Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2023-11-23 | fix: 'R/illparms.R': An ill-defined random effect is now also found if there ↵ | Johannes Ranke | 1 | -0/+27 | |
is only one random effect in the model. Also add a test for this. | |||||
2023-08-17 | Update tests without any openblas variant | Johannes Ranke | 5 | -663/+662 | |
This even a bit faster. | |||||
2023-08-17 | Update platform specific tests on desktop | Johannes Ranke | 5 | -662/+663 | |
2023-08-17 | Point to the dmta pathway vignette in the dmta test | Johannes Ranke | 1 | -2/+5 | |
2023-08-04 | Mesotrione data and vignette | Johannes Ranke | 5 | -663/+662 | |
The vignette illustrates pH dependent degradation (covariate modelling) with some detail for the parent compound. Also, a bug in the saem method of the illparms function was fixed, which prevented to find ill-defined parameters in cases where e.g. slopes of covariate models have a negative estimate. | |||||
2023-04-16 | Skip another test on CRAN to save CPU cycles | Johannes Ranke | 1 | -0/+2 | |
2023-04-16 | Skip the Eigenvalue based saem test | Johannes Ranke | 1 | -5/+6 | |
It took too long, and this solution method is not recommended. We are saving 140 seconds here on my system, so more time is saved on travis. | |||||
2023-04-15 | Increase test coverage | Johannes Ranke | 2 | -1/+52 | |
Also, using mkin analytical solutions for more than one observed variable is not supported (but could be if out_values would be reordered). | |||||
2023-04-15 | Fix logic in test | Johannes Ranke | 1 | -1/+1 | |
2023-04-15 | Fix platform dependence in test | Johannes Ranke | 1 | -1/+5 | |
2023-04-15 | Make predefined symbols safer | Johannes Ranke | 2 | -0/+54 | |
We still need to create a parallel processing cluster _after_ creating a compiled model that is moved to a user defined location, at least I did not find another way to make it work. This is not a problem with parallel processing without a cluster, which is not available on Windows. | |||||
2023-04-15 | Move preprocessing of DMTA data to the test setup | Johannes Ranke | 2 | -11/+12 | |
2023-04-01 | Update tests on 5950x | Johannes Ranke | 4 | -121/+121 | |
Update platform dependent test plots | |||||
2023-03-22 | Fix typo in mkinerrmin | Johannes Ranke | 4 | -121/+121 | |
Thanks to Sebastian Meyer for spotting it. | |||||
2023-02-13 | Merge branch 'main' into custom_lsoda_call | Johannes Ranke | 25 | -401/+344 | |
2022-12-06 | Some parplot improvements | Johannes Ranke | 3 | -90/+85 | |
llquant argument, improved legend text, tests | |||||
2022-12-01 | Possibility to manually specify no_random_effects in mhmkin | Johannes Ranke | 4 | -39/+43 | |
2022-11-29 | Complete starting values in summary for saem.mmkin fits | Johannes Ranke | 5 | -25/+24 | |
Also update tests to the changes in mhmkin (see NEWS) | |||||
2022-11-19 | Confirm which test fail on R-develv1.2.1 | Johannes Ranke | 1 | -1/+2 | |
2022-11-18 | Move two saem fits from setup script to skipped tests | Johannes Ranke | 5 | -12/+18 | |
Save winbuilder/CRAN check time... | |||||
2022-11-18 | Use cluster with two cores to reduce windows check time | Johannes Ranke | 2 | -8/+16 | |
2022-11-18 | Make mixed model test data permanent to ensure reproducibility | Johannes Ranke | 17 | -245/+160 | |
To ensure that tests on different platforms work on the same data, the mixed modelling test data previosly generated in tests/testthat/setup_script.R were generated once using the script in inst/dataset/generation/ds_mixed.R, and are now distributed with the package. | |||||
2022-11-18 | Round parameters with signif() before printing | Johannes Ranke | 2 | -5/+10 | |
This will hopefully make the test pass on all relevant platforms. | |||||
2022-11-18 | Pass error.init to saemix_model, show in parplot | Johannes Ranke | 3 | -4/+8 | |
Due to an oversight, error.init was not really passed to saemix_model in saem.mmkin. The new initial values were reverted to c(1, 1), in order to avoid changing the test results. Initial values for error model parameters are now shown in parplot.multistart. | |||||
2022-11-18 | Avoid ill-determined fit, it is platform dependent | Johannes Ranke | 2 | -12/+11 | |
The previous test lead to an error on the Fedora test systems (gcc and clang) on CRAN: ... Results have changed from known value recorded in 'print_dfop_saemix_1.txt'. old[13:24] vs new[13:24] "" "Fitted parameters:" " estimate lower upper" - "parent_0 100.09 98.94 101.25" + "parent_0 100.17 99.01 101.33" - "log_k1 -2.68 -2.91 -2.45" + "log_k1 -2.66 -2.89 -2.43" "log_k2 -4.12 -4.24 -4.00" - "g_qlogis -0.41 -0.63 -0.20" + "g_qlogis -0.43 -0.65 -0.21" - "a.1 0.91 0.67 1.15" + "a.1 0.91 0.67 1.16" "b.1 0.05 0.04 0.06" - "SD.log_k1 0.36 0.21 0.50" + "SD.log_k1 0.35 0.21 0.50" and 2 more ... | |||||
2022-11-18 | Typo that did not affect test results | Johannes Ranke | 1 | -1/+1 | |
2022-11-16 | Reduce testing on CRAN serversv1.2.0 | Johannes Ranke | 3 | -5/+7 | |
2022-11-16 | Address platform dependence that surfaced on winbuilder | Johannes Ranke | 1 | -1/+6 | |
2022-11-16 | We get about 25% performance gain | Johannes Ranke | 1 | -1/+2 | |
with the custom lsoda call, avoiding repeated getNativeSymbolInfo calls. It's just that we should not be calling foreign functions from different packages, because the may change without notice. Using getNativeSymbolInfo for "call_lsoda" avoids the CRAN note, and a similar call could probably be used for "unlock_solver", avoiding the NOTE in checks for cran, but we should not do this in a CRAN package. | |||||
2022-11-14 | Automatic starting parameters for saem.mmkin | Johannes Ranke | 8 | -1341/+1343 | |
For the case of mkin transformations. This gives faster convergence, and appears to avoid problems with numeric ODE solutions | |||||
2022-11-08 | Show DFOP g equivalent of SFORB in summary, test | Johannes Ranke | 1 | -6/+9 | |
2022-11-04 | Attempt at automatic setting of random effects | Johannes Ranke | 3 | -2/+25 | |
Based on parameters in the separate fits that fail the t-test. | |||||
2022-11-01 | Fix plotting saem fits with covariates | Johannes Ranke | 1 | -1/+1 | |
2022-11-01 | Fix illparms tests | Johannes Ranke | 2 | -4/+4 | |
2022-11-01 | Improved printing for illparms() | Johannes Ranke | 2 | -1/+3 | |
For the case of single fits and no ill-defined parameters found | |||||
2022-10-29 | Another attempt to make the test robust | Johannes Ranke | 2 | -1/+25 | |
As the last update failed on travis. Now we should be more robust, as rounding to one digit works better when parameters are log-transformed (mkin transformations) in the output. | |||||
2022-10-29 | Make test independent of saemix error model parameter patch | Johannes Ranke | 2 | -4/+3 | |
As mentioned in d44c17da6d62568b | |||||
2022-10-28 | Rename parhist to parplot and make it generic | Johannes Ranke | 5 | -5/+6 | |
That parhist name was not the brightest idea, as it does not show histograms. | |||||
2022-10-28 | Update test to renamed method | Johannes Ranke | 1 | -1/+1 | |
2022-10-26 | Really test anova output, test covariates | Johannes Ranke | 2 | -5/+22 | |
2022-10-26 | Don't test parhist and llhist on travis, docs | Johannes Ranke | 1 | -7/+17 | |
2022-10-26 | Reduced sfo model for multistart.saem.mkin tests | Johannes Ranke | 5 | -32/+134 | |
2022-10-26 | Remove kernel density estimate from llhist | Johannes Ranke | 3 | -17/+86 | |
2022-10-26 | Skip llhist and parhist tests as platform dependent | Johannes Ranke | 1 | -0/+1 | |
2022-10-26 | Try to make multistart (more) reproducible | Johannes Ranke | 3 | -143/+139 | |
2022-10-26 | Make multistart work with testthat scopes | Johannes Ranke | 5 | -3/+261 | |
Also skip long-running test with saem using the analytical from mkin | |||||
2022-10-25 | Test multistart method | Johannes Ranke | 2 | -0/+19 | |
2022-10-25 | Round also for nlme print method... | Johannes Ranke | 2 | -7/+8 | |
2022-10-25 | Increase rounding in test for travis | Johannes Ranke | 2 | -2/+2 | |
As the "const" error model parameters is stored on log scale in nlme | |||||
2022-10-25 | Explanatory comment | Johannes Ranke | 1 | -0/+1 | |