

✖ | 1 1    406 | 0_plot_calibration [5480.6s]                                                                                                                                                                                                              

Warning (test-0_plot_calibration.R:40:1): 3. Plots for multinomial outcomes can be created for a dataset with some missing data and a naive model.
'fun.aggregate' is NULL, but found duplicate row/column combinations, so defaulting to length(). That is, the variables [ensemble_model_name, positive_class, data_set, fs_method, learner, type] used in 'formula' do not uniquely identify rows in the input 'data'. In such cases, 'fun.aggregate' is used to derive a single representative value for each combination in the output data.table, for example by summing or averaging (fun.aggregate=sum or fun.aggregate=mean, respectively). Check the resulting table for values larger than 1 to see which combinations were not unique. See ?dcast.data.table for more details.
Backtrace:
    ▆
 1. ├─base::do.call(...) at familiar/R/TestFunctions.R:4828:9
 2. ├─familiar (local) `<nnstndGF>`(object = `<fmlrCllc>`)
 3. └─familiar (local) `<nnstndGF>`(object = `<fmlrCllc>`) at familiar/R/PlotCalibration.R:147:5
 4.   └─familiar:::.plot_calibration_plot(...) at familiar/R/PlotCalibration.R:595:7
 5.     └─familiar:::.create_calibration_plot(...) at familiar/R/PlotCalibration.R:734:5
 6.       ├─data.table::dcast(data = linear_test, cast_formula, value.var = "value") at familiar/R/PlotCalibration.R:873:5
 7.       └─data.table::dcast.data.table(...)
 8.         └─data.table:::warningf(...)
 9.           └─data.table:::raise_condition(...)

Error (test-0_plot_calibration.R:40:1): 3. Plots for multinomial outcomes can be created for a dataset with some missing data and a naive model.
Error in ``[.data.table`(plot_layout_table, , `:=`(figure_id = .I, is_present = sapply(grobs, 
    gtable::is.gtable)))`: Supplied 7 items to be assigned to 6 items of column 'is_present'. If you wish to 'recycle' the RHS please use rep() to make this intent clear to readers of your code.
Backtrace:
    ▆
 1. ├─base::do.call(...) at familiar/R/TestFunctions.R:4828:9
 2. ├─familiar (local) `<nnstndGF>`(object = `<fmlrCllc>`)
 3. └─familiar (local) `<nnstndGF>`(object = `<fmlrCllc>`) at familiar/R/PlotCalibration.R:147:5
 4.   └─familiar:::.plot_calibration_plot(...) at familiar/R/PlotCalibration.R:595:7
 5.     └─familiar:::.update_plot_layout_table(...) at familiar/R/PlotCalibration.R:812:3
 6.       ├─...[] at familiar/R/PlotUtilities.R:989:3
 7.       └─data.table:::`[.data.table`(...) at familiar/R/PlotUtilities.R:989:3

I established that this was an issue with the testing code (and logic). This is fixed with version 2.0.0.