Add elimination algorithms mcs, rcm, and mmd.
#1855
Conversation
|
I saw the issue with |
|
More algorithms here. |
|
Hi, acknowledged thanks -- will look soon. Thanks for the contribution, more elimination algorithms are definitely good! |
|
Verson |
|
Update: I haven't forgotten -- been extra busy here with passing of a family member. Looking to work through this during travel this weekend. Apologies for the delay. |
|
I am sorry to hear that. Please take your time! |
|
Thanks for the contribution. Hangup was new ManifoldsBase v1 update happening simultaneously in related repos. Having to fix them all together. Apologies for the delay |
| @@ -15,8 +15,8 @@ if TEST_GROUP in ["all", "basic_functional_group"] | |||
| include("testSpecialEuclidean2Mani.jl") | |||
| include("testEuclidDistance.jl") | |||
| # gradient / jacobian tests | |||
| include("manifolds/manifolddiff.jl") | |||
| include("manifolds/factordiff.jl") | |||
| #include("manifolds/manifolddiff.jl") | |||
There was a problem hiding this comment.
will resolve these in separate PRs
This pull request adds some additional elimination algorithms:
mcs: maximum cardinality searchrcm: reverse Cuthill-Mckeemmd: multiple minimum degreeThe first two algorithms were implemented by me, and the third was taken from Sparspak.jl and lightly edited. The third algorithm, in particular, is quite good: it was the state of the practice before AMD, and it has the advantage of being MIT licensed.