"Null" Error in CPA

We get the following error when we use any of the graphing functions, but only for certain measurement variables (this will change depending on the pipeline and data analyzed)

An error occurred in the program:
ValueError: invalid literal for float(): NULL

Traceback (most recent call last):
File “Density.pyc”, line 180, in on_update_pressed
File “Density.pyc”, line 229, in setpointslists

We believe this relates back to NULL measurements within the object.csv, which we have confirmed by removing them. Is there a way to get CellProfiler Analyst to ignore either these values or just totally ignore the objects that have multiple NULL measurements so you can plot?

Thanks, Heather

We will add this to the list of bugs to be addressed in the next release. We’ll repost here when that happens.

Regards,
-Mark

Mark,

Do you have a timeframe for the next release of CPA?

Thanks,
Heather

Hi Heather,

Would you mind letting me know if this is still a problem in the latest release since I see you are now using it? If so I’ll make a point to fix it and try to get a new binary out asap.

-Adam

Adam,

Sorry I didn’t get back to you sooner on this. I do in fact still get the same error in the new version.

Heather

Thanks Heather,

Unfortunately I’m having trouble replicating your issue in the latest release of CPA. I have a test database set up with a large number of NULLS in a column of both the per_image and per_object tables, and all of the plotting tools work just fine (the NULLS are ignored and all other values are plotted). Are you sure you’re using the most recent release from the website?

[edit] I was just reminded that we did not package the new windows release with an installer like we did for the last one… which could have been confusing. The file must be unzipped and the .bat file is the relevant one to run. [/edit]

-Adam