Restructure models to use a "version" argument instead of separate functions #178
Conversation
…bout-default_prior-standata-and-stancode
…bout-default_prior-standata-and-stancode
- since it should work with v0.2.1
venpopov
changed the title
- Adapt the `use_model_template` function to the new `version` argument for `bmmodels` - Minor change to the `bmmformula` function to remove redundancies in model specific `bmf2bf` functions
There was a problem hiding this comment.
I found some minor things in the model_imm.R script that I commented. Apart from that I did not run into any issues on my machine and like the cleaner setup for the different models. I also think this will make it easier to accomodate different task versions for these models (e.g. for change detection versus continous reproduction).
Thanks for taking care of this.
tests/testthat/test-fit_model.R
Outdated
There was a problem hiding this comment.
Should we rename this file to test-bmm.R ? I think this would just be for consistency in naming, but it might still be good to just take care of this now.
|
In b3e85cb I added the changes to the |
Looks good! Thanks! |
Summary
This is the follow up to #177. Closes #163
the sdmSimple() model is now sdm()
the IMMfull(), IMMabc() and IMMbsc() are now all imm(), with a version argument:
the "full" is the default
I decided to leave the two mixture models separately. For them almost all the meta information is different - the model name, the citation, etc. So I thought it's easier to keep the documentation separately, otherwise the help page becomes complicated.
This framework should now work for future models that have multiple versions.
Tests
[X] Confirm that all tests passed
[X] Confirm that devtools::check() produces no errors
I reran all the internal reference fits tests as this is a more substantial change. All is good and we replicate previous fits.
Release notes