aboutsummaryrefslogtreecommitdiff
path: root/tests/testthat
AgeCommit message (Collapse)AuthorFilesLines
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 Ranke3-10/+77
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 Ranke7-172/+219
2020-12-09Try other mixed fits for vdiffr on travisJohannes Ranke2-4/+22
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 Ranke2-3/+3
2020-12-08mixed.mmkin and test coverageJohannes Ranke7-5/+242
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 Ranke3-1/+4
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 Ranke2-0/+3
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 Ranke5-24/+24
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
2020-05-12Address solaris platform dependence revealed on CRANJohannes Ranke1-1/+2
2020-05-11Release to CRANv0.9.50.1Johannes Ranke1-1/+2
2020-05-11Add analytical solution for DFOP-SFOJohannes Ranke2-5/+20
This is about twice as fast as deSolve compiled in the case of FOCUS D
2020-05-11Analytical solutions for all SFO variantsJohannes Ranke5-31/+64
2020-05-11Don't run visual tests on r-devel for nowJohannes Ranke2-0/+3
https://github.com/r-lib/vdiffr/issues/86
2020-05-11Reduce precision of traced parameters, formattingJohannes Ranke1-121/+121
2020-05-11Avoid merge() and data.frame() in cost functionJohannes Ranke3-14/+16
also for deSolve and eigenvalue based solutions. This noticeably increases performance for these methods, see test.log and benchmark vignette.
2020-05-10Default to analytical for coupled models if availableJohannes Ranke2-7/+16
This revealed that transforming rates is necessary for fitting the analytical solution of the SFO-SFO model to the FOCUS D dataset. Benchmarks show that fitting coupled models with deSolve got a bit slower through the latest changes
2020-05-09Avoid the call to merge for analytical solutionsJohannes Ranke3-2/+13
This increases performance up to a factor of five!
2020-05-08Roxygenize, adapt testsJohannes Ranke7-106/+111
2020-05-07Another overhaul of analytical solutionsJohannes Ranke7-73/+90
Still in preparation for analytical solutions of coupled models
2020-05-06Change implementation of analytical solutionsJohannes Ranke1-1/+1
Preparing for symbolic solutions for more than one compound

Contact - Imprint