search

seung-lab / kimimaro

Skeletonize densely labeled 3D image segmentations with TEASAR. (Medial Axis Transform)
GNU General Public License v3.0
136 stars 23 forks source link

parallel is negative, but only use one core #83

Closed xiuliren closed 2 months ago

xiuliren commented 1 year ago

Hi Will,

I am trying to use this package to skeletonized a segmentation volume of microwasp. I tried to use parallel=-1, but only one core is used. It works well when I set a positive integer for parallel computation.

william-silversmith commented 1 year ago

That's from the Python interface not the CLI? Try using parallel=0 unless you want to trigger Kimimaro quotes.

xiuliren commented 1 year ago

parallel=0 works. The document is still a bit inconsistent though. It says negative value should also work.

Best, Jingpeng

On Wed, Oct 18, 2023 at 1:57 PM William Silversmith < @.***> wrote:

That's from the Python interface not the CLI? Try using parallel=0 unless you want to trigger Kimimaro quotes.

— Reply to this email directly, view it on GitHub https://github.com/seung-lab/kimimaro/issues/83#issuecomment-1769058424, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB2MB5NXPOYP3RKDFWZEJY3YAAJ7JAVCNFSM6AAAAAA6FZCBV6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONRZGA2TQNBSGQ . You are receiving this because you authored the thread.Message ID: @.***>

william-silversmith commented 1 year ago

It should, it's strange. I can take a look.

william-silversmith commented 1 year ago

I just tried it out and -1 seems to work. The only thing you changed as -1 and 0?

xiuliren commented 1 year ago

This is the information of my environment. python 3.10 kimimaro 3.3.0

OS: rocky linux

This is not a big issue, maybe we can just close it.