8000 ENH: get sparse tensors by stsievert · Pull Request #55 · tensorly/tensorly · GitHub
[go: up one dir, main page]
Skip to content

ENH: get sparse tensors #55

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.

Sign up for GitHub

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

Jump to bottom
Closed
wants to merge 5 commits into from
Closed

ENH: get sparse tensors #55

wants to merge 5 commits into from

Conversation

stsievert
Copy link
Contributor

This implements getting sparse tensors from FROSTT, http://frostt.io/. It stores these tensors as sparse tensors using https://sparse.pydata.org/.

Because it relies on pydata/sparse, these datasets will only be effective with the NumPy backend.

@JeanKossaifi
Copy link
Member

Thanks @stsievert, this is a great addition.

For now, it is probably be best to put in the tensorly/contrib submodule (e.g. tensorly.contrib.frostt) as it breaks the backend system (works only for NumPy).

I am planning to have sparse tensor support added in TensorLy so this would be a great match (ideally supporting all backends but in the short term might need to implement a backend specific solution). Until this is added, we should probably check that the current backend is NumPy. I could also add a tensorly-wide variable to store the folder in which data is stored.

@JeanKossaifi JeanKossaifi mentioned this pull request Jul 3, 2018
Closed
@stsievert
Copy link
Contributor Author

I am planning to have sparse tensor support added in TensorLy

I have similar plans – there's some interest in sparse tensor decomposition, and I'm involved with some of that.

This PR doesn't have support for column labels (though they're provided in the dataset), though there is a relevant xarray issue: pydata/xarray#1375.

@coveralls
Copy link

Coverage Status

Coverage increased (+0.04%) to 96.974% when pulling 3622c4e on stsievert:datasets-frostt into d93d280 on tensorly:master.

@coveralls
Copy link
coveralls commented Jul 4, 2018 •
edited
Loading

Coverage Status

Coverage increased (+0.04%) to 96.974% when pulling 3622c4e on stsievert:datasets-frostt into d93d280 on tensorly:master.

@stsievert stsievert mentioned this pull request Aug 6, 2018
Closed
@stsievert stsievert closed this Aug 6, 2018
5DF5
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@stsievert @JeanKossaifi @coveralls
0