Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2020-05-11 | Add analytical solution for DFOP-SFO | Johannes Ranke | 1 | -8/+8 | |
This is about twice as fast as deSolve compiled in the case of FOCUS D | |||||
2020-05-11 | Analytical solutions for all SFO variants | Johannes Ranke | 1 | -27/+16 | |
2020-05-11 | Don't run visual tests on r-devel for now | Johannes Ranke | 1 | -14/+25 | |
https://github.com/r-lib/vdiffr/issues/86 | |||||
2020-05-11 | Avoid merge() and data.frame() in cost function | Johannes Ranke | 1 | -10/+10 | |
also for deSolve and eigenvalue based solutions. This noticeably increases performance for these methods, see test.log and benchmark vignette. | |||||
2020-05-10 | Default to analytical for coupled models if available | Johannes Ranke | 1 | -10/+10 | |
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 | |||||
2020-05-09 | Avoid the call to merge for analytical solutions | Johannes Ranke | 1 | -16/+17 | |
This increases performance up to a factor of five! | |||||
2020-05-08 | We do not need the n.outtimes argument for mkinfit | Johannes Ranke | 1 | -9/+9 | |
As we set the tolerance for ode() appropriately | |||||
2020-05-08 | Avoid duplicate merge in tc error model fit | Johannes Ranke | 1 | -11/+11 | |
This increases the performance in the complete test suite by about 20 secs from 120 to around 100 secs. I tried improving merge speed by using data.table on another branch, but this did not give a noticeable performance gain. | |||||
2020-05-08 | Slight performance gain by reducing default n.outtimes | Johannes Ranke | 1 | -14/+14 | |
2020-05-08 | Roxygenize, adapt tests | Johannes Ranke | 1 | -7/+7 | |
2020-05-07 | Another overhaul of analytical solutions | Johannes Ranke | 1 | -11/+11 | |
Still in preparation for analytical solutions of coupled models | |||||
2020-05-06 | Change implementation of analytical solutions | Johannes Ranke | 1 | -15/+15 | |
Preparing for symbolic solutions for more than one compound | |||||
2020-04-22 | Support SFORB with formation fractions | Johannes Ranke | 1 | -9/+9 | |
2020-04-20 | Increase test tolerance for residuals | Johannes Ranke | 1 | -4/+4 | |
This was done to address the test failure on r-devel-linux-x86_64-debian-gcc on CRAN | |||||
2020-04-18 | Avoid assignment to .GlobalEnv for CRAN | Johannes Ranke | 1 | -9/+7 | |
2020-04-17 | Fix summarizing old objects | Johannes Ranke | 1 | -7/+9 | |
2020-04-16 | Increase test coverage | Johannes Ranke | 1 | -10/+10 | |
2020-04-09 | Add another visual test for coverage | Johannes Ranke | 1 | -16/+16 | |
2020-04-09 | Improve test coverage and fix nlme test | Johannes Ranke | 1 | -16/+16 | |
No idea why I had to do more assignments all of a sudden in test_nlme.R | |||||
2020-04-08 | Use two assignments to avoid skipping another test | Johannes Ranke | 1 | -15/+11 | |
2020-04-08 | Make the nlme test file work with testthat | Johannes Ranke | 1 | -19/+10 | |
2020-04-07 | Add some tests for the nlme convenience functions | Johannes Ranke | 1 | -9/+19 | |
2020-04-03 | Compiled models article, reduce distractions | Johannes Ranke | 1 | -15/+15 | |
- Added a section with platform specific notes on getting compiled models to work to the compiled models article - Don't return empty SFORB parameter list from endpoints() if there is no SFORB model - Avoid warnings when using standardized = TRUE in plot.mmkin() | |||||
2020-03-31 | Use inline documentation for R6 class mkinds | Johannes Ranke | 1 | -30/+15 | |
2020-03-31 | Static documentation rebuilt by pkgdown | Johannes Ranke | 1 | -18/+33 | |
2019-11-13 | Calculate Akaike weights | Johannes Ranke | 1 | -10/+11 | |
2019-11-09 | Add a lack-of-fit test | Johannes Ranke | 1 | -11/+15 | |
- Switch an example dataset in the test setup to a dataset with replicates, adapt tests - Skip the test for lrtest with an update specification as it does not only fail when pkgdown generates static help pages, but also in testthat | |||||
2019-11-05 | Fix standardized residual plots in 'plot_res' | Johannes Ranke | 1 | -2/+2 | |
2019-11-04 | Scaled residual plots | Johannes Ranke | 1 | -10/+10 | |
2019-11-01 | Fix bug in yesterdays release, add methods for BIC | Johannes Ranke | 1 | -12/+12 | |
2019-10-31 | Reorganize tests to reduce time on winbuilder | Johannes Ranke | 1 | -10/+9 | |
2019-10-31 | Add likelihood ratio test and other methods, fixes | Johannes Ranke | 1 | -7/+10 | |
The likelihood ratio test method is lrtest, in addition, methods for update and residuals were added. | |||||
2019-10-28 | Fix the cutoff for likelihood based intervals | Johannes Ranke | 1 | -8/+8 | |
The cutoff now matches what is given by Venzon and Moolgavkar (1988). Also, confidence intervals closely match intervals obtained with stats4::confint in the test case where an stats4::mle object is created from the likelihood function in one test case. Static documentation rebuilt by pkgdown | |||||
2019-10-26 | Fix checks and tests | Johannes Ranke | 1 | -9/+9 | |
Static documentation rebuilt by pkgdown | |||||
2019-10-25 | Use roxygen for functions and methods | Johannes Ranke | 1 | -4/+4 | |
2019-10-22 | Two more plotting tests | Johannes Ranke | 1 | -4/+4 | |
2019-10-22 | Go back to d_3 for errmod = "obs" | Johannes Ranke | 1 | -2/+2 | |
in the hope that this makes plotting cross-platform also for this error model | |||||
2019-10-22 | Improved visual testing | Johannes Ranke | 1 | -5/+5 | |
2019-10-21 | Fix platform dependence of the correlation matrix | Johannes Ranke | 1 | -9/+5 | |
Especially on winbuilder (i386 and amd64) | |||||
2019-10-21 | Explicit solution_type for FOCUS D in tests | Johannes Ranke | 1 | -5/+9 | |
Because on winbuilder obviously gcc was not found, so the Eigenvalue based solution method was used, leading to a test failure when comparing the summary, as the solution method is listed | |||||
2019-10-21 | Tests all pass locally | Johannes Ranke | 1 | -1/+1 | |
2019-10-21 | Improve some plotting routines, more tests | Johannes Ranke | 1 | -5/+5 | |
Static documentation rebuilt by pkgdown | |||||
2019-10-21 | Improve tests, remove geometric_mean | Johannes Ranke | 1 | -10/+11 | |
2019-10-21 | Refactor mkinfit, infrastructure work | Johannes Ranke | 1 | -21/+21 | |
mkinfit objects now include an ll() function to calculate the log-likelihood. Part of the code was refactored, hopefully making it easier to read and maintain. IRLS is currently the default algorithm for the error model "obs", for no particular reason. This may be subject to change when I get around to investigate. Slow tests are now in a separate subdirectory and will probably only be run by my own Makefile target. Formatting of test logs is improved. Roundtripping error model parameters works with a precision of 10% when we use lots of replicates in the synthetic data (see slow tests). This is not new in this commit, but as I think it is reasonable this closes #7. | |||||
2019-07-08 | Support summarizing older fits | Johannes Ranke | 1 | -7/+8 | |
generated with mkin < 0.9.49.5 | |||||
2019-07-05 | Static documentation rebuilt by pkgdown | Johannes Ranke | 1 | -6/+6 | |
2019-07-03 | Add check and test logs | Johannes Ranke | 1 | -6/+6 | |
2019-06-05 | Adapt tests to new algorithms and output | Johannes Ranke | 1 | -377/+9 | |
One of the tests exceeded the number of iterations when using the d_3 error model algorithm, so only use "direct" in this case. | |||||
2019-06-04 | Additional algorithm "d_c", more tests, docs | Johannes Ranke | 1 | -9/+377 | |
The new algorithm tries direct optimization of the likelihood, as well as a three step procedure. In this way, we consistently get the model with the highest likelihood for SFO, DFOP and HS for all 12 new test datasets. | |||||
2019-05-08 | Add functionality to plot the error model | Johannes Ranke | 1 | -18/+13 | |
by plotting squared residuals against predicted values, and showing the variance function used in the fitted error model. Rebuild docs |