nhejazi / biotmle

:package: :microscope: R/biotmle: Targeted Learning with Moderated Statistics for Biomarker Discovery
https://code.nimahejazi.org/biotmle/
Other
4 stars 2 forks source link

Parallelization via BiocParallel #18

Closed nhejazi closed 7 years ago

nhejazi commented 7 years ago

The original (and current) implementation of this technique used the "foreach" construct to perform parallel computation to assess the large numbers of biomarkers that are generally dealt with. While support for "foreach"-style computation should be kept intact, it would likely be worth adding a flag to the main functions, to allow the (optional) use of BiocParallel.

nhejazi commented 7 years ago

The BiocParallel package seems to concern itself mostly with the registration of backends, and, in fact, claims to fully support the use of foreach. This means that switching over to BiocParallel should be a fairly easy process to complete.

Note that the only uses of functions from the foreach and parallel packages are in the core function biomarkertmle (and nowhere else in this package!), meaning that only a few lines of code likely need to be altered to migrate to using BiocParallel.

nhejazi commented 7 years ago

Resolved by f494b63.