Runtime Error

[ddshin]

Hi - I am using test data downloaded from OpenfMRI repo (accession # ds000009) to run ndmg with. I used the BIDS online validator to confirm that the directory is BIDS compliant (contains DTI and T1, T2 structural).

I am getting a runtime error.. Something about "IndexError: too many indices"

Any ideas? The screenshot of the error is attached.

Thanks!

[gkiar]

Hi @ddshin - thanks for reaching out! I've seen that bug before, though I thought it had been fixed in the latest version, so this is interesting to see... Could you please try re-running with the pipeline version bids/ndmg:v0.0.49-1-1 and let me know how it goes?

I'm at the open science room today so we can debug and iterate together, if needed 😄

[ddshin]

Hi @gkiar - Awesome, thanks. I am now running with "ndmg:v0.0.49-1-1". I will let you know how it goes. I would love to be at the open science room working on this together with you, except I am in San Diego, CA (UCSD CFMRI). Where are you located?

[gkiar]

Ah - I'm in Montreal these days (McGill/MNI). I'll try to download a subset of the data you mentioned and will dig in to see what may be causing the bug to pop back up.

For what it's worth the bug was originally caused by sessions which have multiple b0 scans within a stack. I know that this was fixed at one point since I've run many datasets with multiple b0 acquisitions, but must have reintroduced it somehow recently if your current attempt doesn't succeed. Either way, I know how to fix so we'll get it resolved in no time 👍

[ddshin]

Thanks @gkiar!

I do get the same error with "ndmg:v0.0.49-1-1". Do you think it would be helpful if I send you the data I am using? If so, let me know and I can leave a dropbox link for you.

[gkiar]

Hey @ddshin - thanks for confirming, it seems I did indeed bork it recently... I'll pop back tomorrow (unfortunately booked the rest of today) and will see where the bug was reintroduced and resolve it. A dropbox link with 1 or 2 representative sessions from your dataset would be greatly appreciated! From my website gkiar.github.io you can find my email and send me the link there, please. And thank you for patiently working with me!

[ddshin]

Cool. I sent you an email with a link to the data. Talk soon!

[chrisgorgo]

You can also upload the data to OpenNeuro.

On Tue, Jun 27, 2017 at 3:48 PM, ddshin notifications@github.com wrote:

Cool. I sent you an email with a link to the data. Talk soon!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/BIDS-Apps/ndmg/issues/12#issuecomment-311507600, or mute the thread https://github.com/notifications/unsubscribe-auth/AAOkp9C_Aih6PET7vMi4nSvqXd7_zbKeks5sIYavgaJpZM4OF-4i .

[gkiar]

resolved via email - we found an issue in FSL's applywarp function when running on Mac with insufficient resources allocated when using the epireg function. We're digging further into the problem, but the issue here was just a manifestation of FSL + Docker + Mac not playing nicely with one another.

[ddshin]

Thank you, Greg. I would like to add that the problem also applies to docker instances on Windows platform.

On Aug 31, 2017 9:26 AM, "Greg Kiar" notifications@github.com wrote:

resolved vial email - we found an issue in FSL's applywarp function when running on Mac with insufficient resources allocated when using the epireg function. We're digging further into the problem, but the issue here was just a manifestation of FSL + Docker + Mac not playing nicely with one another.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/BIDS-Apps/ndmg/issues/12#issuecomment-326349969, or mute the thread https://github.com/notifications/unsubscribe-auth/AJfYvArOYCTBUD5REwPoW4P3iE4M-Ed-ks5sdt7CgaJpZM4OF-4i .