aboutsummaryrefslogtreecommitdiff
path: root/tests
AgeCommit message (Collapse)AuthorFilesLines
2021-02-24Merge branch 'master' into saemixJohannes Ranke2-0/+10
2021-02-24Reset graphical parameters with on.exit()Johannes Ranke2-0/+10
plot.mixed.mmkin did not reset graphical parameters at all. The other plotting functions did not use on.exit, so this change should make the use of the plotting functions safer.
2021-02-06Reintroduce interface to saemixJohannes Ranke6-3009/+3248
Also after the upgrade from buster to bullseye of my local system, some test results for saemix have changed.
2021-02-06Increase test tolerance for parameter comparisonsv1.0.1Johannes Ranke1-4/+4
Platform dependence also revealed after upgrade to bullseye
2021-02-06Address issue revealed by additional tests on CRANJohannes Ranke3-17/+6
Remove tests relying on non-convergence of the FOMC fit to the FOCUS A dataset, as this is platform dependent. After the upgrade, the fit converges on this system as well, although neither ATLAS is used, nor R was configured disabling long doubles (these were the conditions under which the tests failed on CRAN).
2021-02-03Prepare for v1.0.0v1.0.0Johannes Ranke4-237/+0
- Improve authorship and copyright information - Prepare pkgdown config - Remove dependence on saemix as we need the development version which is not ready for CRAN - Temporarily remove saemix interface to check code coverage of the rest
2021-01-25Update visual testsJohannes Ranke3-1895/+1895
2021-01-25Change default ylab in plot.mkinfit, explicit ylab for plot.mmkinJohannes Ranke12-20/+20
See NEWS.md. Closes #12
2021-01-06Make saemix and corresponding tests optionalJohannes Ranke7-50/+71
Address release critical check and test issues
2020-12-19Dimethenamid data, normalisation optionsJohannes Ranke3-2/+11
2020-12-17Remove two skipped (on Travis) tests to reduce noiseJohannes Ranke3-217/+1
2020-12-10Tried to find a fast test for deSolve in saemixJohannes Ranke2-2/+7
But could not find one.
2020-12-10Add a test for saemix with deSolveJohannes Ranke2-8/+16
But skip the test as it takes too long to always run
2020-12-09Some test fixesJohannes Ranke3-6/+4
2020-12-09Add more tests and fix HS in saemJohannes Ranke4-1213/+1276
2020-12-09Fix a test I accidentally brokeJohannes Ranke1-1/+1
2020-12-09And moreJohannes Ranke3-1/+16
2020-12-09Any yet more testsJohannes Ranke4-3/+17
2020-12-09Update test output and reduce precision for travisJohannes Ranke4-27/+27
2020-12-09More testsJohannes Ranke8-2608/+1528
2020-12-09Try other mixed fits for vdiffr on travisJohannes Ranke4-4810/+1290
2020-12-09Test print method for mixed.mmkinJohannes Ranke3-2/+32
2020-12-09Make tests more robust, remove two redundant onesJohannes Ranke4-44/+22
2020-12-09Fix tests and rerunJohannes Ranke4-83/+88
2020-12-09Some more tests that may work on travis nowJohannes Ranke2-27/+28
2020-12-09Some more testsJohannes Ranke5-67/+194
2020-12-09Don't test on devel, skip failing vdiffr testsJohannes Ranke1-1/+1
I can test on R-devel locally for preparing releases
2020-12-08Test plots on R-release and cache packages againJohannes Ranke22-3137/+3137
2020-12-08mixed.mmkin and test coverageJohannes Ranke11-5/+10170
2020-12-02Possibility to specify random effects structuresJohannes Ranke1-6/+6
The default is pdDiag again, as we often have a small number of datasets in degradation kinetics.
2020-11-30Log-Cholesky parameterisation as default in nlme.mmkinJohannes Ranke1-9/+11
2020-11-19Update tests, improve CAKE_export testJohannes Ranke2-4/+3
2020-11-12Update test outputJohannes Ranke1-1/+2
2020-11-11Skip vdiffr tests on travis as they are not robustJohannes Ranke5-3/+6
As can be seen in the miniscule change seen on R-devel in the reference plot updated with this commit
2020-11-08Improve saem method, add summaryJohannes Ranke1-1/+1
Also make the endpoints function work for saem objects.
2020-11-07Create saem generic for fitting saemix modelsJohannes Ranke1-1/+1
The reasons for this decision were - Creating an saemix generic in the saemix package caused problems with roxygen, because functions like saemix.plot.xy were documented in their help files as S3 methods, although explicitly exported with @export - Creating an saemix generic in this package is possible, but would make it necessary to load samix with exclude = "saemix" in order to avoid overwriting the generic when loading saemix. - The return object of such an saemix generic in this package cannot be an S3 class with class attribute c("saemix.mmkin", "SaemixObject") similar to nlme.mmkin, as saemix returns an S4 class. - Extending the S4 class SaemixObject using simple inheritance to a class SaemixMmkinObject with additional slots did not work as expected. When the initialize method was left untouched, it prevented creation of an SaemixMmkinObject even if it was based on an initialised SaemixObject, as the initialize method seems to always be called by new(). This could potentially be circumvented by a coerce method. If an alternative initialize method was used, an SaemixMmkinObject could be created. However, the methods written for SaemixObjects only worked in some instances, either because they checked for the class, and not for class inheritance (like compare.saemix), or because the initialize method was called for some reason. Therefore, the idea of creating a derived S4 class was abandoned. - A side effect of this decision is that the introduction of the saem generic opens the possibility to use the same generic also for other backends like nlmixr with the SAEM algorithm.
2020-11-06Fast analytical solutions for saemix, update.mmkinJohannes Ranke4-162/+197
Also, use logit transformation for g and for solitary formation fractions, addressing #10.
2020-11-04Revert to use starting values for random effects in nlme.mmkinJohannes Ranke5-14/+4
2020-10-27Add summary method for nlme.mmkin objectsJohannes Ranke1-1/+1
Improve and update docs
2020-10-08Fix issues for releaseJohannes Ranke2-3/+3
2020-07-17mkinfit: Ignore unused components of state.iniJohannes Ranke2-1/+11
2020-07-15Improve testsJohannes Ranke1-44/+44
- Reduce significant digits for the objective function output in mkinfit(..., quiet = FALSE) as R and R-devel gave different output on my system - Add makefile target 'devtest' for testing with R-devel, in order to fix problems showing up with R-devel on Travis
2020-07-15Improve handling of (partially) failing fitsJohannes Ranke1-1/+1
2020-06-15Test plots on r-devel onlyJohannes Ranke19-222/+225
following the arguments of Xavier Robin https://github.com/r-lib/vdiffr/issues/86#issuecomment-636447231
2020-06-15Fix a bug, update for current vdiffrJohannes Ranke3-4/+1
Vignette FOCUS_L failed as I had introduced a bug in the handling of warnings. Current vdiffr only runs visual tests if R < 4.1.0, skipping r-devel for now, see https://github.com/r-lib/vdiffr/commit/630a29d013361fd63fea242f531e2db6aef37919
2020-05-29Improve handling of warnings, reorganize testsJohannes Ranke8-129/+146
2020-05-29Warn if standardized residuals are unlikely normalJohannes Ranke8-32/+51
This revealed a bug in the data returned in mkinfit$data in the case of the d_3 algorithm, which also affected the residual plot - the data from the direct fitting was not returned even if this was the better method.
2020-05-28use_of_ff = "max" for shorthand models like "SFO"Johannes Ranke6-25/+25
2020-05-26Use all cores per default, confint toleranceJohannes Ranke1-1/+1
Also, use more intelligent starting values for the variance of the random effects for saemix. While this does not appear to speed up the convergence, it shows where this variance is greatly reduced by using mixed-effects models as opposed to the separate independent fits.
2020-05-12Rerun tests and checks, CRAN commentsJohannes Ranke1-1/+1

Contact - Imprint