TSVD.jl
TSVD.jl copied to clipboard
Truncated singular value decomposition with partial reorthogonalization
I am not exactly sure if this is correct, but I would expect the result to be close to zero. ```julia using TSVD, LinearAlgebra A = randn(10, 10) U, S,...
Hey there, would it be possible to support arbitrary types? As it stands this doesn't work: ```julia-repl julia> using TSVD julia> using DoubleFloats julia> a = rand(Double64, 4, 5) 4×5...
If a sparse matrix isn't in the local MatrixDepot cache then the `matrixdepot` function outputs some extra lines related to the download that causes doctests failure. To work around that,...
Hi, I've been using TSVD extensively -- thanks for this great package. However, I keep getting the following error when calling tsvd. Below is a minimal example in which failure...
using LinearAlgebra, TSVD X=rand(4,5) X=X'X F=TSVD.teig(X) X*F[2] ≈ F[2]*diagm(F[1]) false
See #25.
In the singular case, TSVD.jl appears to produce an error, while [PROPACK.jl](https://github.com/JuliaSmoothOptimizers/PROPACK.jl?tab=readme-ov-file) provides the expected results. I’m wondering if this issue is something that could be easily addressed. ```julia julia>...
```julia julia> U2,S2,V2=tsvd(A) ([-0.6544776144364567; -0.2676920846823157; -0.26769208468231576; -0.6544776144364568;;], [2.853906087544359], [-0.6544776144364567; -0.2676920846823159; -0.2676920846823158; -0.6544776144364568;;]) julia> S2 1-element Vector{Float64}: 2.853906087544359 julia> U2,S2,V2=tsvd(A,Dcut) ([0.654477614436457 0.6544776144364569 … -0.49252301879505456 -0.19490429776098736; 0.2676920846823159 0.2676920846823158 … 0.0845812528300824 0.9612271076541434; 0.267692084682316...
Bumps [codecov/codecov-action](https://github.com/codecov/codecov-action) from 4 to 5. Release notes Sourced from codecov/codecov-action's releases. v5.0.0 v5 Release v5 of the Codecov GitHub Action will use the Codecov Wrapper to encapsulate the CLI....