EpistasisLab / EpistasisLab/tpot
Number of simultaneously running jobs fluctuates
- Dominant language
- Jupyter Notebook
- Stars
- 10.1k
- Forks
- 1.6k
- PR merge metrics
- No merged PRs in 30d
Description
Hello,
I noticed that when I am running TPOT, with njobs=8 for example, and checking the processes using *top* in linux, I see 8 different processes running. However after a while that number starts decreasing until only 1 is remaining and finally, after that process has completed, 8 new processes start again and the cycle repeats.
Since the last pipeline remaining can take significantly longer than the remaining pipelines (or even cross the maximum minutes threshold), this doesn't seem to be efficient.
To me it seems that TPOT isn't starting another process until all of the initial 8 different processes have been completed. Is this the case?
Thanks !
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.