Image Processing JupyterLab not Opening
Hello,
I've been trying to access the IMAGE_PROCESSING flavour of JupyterLab in order to interact with Brain MRI data, but I'm having some trouble getting on. I'm requesting the following instance:
Whenever I run this, however, the job appears to get stuck. In the Monitor tab I'm told that the job is ‘Runnnig’, and that I'm actively being charged for it:
However the job seems to be officially stuck as ‘Launching’:
I'm not able to actually open it, I just receive the following message:
I've tried this several times now and have ended up stuck in this position for up to 30 minutes while being charged but not being able to access it. Does anyone know what might be going wrong here? I've been able to succesfully open a PYTHON_R flavour of JupyterLab, but no luck with IMAGE_PROCESSING.
Many thanks,
Nick
Comments
4 comments
Hi Nick,
I've just replicated this issue. We will escalate this to DNAnexus and one of us will come back to you when we have a resolution.
Many thanks,
Ollie
Hi Nick,
I'm a student who has been studying UKB imaging data recently. I have also encountered the same problem as yours. Anyway, May I ask if you know how to use JupyterLab on RAP to obtain the raw images? I would be most grateful if you could offer me any guidance.
Many thanks,
Mei
Hello,
This issue was turned into a ‘Request’ where I've now received a reply. Advice included switching instance Type from HDD to SSD. For example, using the settings below, and instance does open in about 9 minutes.
I havent; actually played around with the data yet, but I'm planning to use this documentation on GitHub to figure it out. They also have pages specifically on how to download data from the RAP to these notebooks.
So basically I think my plan is to write a script in R Studio (which opens faster and is cheaper) to loop over Bulk files and find the file locaiton of all subjects with fMRI NIFTI files, and create a list of filepaths. Then I'll access this document in Jupyter so I can run a ‘dx download’ command on each of the filepaths, and use FSL on them.
I'm sure there are easier ways to do it, but this is my best plan for now! Afraid I probably can't be of much more help, but will post here if I make significant progress.
Nick
Thanks for following up here Nick. We will work with DNAnexus to update the documentation associated with the Image Processing JupyterNotebook option (https://dnanexus.gitbook.io/uk-biobank-rap/working-on-the-research-analysis-platform/running-analysis-jobs/tools-library#image-processing) to provide these details.
Thanks,
Ollie
Please sign in to leave a comment.