Skip to content
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.

Sign up for GitHub

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

Jump to bottom

make similar(A<:HermOrSym, opts...) consistently preserve A's storage type and uplo flag #24163

Merged
merged 1 commit into from
Oct 19, 2017
Merged

make similar(A<:HermOrSym, opts...) consistently preserve A's storage type and uplo flag #24163

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 11 additions & 8 deletions base/linalg/symmetric.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -137,16 +137,19 @@ function setindex!(A::Hermitian, v, i::Integer, j::Integer)
end
end

similar(A::Symmetric, ::Type{T}) where {T} = Symmetric(similar(A.data, T))
# Hermitian version can be simplified when check for imaginary part of
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please keep some version of the comment

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Comment restored :). Thanks!

# diagonal in Hermitian has been removed
# For A<:Union{Symmetric,Hermitian}, similar(A[, neweltype]) should yield a matrix with the same
# symmetry type, uplo flag, and underlying storage type as A. The following methods cover these cases.
similar(A::Symmetric, ::Type{T}) where {T} = Symmetric(similar(parent(A), T), ifelse(A.uplo == 'U', :U, :L))
# If the the Hermitian constructor's check ascertaining that the wrapped matrix's
# diagonal is strictly real is removed, the following method can be simplified.
function similar(A::Hermitian, ::Type{T}) where T
B = similar(A.data, T)
for i = 1:size(A,1)
B[i,i] = 0
end
return Hermitian(B)
B = similar(parent(A), T)
for i in 1:size(B, 1) B[i, i] = 0 end
return Hermitian(B, ifelse(A.uplo == 'U', :U, :L))
end
# On the other hand, similar(A, [neweltype,] shape...) should yield a matrix of the underlying
# storage type of A (not wrapped in a symmetry type). The following method covers these cases.
similar(A::Union{Symmetric,Hermitian}, ::Type{T}, dims::Dims{N}) where {T,N} = similar(parent(A), T, dims)

# Conversion
convert(::Type{Matrix}, A::Symmetric) = copytri!(convert(Matrix, copy(A.data)), A.uplo)
Expand Down
14 changes: 14 additions & 0 deletions test/linalg/symmetric.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -444,3 +444,17 @@ end
@test norm(inv(Hermitian(H))*(H*ones(8)) .- 1) ≈ 0 atol = 1e-5
end
end

@testset "similar should preserve underlying storage type and uplo flag" begin
m, n = 4, 3
sparsemat = sprand(m, m, 0.5)
for SymType in (Symmetric, Hermitian)
symsparsemat = SymType(sparsemat)
@test isa(similar(symsparsemat), typeof(symsparsemat))
@test similar(symsparsemat).uplo == symsparsemat.uplo
@test isa(similar(symsparsemat, Float32), SymType{Float32,<:SparseMatrixCSC{Float32}})
@test similar(symsparsemat, Float32).uplo == symsparsemat.uplo
@test isa(similar(symsparsemat, (n, n)), typeof(sparsemat))
@test isa(similar(symsparsemat, Float32, (n, n)), SparseMatrixCSC{Float32})
end
end