Description
I encounter examples where running morphology.reconstruction() crashes python without error message - it just throws a python system error in windows after which the kernel has to be restarted. The input arrays do not seem to be particular in any way, and not getting any error message makes it really hard for me to see where things go wrong. This is only happening for specific arrays and - surprisingly - stochastically after 1-5 runs of this command.
Way to reproduce
se = morphology.disk(1)
Ie = morphology.erosion(map_, se)
fmap = morphology.reconstruction(Ie, map_)
where map_ has shape (32, 32) is dtype('float64') and contains nans.
Version information
3.6.9 |Anaconda, Inc.| (default, Jul 30 2019, 13:42:17)
[GCC 4.2.1 Compatible Clang 4.0.1 (tags/RELEASE_401/final)]
Darwin-18.6.0-x86_64-i386-64bit
scikit-image version: 0.15.0
numpy version: 1.16.4
Description
I encounter examples where running morphology.reconstruction() crashes python without error message - it just throws a python system error in windows after which the kernel has to be restarted. The input arrays do not seem to be particular in any way, and not getting any error message makes it really hard for me to see where things go wrong. This is only happening for specific arrays and - surprisingly - stochastically after 1-5 runs of this command.
Way to reproduce
where
map_has shape (32, 32) is dtype('float64') and contains nans.Version information