Controlling the number of threads

gagneurlab / MMSplice_MTSplice

Tissue-specific variant effect predictions on splicing

MIT License

40 stars 21 forks source link

Controlling the number of threads #42

Closed tstohn closed 4 years ago

tstohn commented 4 years ago

MMSplice version: 2.0.0
Python version: 3.6
Operating System: Ubuntu

Is there any way to control the number of used threads in MMSplice? Setting OMP_NUM_THREADS seems to control tensorflows thread number, however MMSplice still uses all available cores.

All the best & Thanks, Tim

s6juncheng commented 4 years ago

Hi Tim,

thanks for using MMSplice. Have you tried this? https://github.com/keras-team/keras/issues/4740#issuecomment-268197982

BTW, I'm also looking into the writeVCF option.

Best, Jun

tstohn commented 4 years ago

Hi Jun,

Thanks a lot for the quick reply. I've tried it and it also created a bunch of threads (more than specified). However, I just saw that actually only the specified number of threads are running - so that is fine with me and anyway tensorflow dependant.

Thanks a lot also for looking into the writeVCF option, Cheers, Tim