Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
|
|
I am postponing my attempts to get the nlmixr interface to CRAN, given
some problems with nlmixr using R-devel under Windows, see
https://github.com/nlmixrdevelopment/nlmixr/issues/596
and
https://github.com/r-hub/rhub/issues/512,
which is fixed by the removal of nlmixr from the testsuite.
For the tests to be more platform independent, the biphasic mixed
effects models test dataset was defined in a way that fitting
should be more robust (less ill-defined).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
By actually passing dot arguments (that can include fixed.estim) to
saemixModel, as advertised in the docs.
|
|
as obtained by mean_degparms(..., random = FALSE), instead
of the list obtained by mean_degparms(, ... random = TRUE) that
is passed by nlme.mmkin to nlme by default.
|
|
Respect digits argument when printing correlations in summaries
|
|
|
|
|
|
|
|
This avoids numerical instabilities that sometimes occur with the FOCEI
algorithm in nlmixr when the initial values are very close to the
optimum values
|
|
|
|
|
|
|
|
|
|
The EFSA URL failed on winbuilder because some cookie sets a different
domain, so I am now using \href{} with the full link as link text instead
of \url{}
|
|
The data given in previous mkin versions for "Borstel 2" (Staudenmaier
2013) are in fact a subset of the "Borstel 1" data, with the "Borstel 2"
data simply being further analysed for isomer composition by
Staudenmaier (2013).
While correcting this, the URL of the RAR published by EFSA was
updated, as the EFSA Register of Questions was replaced by the
Open EFSA portal.
This also affects the recent publication in "Environments" [1], where
these two datasets were pooled. However, the influence on the final
results can be expected to be minimal. Out of the 16 observations for
the Borstel soil, five were erroneously duplicated which just changes
the weight of the different sampling points and modifies the variance.
The conclusions of the publication are qualitative and are not
considered to be affected.
[1] https://doi.org/10.3390/environments8080071
|
|
Evaluations with nlme, saemix and nlmixr are included
|
|
|
|
|
|
Also adapt summary.nlmixr.mmkin to correctly handle the way
formation fractions are translated to nlmixr
|
|
Works for the dimethenamid data, at least for FOCEI. Very little testing
yet. The summary function does not yet handle the new transformations
of formation fractions (that are in fact very old, as they were used
in the very first version of mkin). The test file has no tests yet, just
some code that may be used for testing.
|
|
|
|
|
|
- All degradation models are specified as ODE models. This appears to be
fast enough
- Error models are being translated to nlmixr as close to the mkin error
model as possible. When using the 'saem' backend, it appears not to be
possible to use the same error model for more than one observed variable
- No support yet for models with parallel formation of metabolites, where
the ilr transformation is used in mkin per default
- There is a bug in nlmixr which appears to be triggered if the data are
not balanced, see nlmixrdevelopment/nlmixr#530
- There is a print and a plot method, the summary method is not finished
|
|
|
|
|
|
In residual plots, use xlab and xlim if appropriate
|
|
|
|
|
|
|
|
- fit_with_errors for saem()
- test_log_parms for mean_degparms() and saem()
|
|
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.
|
|
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.
|
|
|
|
Also bump version to 1.0.3.
|
|
Run make testcheck to regenerate logs with merge conflicts
|
|
mkinfit: Keep model names stored in mkinmod objects, avoiding their loss in gmkin
|
|
Also after the upgrade from buster to bullseye of my local system, some
test results for saemix have changed.
|
|
The errors in the example code were in the \dontrun sections, so they
were not caught by CRAN checks. In addition, the static help files
generated with pkgdown were cached, so I noticed the errors only
after completely regenerating the documentation for version 1.0.0.
|
|
- 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
|
|
|
|
See NEWS.md. Closes #12
|
|
This makes fitting with saem within parallel::mclapply much faster
and, surprisingly, much less hungry for RAM.
|