-
Notifications
You must be signed in to change notification settings - Fork 3
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
Return Symmetric
Hessians by default?
#154
Comments
For SparseMatrixColorings, the type of the sparsity pattern However, the sparsity pattern is also an important information for the type of the future decompressed matrix In my view, we're rather unlikely to decompress in-place into a Related: |
Solvers (with recency bias of ipopt) typically only require the lower or upper triangular, and for me the breakage wouldn't be a big problem yet. Based on the results in the linked PR this gives a median ~15% improvement in the performance which seems pretty useful to me! |
The issue is in the coloring, we need both triangles. |
Alright, let's keep the default |
In #151, the question popped up whether
hessian_sparsity
should return aSymmetric{Bool, SparseMatrixCSC{Bool, Int64}}
instead of the currentSparseMatrixCSC{Bool, Int64}
.Since this can be considered a breaking change, this feature would have to wait until we have more reasons to tag a breaking
v0.7
release.@gdalle @Vaibhavdixit02 @amontoison I would be interested in your opinions on this!
Related issue: #14
The text was updated successfully, but these errors were encountered: