You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Still getting a NOTE about fitode : objfun: no visible binding for '<<-' assignment to ‘oldnll’ (for oldnll, oldgrad, and oldpar, for both objfun and gradfun. This is harmless, but needs to be cleaned up if at all possible before we submit to CRAN.
Data:
document all data objects (e.g. blowfly) properly (or remove them for now, via .Rbuildignore)
download lynx/hare data, store it in the package, document
add David Earn as contributor ("ctb")?
Vignettes/documentation:
probably put paper stuff (in inst/doc and vignettes) in .Rbuildignore, leave only the intro vignette
clean up FIXME items in the intro vignette. (Rename to fitode-intro ?)
Ideally before CRAN release but not necessarily:
Maybe add a little bit more to the README file and the intro vignette about the comparison between fitode and other packages for fitting? (Advantages: use of sensitivity equations allows automatic deriv-based optimization; formula-based definition for flexibility, link functions, priors/regularization. Disadvantages: R-only (cf. odin, various other C++-based ODE packages); frequentist only (cf. debInfer or whatever it's called).
The documentation of the S4 methods is technically complete but messy (I ran out of steam for understanding the interaction between roxygen and S4 methods). Ideally the methods documentation would be appropriately combined into a single clear .Rd page ...
Later:
lots of FIXME items scattered through the code which we can talk about.
The text was updated successfully, but these errors were encountered:
To do before CRAN release:
Still getting a NOTE aboutfitode : objfun: no visible binding for '<<-' assignment to ‘oldnll’
(foroldnll
,oldgrad
, andoldpar
, for bothobjfun
andgradfun
. This is harmless, but needs to be cleaned up if at all possible before we submit to CRAN.blowfly
) properly (or remove them for now, via.Rbuildignore
)inst/doc
andvignettes
) in.Rbuildignore
, leave only the intro vignetteFIXME
items in the intro vignette. (Rename to fitode-intro ?)Ideally before CRAN release but not necessarily:
fitode
and other packages for fitting? (Advantages: use of sensitivity equations allows automatic deriv-based optimization; formula-based definition for flexibility, link functions, priors/regularization. Disadvantages: R-only (cf.odin
, various other C++-based ODE packages); frequentist only (cf. debInfer or whatever it's called).Later:
FIXME
items scattered through the code which we can talk about.The text was updated successfully, but these errors were encountered: