Release of constraint in labrids?

Started writing outline, assembling references

  • Reviewed brownie-lite as an alternative model to compare the release-of-constraint, but looks like it will be easier just to modify wrightscape code to keep parameters fixed or global.

  • Writing a wrapper to move the likelihood function for wrightscape to up to the R level.

  • Figure out which function to pass up: easiest is to build off of fit_model, but just evaluate, don’t optimize the likelihood.

Doxygen Documentation

Completing the doxygen documentation of the wrightscape source code to make it easier to navigate

  • doxygen -g makes the Doxyfile

  • doxygen Doxyfile makes the documentation

  • doxygen -u Doxyfile updates the Doxyfile. Not sure when this is necessary

    EXTRACT_ALL = YES Extract documentation even from those elements you haven’t yet commented. HAVE_DOT = YES Use Graphviz for class and collaboration diagrammes. CALL_GRAPH = YES Generate a dependency graph for functions and methods. OUTPUT_DIRECTORY = ../doc/Doxygen

  • Vim Doxygen toolkit:

    :Dox

for function documentation,

:DoxLic

for License,

:DoxAuthor

for file header (author, title, etc).  See link above for details.

  • erg! any typo in latex equation that won’t parse will silently break documentation for rest of file in Doxygen

Code updates

Moving likelihood up to the R level.  Steps:

  1. Update function call for calc_lik to conform to the R .C interface (double and integer pointers only).  DONE

  2. Write a C wrapper function to call the calculation of the last common ancestor, since we probably don’t want to be repeating this calculation each time. DONE

  3. Write wrapper R function for the calc_lik function DONE

  4. Write example functions calling the optimizer routine on a function defined with a subset of parameters from the full version, to mimic ouch, brownie, etc. DONE

  5. Compare convergence quality and speed