Dominik Kutra
Dominik Kutra
Hey @wolny, I reproduced the bug from today's meeting. ## Bug description When deselecting all object features an exception is raised. ## To Reproduce Steps to reproduce the behavior: Within...
was there any followup on this issue? if not we should maybe close this. I don't see any actionable item coming from this right now.
Hello @samuelsuntree, [Pixel Classification](https://www.ilastik.org/documentation/pixelclassification/pixelclassification) and Voxel Segmantation are both CPU-bound. So for good performance you'd want something like 4 or 8 cores (intel will probably be faster compared to amd)....
Hey, one of the main reasons is mkl, which is involved in many numpy and pytorch operations. This library comes from intel and it is notorious for underperforming on amd...
Hello @jospell, sorry you ran into this. The error message is not really helpful. ilastik will always append a channel axis to 2d data (during training), so it expects three...
Hello @jospell, did it help specifying the input axes?
Hello @jospell, do you get an error when opening the file (in Gui mode?!)? Well when the data you want to batch process is somewhat similar to the training data...
will also fail for `int` (hearsay)
this post is still wip, especially regarding performance analysis on osx, windows. I looked into this a bit - here are some findings: ### Summary: * `fastfilters` currently, uses some...