-
Notifications
You must be signed in to change notification settings - Fork 24.3k
[BUG] Incorrect behavior of sparse matrix-matrix multiplication #6219
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
Labels
Comments
This was referenced Apr 3, 2018
This issue is caused by the following:
After chatted to @zou3519, we can define a sparse tensor with coalesced=true when:
And set |
zdevito
pushed a commit
to zdevito/ATen
that referenced
this issue
Aug 15, 2018
…iants (#10496) Summary: - fixes pytorch/pytorch#6219 - removed invariants at pytorch/pytorch#4707 - assume a sparse tensor with coalesced=true when: 1. its elements are unique and 2. the indices are in sorted order Pull Request resolved: pytorch/pytorch#10496 Differential Revision: D9311214 Pulled By: weiyangfb fbshipit-source-id: 167fa5a8e9e5f9c800db02f728a1194029f7e4f3
goodlux
pushed a commit
to goodlux/pytorch
that referenced
this issue
Aug 15, 2018
…iants (pytorch#10496) Summary: - fixes pytorch#6219 - removed invariants at pytorch#4707 - assume a sparse tensor with coalesced=true when: 1. its elements are unique and 2. the indices are in sorted order Pull Request resolved: pytorch#10496 Differential Revision: D9311214 Pulled By: weiyangfb fbshipit-source-id: 167fa5a8e9e5f9c800db02f728a1194029f7e4f3
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
This PyTorch bug was introduced by #4707.
Unlike claimed there,
indices
should not only be unique, but also be sorted in coalesced matrix for correct behavior of matrix multiplication. This raises a major issue when performing matrix multiplication with coalesced then transposed matrix.The following is tested on 2e156f3, but any version after 5e72d7a will have the same behavior.
I will create a pull request regarding this and #6171 later today.
The text was updated successfully, but these errors were encountered: