Use parallel processing in semlbci

sfcheung / semlbci

Find the likelihood based confidence intervals for parameters in structural equation modeling

https://sfcheung.github.io/semlbci/

1 stars 0 forks source link

Closed sfcheung closed 3 years ago

sfcheung commented 3 years ago

Done:

Updated semlbci such that it can select all free parameters automatically if so desired. This is a proper use case if parallel processing is used.
Tested.

Not To Do:

Add a warning when the memory of the workspace is large in size, because the whole parent environment will be copied to each worker.
- Not easy to implement. May do it in the future.
Add tryCatch in parallel to avoid quitting without stopping the workers.
- Not easy to implement. May do it in the future.

sfcheung commented 3 years ago

Moved to this milestone (https://github.com/sfcheung/semlbci/milestone/1) because it may work soon.