At the OpenMx Developers Meeting on 1/13/12 we discussed the following:
- The grant application submitted to NIH, that if accepted will help fund the OpenMx project, has been successfully submitted.
- The group discussed the content of a grant application through NSF, that if funded will fund the OpenMx project.
- Dan Hackett has checked in a front-end working version (passes all test cases) of PPML to the OpenMx code base. Because this implementation is in the front-end it is unclear the degree to which speedups will occur. However, this represents an excellent iterative step towards a back-end implementation of PPML.
- Mike Hunter has checked in some code that calculates the covariance and means helping move towards the implementation of LISREL objective functions in OpenMx. Mike is currently experiencing some issues "merging"/"augmenting" matrices. The group discussed using calls to
rbind. The group also discusses using filter matrices to achieve this functionality.
- The group continued to discuss how to make OpenMx behave nicely on a cluster. The group continues to believe this can be achieved by reading and setting several PBS environment variables.
- Empirical results from Mike Neale show that hyper-threading is not particularly useful in OpenMx. The decrease in efficiency occurs at 13 threads in each of the linked graphs, which the group believes is when hyper-threading comes into play.
- The group discussed several possible color contrast issues in the Beginner's Guide in the Documentation. The group decided to explore posting to the Sphinx forums for solutions.
- The group also discussed creating a post and sending email to SEMNET to encourage OpenMx 1.2 Beta users (FIML parallelization) to tell us about their experiences speedups or lack thereof.