search

JuliaSmoothOptimizers / LimitedLDLFactorizations.jl

Limited-Memory Factorization of Symmetric Matrices
Other
21 stars 10 forks source link

Lower triangular with diagonal input matrix #63

Closed geoffroyleconte closed 2 years ago

geoffroyleconte commented 2 years ago

The matrix used by the factorization is now lower triangular with the diagonal, instead of being separated in a strict lower triangular matrix and the diagonal elements.

The lldl and LimitedLDLFactorization function arguments have been modified to be more convenient to use with these changes. In particular, the permutation vector P is now a keyword argument.

adiag was previously computed with diag(A), which produces a SparseVector. It is now a Vector that may have some zeros, because I thought it would be faster to access its elements. This change forces the types of pos and neg to be always Int64 because they are computed with findall, but I don't think this is an issue.

codecov[bot] commented 2 years ago

Codecov Report

Base: 85.99% // Head: 88.20% // Increases project coverage by +2.20% :tada:

Coverage data is based on head (487d28f) compared to base (83b91a5). Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #63 +/- ## ========================================== + Coverage 85.99% 88.20% +2.20% ========================================== Files 1 1 Lines 357 373 +16 ========================================== + Hits 307 329 +22 + Misses 50 44 -6 ``` | [Impacted Files](https://codecov.io/gh/JuliaSmoothOptimizers/LimitedLDLFactorizations.jl/pull/63?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaSmoothOptimizers) | Coverage Δ | | |---|---|---| | [src/LimitedLDLFactorizations.jl](https://codecov.io/gh/JuliaSmoothOptimizers/LimitedLDLFactorizations.jl/pull/63/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaSmoothOptimizers#diff-c3JjL0xpbWl0ZWRMRExGYWN0b3JpemF0aW9ucy5qbA==) | `88.20% <100.00%> (+2.20%)` | :arrow_up: | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaSmoothOptimizers). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaSmoothOptimizers)

:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.

geoffroyleconte commented 2 years ago

Thanks for the reviews!