We are pleased to announce the next major release of the OpenMx library. Use the installation command in an R terminal
source('http://openmx.psyc.virginia.edu/getOpenMx.R'). There have been several minor enhancements to the OpenMx library interface (all changes are backwards compatible) and several performance improvements. We have implemented a dependency tracking system in the OpenMx backend. If an algebra is independent of all the definition variables in a FIML calculation, then the algebra is computed once per FIML iteration. Some of the test cases in our performance suite are up to x20 faster. You can follow the performance testing on this page.
Release 1.3.0-2168 (September 17, 2012)
- added 'cache' and 'cacheBack' arguments to mxEval()
- added omxLocateParameters() function (see ?omxLocateParameters)
- type='RAM' allowing 'manifestVars' argument to appear in a different order than in the observed covariance matrix.
- the configuration mxRun(model, checkpoint = TRUE) writes a line in the checkpoint file at the conclusion of model optimization.
- added "Always Checkpoint" to mxOptions() with values "Yes" or "No"
- header for the checkpoint file will identify anonymous free parameters with the string modelName.matrixName[row,col]
- omxGetParameters() and omxSetParameters() support anonymous free parameters
- implemented cov2cor in the OpenMx backend
- mxOption "Major iterations" accepts either a value or a function
- now tracking the MxAlgebra and MxMatrix objects that need to be updated when populating free parameters or updating definition variables.
- performance improvements in mxModel() when building RAM models
- performance improvements in mxRun() frontend for large matrices
- rewrite on the processing of objective functions in the frontend
- added R functions omxCbind(), omxRbind(), and omxTranspose()
- added 'fetch' argument to omxGetParameters()