BlockBandedMatrices.jl icon indicating copy to clipboard operation
BlockBandedMatrices.jl copied to clipboard

A Julia package for representing block-banded matrices and banded-block-banded matrices

Results 23 BlockBandedMatrices.jl issues
Sort by recently updated
recently updated
newest added

Starting from Julia v1.9, `axpy!` is owned by `LinearAlgebra`, and `BLAS.axpy!` does no longer work for non-strided arrays. Also, in the future `AbstractQ` will no longer subtype `AbstractMatrix`, so in...

I have a big `BanedBlockBandedMatrix` and need to multiply it with a vector. However it seems to be rather slow and require a lot of allocation. Here's a MWE ```...

Ideally we'd have: ```julia julia> B = BlockBandedMatrix{Float64}(undef, Fill(2,10), Fill(3, 10), (1, 2)); julia> MemoryLayout(view(B, getindex.(Block.(1:10), 1), getindex.(Block.(1:1), 2))) isa BandedColumns{StridedLayout} ``` To do this, we have several things to...

Allocations increase drastically with newer versions of BandedMatrices.jl, BlockbandedMatrices.jl, LazyArrays.jl and LazyBandedMatrices.jl I do not know where the problem is. $ julia-15 mwe.jl 20.566046 seconds (1.11 M allocations: 65.356 MiB,...

```julia julia> A=BlockBandedMatrix(Zeros(1,1),[1],[1],(0,0)) 1×1-blocked 1×1 BlockSkylineMatrix{Float64,Array{Float64,1},BlockBandedMatrices.BlockSkylineSizes{Tuple{BlockedUnitRange{Array{Int64,1}},BlockedUnitRange{Array{Int64,1}}},Fill{Int64,1,Tuple{Base.OneTo{Int64}}},Fill{Int64,1,Tuple{Base.OneTo{Int64}}},BandedMatrices.BandedMatrix{Int64,Array{Int64,2},Base.OneTo{Int64}},Array{Int64,1}}}: 0.0 julia> A[Block(1)] ERROR: StackOverflowError: Stacktrace: [1] view(::BlockSkylineMatrix{Float64,Array{Float64,1},BlockBandedMatrices.BlockSkylineSizes{Tuple{BlockedUnitRange{Array{Int64,1}},BlockedUnitRange{Array{Int64,1}}},Fill{Int64,1,Tuple{Base.OneTo{Int64}}},Fill{Int64,1,Tuple{Base.OneTo{Int64}}},BandedMatrices.BandedMatrix{Int64,Array{Int64,2},Base.OneTo{Int64}},Array{Int64,1}}}, ::Block{1,Int64}) at C:\Users\pty\.julia\packages\BlockArrays\g01fJ\src\abstractblockarray.jl:129 (repeats 79984 times) ``` An error is expected, but not stack overflow maybe?

I have a banded block banded matrix where the blocks themselves can have different bandwidths. Specifically, the bandwidth within the blocks is dependent on which block-diagonal the block is in....

Usually these types of matrices from from PDEs where they can be constructed from the Kronecker product of the actions along each dimension, like: ```julia using SparseArrays Iy = SparseMatrixCSC(I,N,N)...

```julia using FillArrays, BlockBandedMatrices l,u = 1,1 # block bandwidths N = M = 2 # number of row/column blocks cols = rows = 1:N # block sizes jac_bbm =...