You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Related to my other issue (#80), I also noticed that mc.cores is not set when mclapply runs within gene_summary. According to its documentation, it's set to either the MC_CORES environmental variable, or to 2. I don't think this default will work for the typical user, whose computer likely have more than 2 cores but who will also likely not know to set MC_CORES to make use of those cores effectively.
One option is to use a simple fix by setting mc.cores = parallel::detectCores(). Another is to let the user the decide on the number of cores. In either case, the memory footprint may be too big for some computers, depending on their specs (on my machine, the job used roughly ~900 MB RAM per core), and I would guess that the error thrown when the RAM is full will be uninformative, so you may need to consider adding in a safeguard to let users know if their RAM can't handle the number of cores used.
The text was updated successfully, but these errors were encountered:
Related to my other issue (#80), I also noticed that
mc.cores
is not set whenmclapply
runs withingene_summary
. According to its documentation, it's set to either the MC_CORES environmental variable, or to 2. I don't think this default will work for the typical user, whose computer likely have more than 2 cores but who will also likely not know to set MC_CORES to make use of those cores effectively.One option is to use a simple fix by setting
mc.cores = parallel::detectCores()
. Another is to let the user the decide on the number of cores. In either case, the memory footprint may be too big for some computers, depending on their specs (on my machine, the job used roughly ~900 MB RAM per core), and I would guess that the error thrown when the RAM is full will be uninformative, so you may need to consider adding in a safeguard to let users know if their RAM can't handle the number of cores used.The text was updated successfully, but these errors were encountered: