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
A workaround on MacOSX is to compile numpy and use ATLAS instead of the native Accelerate.Framework. Further, there needs to be an additional import numpy statement in the core_helper function.
Otherwise the fastest process wins and all other segfault.
The Problem comes from numpy that cannot use the
Accelerate.Framework
in python multiprocessing.The segfaults are related to
A workaround on MacOSX is to compile numpy and use
ATLAS
instead of the nativeAccelerate.Framework
. Further, there needs to be an additionalimport numpy
statement in the core_helper function.Otherwise the fastest process wins and all other segfault.
Here is an installation guide for ATLAS and numpy (https://github.com/CellCognition/cecog/wiki/Build-build-accelerated-numpy-using-ATLAS-on-Mac-OSX).
The text was updated successfully, but these errors were encountered: