Dear Jason,
dependencies are agnostic of the files or other data items that are
actually created or passed through them when the job is run. This is a
feature that makes batches re-usable for many subjects without linking
the outputs to inputs for each subject over and over again.
You can save a copy of a job with all dependencies replaced by actual
values by calling
cfg_util('savejobrun', jobid, filename)
right after the job has been run. jobid is the number that is logged in
the MATLAB command window when a job is started: "Running job X".
filename is the name of the file you want to save the job to. Note that
this job does not have any information about data flow any more, and you
can no longer use it as a template to process further datasets.
Best,
Volkmar
Dear All,
Hello all.
I am trying to create some provenance info from an SPM batch file. So
I would like to take an SPM job file and read through it to pull out
the different steps, the data and the parameters. I have no problem
doing this EXCEPT where I have specified some dependencies. I am
having trouble decoding what the dependencies refer to.
Essentially if I have the following steps:
Realign << Input data
Reslice >> Output data
smooth << Output data from reslice
I would like the following:
Step1 Realign
input data: FILENAME
parameters :XX
Step2 Reslice
input data: output from Step 1
output data: rFILENAME
parameters: XX
Step3 Smooth
input data: rFILENAME
output data: srFILENAME
parameters: XX
<https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=SPM;e13bd001.1102>
|