Error while processing Identify Secondary Objects


I’ve been trying to set-up a new pipeline which I’m usually capable of doing - but with a new file I have (its a 2500x2500 pixel image) I keep on getting the same error code:

Traceback (most recent call last):
File “cellprofiler\gui\pipelinecontroller.pyc”, line 2826, in do_step
File “cellprofiler\pipeline.pyc”, line 2067, in run_module
File “cellprofiler\modules\identifysecondaryobjects.pyc”, line 672, in run
File “cellprofiler\objects.pyc”, line 291, in relate_children
File “cellprofiler\objects.pyc”, line 315, in relate_histogram
File “numpy\core\fromnumeric.pyc”, line 943, in argmax

While identify primary objects runs fine, it always stops here. Is this really a memory issue of my PC or is there something wrong in the pipeline? I’m running an i7-6000@2.8Ghz and 8gb rams on a 64-bit system.


Cellprofiler_pipeline.cpproj (534.5 KB)


The version of the pipeline you sent doesn’t have your ID Secondary module in it and you didn’t include any images, so it’s not possible to troubleshoot the settings, but it’s possible if you’ve got a lot of primary objects in your large-ish image that 8GB of RAM might not be enough. One way to tell if this is likely true is to set CP to only analyze one image at at time (in File->Preferences) rather than scaling up to use however many cores you have; that’ll keep memory requirements at a minimum, and you can use the TaskManager to see just how much memory per image you need.


Sorry for that! - you can find a screenshot of what my primary objects identification looks like below…i tried reducing the cores to 1 as suggested but it didn’t fix the problem…I actually tried cropping the picture and I was capable to get a seconday object identification running so I’m guessing RAM is the culprit. As a ballpark figure would you be able to tell me how much RAM processing these types of pictures would take?
Thanks so much for your help


I’m really not sure; probably the best thing to do would be to see if your lab/university/company etc has a loaner computer with a larger amount of RAM (16GB at minimum) that you can borrow to try the pipeline out on and see if a) it’ll run and if so b) what TaskManager says it consumes for memory. Given that you’ve got tens of thousands of primary objects though, I’m not at all surprised you’re running into memory issues.