[Neuroimaging] Basic Nipype question

John Pellman pellman.john at gmail.com
Tue Dec 26 23:37:10 EST 2017


Apologies for the duplicate message- I didn't notice that my initial e-mail
for through and decided to resend with the account that belongs to this
list.

On Dec 26, 2017 11:35 PM, "John Pellman" <pellman.john at gmail.com> wrote:

>
> In general yes nodes do check if they have been successfully run, but the
> accuracy of the checks depend upon how you've configured your nipype
> installation / the nipype config you're using.  There are two methods that
> a node uses to determine if it has already been run- hashing (slower but
> more accurate) and timestamp (faster but much less accurate; easily
> confused if your pipeline's sink directory is copied from one disk volume
> to another without preserving timestamps).
>
> See:
> http://nipype.readthedocs.io/en/latest/users/config_file.html#execution (under
> 'hash_method')
> http://nipype.readthedocs.io/en/latest/users/debug.html (referenced in
> point 7)
>
> This isn't my line of work anymore so others in this group might be able
> to answer your questions more effectively in the future (I just saw this in
> my inbox and recognized it as a question I could answer based off previous
> experience).  You might also want to check out https://neurostars.org/ as
> a venue for getting nipype-related questions answered.
>
> On Dec 22, 2017 6:52 PM, "Luke Bloy" <luke.bloy at gmail.com> wrote:
>
>> Hi,
>>
>> I'm just getting starting with nipype and was wondering if  individual
>> nodes checked if they had already been successful run.
>>
>> Basically will a node reexecute if its outputs already exist?
>>
>> Thanks,
>> Luke
>>
>> _______________________________________________
>> Neuroimaging mailing list
>> Neuroimaging at python.org
>> https://mail.python.org/mailman/listinfo/neuroimaging
>>
>>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20171226/418edfa5/attachment.html>


More information about the Neuroimaging mailing list