Obtaining .m files for batch processing


#1

I am trying to get .m files to launch batch processes in a cluster.
I am “playing” with two images before going to a huge experiment.
When I “analyze images” in my current pipeline, both of them are correctly analyzed. On the other hand, when I add the module CreatBatchFiles at the end of the pipeline, only one image (or one cycle) is analyzed. Hence, when I ask for 1 batch file for each cycle, only one Batch_ file is created.

I had a look to the parameters of CreateBatchFiles and did not find the explanation. Any idea of what could be happening?

Thank you for your help.


#2

If you are only analyzing two images, when you add CreateBatchFiles the output should be as follows:

Batch_data.mat
Batch_2_to_2.m

The first cycle is analyzed and the data for that cycle is stored in Batch_data.mat. Then, you submit the .m files to the cluster and they load Batch_data.mat and analyze whichever set of images it specifies, in this case, it is only image 2.

If you would like to run analysis on a cluster, I would first read the help for batch processing by going to Help -> Getting Started -> BatchProcessing (it can also be found on the forum). This explains how to set up a cluster for analysis and how to submit all the jobs to the cluster at once.

Hope this helps,
Mike