Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Ship fsaverage5 with nilearn #1705

Closed
GaelVaroquaux opened this issue Jul 26, 2018 · 12 comments
Closed

Ship fsaverage5 with nilearn #1705

GaelVaroquaux opened this issue Jul 26, 2018 · 12 comments
Labels
Datasets This issue is related to datasets or the :mod:`nilearn.datasets` module. Good first issue Good for newcomers. Equivalent to "very low" effort. Priority: high The task is urgent and needs to be addressed as soon as possible.

Comments

@GaelVaroquaux
Copy link
Member

We now need fsaverage5 for basic interactive plotting functionality. It is retrieved via datasets.fetch_surf_fsaverage.

However, as any Internet access, this is brittle and can fail. For example:
https://ci.appveyor.com/project/nilearn-ci/nilearn/build/1.0.2612/job/0bn4wduir9icwafu

I suggest that we embed in nilearn the files needed for plotting, after compression with "gzip --best". It gives the following sizes:

$ ll -h
total 784K
-rw-r--r-- 1 varoquau varoquau 196K Jul 26 14:49 pial_inflated.left.gii.gz
-rw-r--r-- 1 varoquau varoquau 196K Jul 26 14:49 pial_inflated.right.gii.gz
-rw-r--r-- 1 varoquau varoquau 196K Jul 26 14:49 pial.left.gii.gz
-rw-r--r-- 1 varoquau varoquau 196K Jul 26 14:49 pial.right.gii.gz

This will make the download bigger by 784K, but the addition of the javascript libraries already added 1.3M (coming mostly from plotly-gl3d-latest.min.js).

The code of datasets.fetch_surf_fsaverage will need to be modified to load these files when it is fsaverage5 pial.

@GaelVaroquaux GaelVaroquaux added Good first issue Good for newcomers. Equivalent to "very low" effort. Datasets This issue is related to datasets or the :mod:`nilearn.datasets` module. labels Jul 26, 2018
@GaelVaroquaux
Copy link
Member Author

This has become a major issue: it is breaking all our CI. We unfortunately need to address it urgently.

@KamalakerDadi : do you have some spare cycles?

@KamalakerDadi
Copy link
Contributor

What are files needed to be shipped ? Everything surface plotting uses ?

@GaelVaroquaux
Copy link
Member Author

Listed above. And they need to be gzipped with best compression.

@KamalakerDadi
Copy link
Contributor

Ok. I am on it.

@kchawla-pi kchawla-pi added the Priority: high The task is urgent and needs to be addressed as soon as possible. label Jul 31, 2018
@GaelVaroquaux
Copy link
Member Author

GaelVaroquaux commented Jul 31, 2018 via email

@KamalakerDadi
Copy link
Contributor

Shall we do the same for brainomics localizer data. Shipping only part of data where examples are dependant on ?

Right now it is down again http://brainomics.cea.fr/localizer

@GaelVaroquaux
Copy link
Member Author

GaelVaroquaux commented Jul 31, 2018 via email

@KamalakerDadi
Copy link
Contributor

Does this needs to be uncompressed in home directory in nilearn_data folder ?

@GaelVaroquaux
Copy link
Member Author

GaelVaroquaux commented Jul 31, 2018 via email

@KamalakerDadi
Copy link
Contributor

They can be gzipped separately, and then they don't need to be unzipped: nibabel knows how to read from gzipped files.

Meaning if I give directly the gzipped files to plotting functions it should work, right ?

@KamalakerDadi
Copy link
Contributor

gzipped gifti files doesn't work with nibabel.load or gifti.read ?

@GaelVaroquaux
Copy link
Member Author

GaelVaroquaux commented Jul 31, 2018 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Datasets This issue is related to datasets or the :mod:`nilearn.datasets` module. Good first issue Good for newcomers. Equivalent to "very low" effort. Priority: high The task is urgent and needs to be addressed as soon as possible.
Projects
None yet
Development

No branches or pull requests

3 participants