New DD models #34
New DD models #34
Conversation
… DD models in dd_lamuN()
Merge branch 'develop' into new-ddmodels # Conflicts: # R/dd_loglik.R # tests/testthat/test_z_DDD.R
The problem is with |
| #if (ddep > 5) { | ||
| # stop("This DD model is not implemented for the analytical method yet.") | ||
| #} |
There was a problem hiding this comment.
| #if (ddep > 5) { | |
| # stop("This DD model is not implemented for the analytical method yet.") | |
| #} |
Commented out code is not desirable
There was a problem hiding this comment.
Only put stop for user-accessible functions
| abstol = 1e-10 | ||
| reltol = 1e-8 | ||
| abstol = 1e-16 | ||
| reltol = 1e-10 |
There was a problem hiding this comment.
keep reduced tolerance for CR model?
| y = dd_integrate(probs,brts[(k-1):k],rhs_func_name,c(pars1,k1,ddep),rtol = reltol,atol = abstol,method = methode) | ||
| probs = y[2,2:(lx+1)] | ||
| #y = deSolve::ode(probs,brts[(k-1):k],rhs_func,c(pars1,k1,ddep),rtol = reltol,atol = abstol,method = "analytical") | ||
| #probs2 = y[2,2:(lx+1)] | ||
| probs = dd_loglik_M(pars1,lx,k1,ddep,tt = abs(brts[k] - brts[k-1]),probs) |
There was a problem hiding this comment.
I'm unsure which version should be kept here
There was a problem hiding this comment.
No, this is just git mismatching the versions because of the large size of the file and the redundancies in the script. In reality, new-dd-models retains the same version as develop
|
|
||
| if (ddmodel > 5) { | ||
| if (methode == "analytical" || cond == 3) { | ||
| stop("Sorry, ddmodel options > 5 have not been developed for method = \"analytical\" or cond = 3.") |
There was a problem hiding this comment.
| stop("Sorry, ddmodel options > 5 have not been developed for method = \"analytical\" or cond = 3.") | |
| stop("Sorry, ddmodel options > 5 have not been tested for method = \"analytical\" or cond = 3.") |
|
Hi @rsetienne, are there still some standing issues we need to address here, or is this branch ready to be merged to |
|
Hi, develop needed to be fixed first. I think we're almost ready.
Op vr 10 feb. 2023 18:07 schreef Théo Pannetier ***@***.***>:
… Hi @rsetienne <https://github.com/rsetienne>, are there still some
standing issues we need to address here, or is this branch ready to be
merged to develop?
—
Reply to this email directly, view it on GitHub
<#34 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADMTHGP5NKTGDYDLSW3CRUTWWZYVHANCNFSM5ETK5I6Q>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
There are still errors on GHA when trying to merge. @TheoPannetier, can you please check? |
I expanded the possible of values of arguments
ddmodel/ddepfrom 5 to 15 possible DD models, covering all combinations of linear, power, or exponential DD; or CR speciation / extinction (except of course the CR speciation + CR extinction model).Updated the corresponding doc
Added tests to make sure that
dd_loglik()doesn't crash and returns a non-infinite loglik when using themdd_loglik()return-Infinstead of an errorAdded a few helper functions (
both_rates_vary(),is_speciation_linear(),get_Kprime()) to help with model-dependent conditions