I’ve got CellProfiler generating data from a linux cluster, and am exporting the data to a SQLite db file at the end of the run (and generating a properties file). I’d like to analyze this data with CPA.
Am I right in thinking that this will be hard with the windows version of CPA, due to the differences in Windows and Unix path names? So far I’ve only gotten the windows version of CPA to work with data output by the windows version of CP, but I’m sure I haven’t tried everything.
I’ve also tried to compile CPA under Ubuntu/Debian and got it to open up my linux-generated CP data. This works, but it crashes frequently with errors in bits of python code, which I interpret to mean that it’s buggy and/or that we didn’t build it correctly (right?).
The last option appears to be using the Mac OS version of CPA to analyze the linux-generated CP data. I’m working on getting a mac set up to try this, but for people that don’t have a mac it seems like they’d be out of luck with CPA.
I came across a request on github for a tool within CPA that would allow changes to path names in the properties and db files, which seems incredibly useful. But it looks like this hasn’t seen much activity lately.
Has anyone else gotten a version of CPA to work with linux-generated CP data? I’d be grateful to hear about things I haven’t tried yet.