search

geomstats / challenge-iclr-2021

GitHub repository for the ICLR Computational Geometry & Topology Challenge 2021
MIT License
51 stars 25 forks source link

Shape analysis with skeletal representations and principal nested spheres #17

Closed ZhiyLiu closed 3 years ago

ZhiyLiu commented 3 years ago

This work provides python interfaces and examples that contain (1) geometric modeling of non-branching shapes (e.g., the hippocampus) with skeletal representations (s-reps) and (2) a non-Euclidean statistical method principal nested spheres (PNS).

In the experiments, we simulated 3D shapes with deformed ellipsoids. Meanwhile, we compared our methods with other statistical methods. We have included the dataset as well as the script of simulating data in this submission.

review-notebook-app[bot] commented 3 years ago

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

ninamiolane commented 3 years ago

Thank you for this great submission! 🎉

It seems that the Travis test is failing, see error:

jupyter_client.kernelspec.NoSuchKernel: No such kernel named py37

here: https://travis-ci.org/github/geomstats/challenge-iclr-2021/builds/769187653

It may be that you used Python 3.7 for the notebook, whereas the challenge guidelines and tests expect Python 3.8. Could you see if that could be the problem? Thanks, looking forward to merge this!

ZhiyLiu commented 3 years ago

Thanks for the notice. This submission is recreated in #20.

ninamiolane commented 3 years ago

Hi @ZhiyLiu , if you haven't replied yet, please check the issue #39 about asking your permission to be listed as a co-author on the challenge's white paper. Thank you!