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

Wrote additional high level API assembly functions #652

Merged
merged 6 commits into from
Sep 6, 2021
Merged

Wrote additional high level API assembly functions #652

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
dd8b9f9
Wrote additional high level API assembly functions
amartinhuertas Sep 5, 2021
00a8d3e
Update NEWS.md
amartinhuertas Sep 5, 2021
296ff28
Merge branch 'master' of github.com:gridap/Gridap.jl into developed_a…
amartinhuertas Sep 6, 2021
2e7c92c
* Type annotated linear algebra argumentsin the new assembly functions.
amartinhuertas Sep 6, 2021
3455363
Missing type annotations
amartinhuertas Sep 6, 2021
1cc0816
Put f,b args first in all the assembly functions
amartinhuertas Sep 6, 2021
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
3 changes: 3 additions & 0 deletions NEWS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Fixed
- Inheritance relationship for DiscreteModelPortion. Since PR [#645](https://github.com/gridap/Gridap.jl/pull/645).

### Added
- Additional high level API assembly functions. Since PR [#652](https://github.com/gridap/Gridap.jl/pull/652).

## [0.16.4] - 2021-08-17

### Added
Expand Down
44 changes: 44 additions & 0 deletions src/FESpaces/Assemblers.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -270,44 +270,88 @@ function assemble_matrix(f::Function,a::Assembler,U::FESpace,V::FESpace)
assemble_matrix(a,collect_cell_matrix(U,V,f(u,v)))
end

function assemble_matrix!(A,f::Function,a::Assembler,U::FESpace,V::FESpace)
Copy link
Member

Choose a reason for hiding this comment

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

Start the argument list with ::Function objects so that we can use do-block syntax. Also type-annotate the new array arguments, I.e.

assemble_matrix!(f::Function,A::AbstractMatrix,a::Assembler,U::FESpace,V::FESpace)
assemble_matrix_and_vector!(f::Function,b::Function,M::AbstractMatrix,r::AbstractVector,a::Assembler,U::FESpace,V::FESpace)

Apply this minor changes to all new introduced assembly functions.

Copy link
Member Author

Choose a reason for hiding this comment

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

Start the argument list with ::Function objects so that we can use do-block syntax

I though that the parameters that are modified in a !-like function should appear first in the list of parameters by convention. From your comment, I guess this is not strictly necessary, right?

Copy link
Member

Choose a reason for hiding this comment

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

right, see e.g. broadcast

v = get_fe_basis(V)
u = get_trial_fe_basis(U)
assemble_matrix!(A,a,collect_cell_matrix(U,V,f(u,v)))
end

function assemble_vector(f::Function,a::Assembler,V::FESpace)
v = get_fe_basis(V)
assemble_vector(a,collect_cell_vector(V,f(v)))
end

function assemble_vector!(b,f::Function,a::Assembler,V::FESpace)
v = get_fe_basis(V)
assemble_vector!(b,a,collect_cell_vector(V,f(v)))
end

function assemble_matrix_and_vector(f::Function,b::Function,a::Assembler,U::FESpace,V::FESpace)
v = get_fe_basis(V)
u = get_trial_fe_basis(U)
assemble_matrix_and_vector(a,collect_cell_matrix_and_vector(U,V,f(u,v),b(v)))
end

function assemble_matrix_and_vector!(M,r,f::Function,b::Function,a::Assembler,U::FESpace,V::FESpace)
v = get_fe_basis(V)
u = get_trial_fe_basis(U)
assemble_matrix_and_vector!(M,r,a,collect_cell_matrix_and_vector(U,V,f(u,v),b(v)))
end

function assemble_matrix(f,a::Assembler,U::FESpace,V::FESpace)
assemble_matrix(a,collect_cell_matrix(U,V,f))
end

function assemble_matrix!(A,f,a::Assembler,U::FESpace,V::FESpace)
assemble_matrix!(A,a,collect_cell_matrix(U,V,f))
end

function assemble_vector(f,a::Assembler,V::FESpace)
assemble_vector(a,collect_cell_vector(V,f))
end

function assemble_vector!(b,f,a::Assembler,V::FESpace)
assemble_vector!(b,a,collect_cell_vector(V,f))
end

function assemble_matrix_and_vector(f,b,a::Assembler,U::FESpace,V::FESpace)
assemble_matrix_and_vector(a,collect_cell_matrix_and_vector(U,V,f,b))
end

function assemble_matrix_and_vector!(M,r,f,b,a::Assembler,U::FESpace,V::FESpace)
Copy link
Member

Choose a reason for hiding this comment

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

and here (and type-annotate array inputs)

Copy link
Member Author

Choose a reason for hiding this comment

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

oops ... I forgot this one

Copy link
Member Author

Choose a reason for hiding this comment

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

f is not a function, isnt?

assemble_matrix_and_vector!(M,r,a,collect_cell_matrix_and_vector(U,V,f,b))
end

function assemble_matrix(f,U::FESpace,V::FESpace)
a = SparseMatrixAssembler(U,V)
assemble_matrix(f,a,U,V)
end

function assemble_matrix!(A,f,U::FESpace,V::FESpace)
Copy link
Member

Choose a reason for hiding this comment

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

and here

a = SparseMatrixAssembler(U,V)
assemble_matrix!(A,f,a,U,V)
end

function assemble_vector(f,V::FESpace)
a = SparseMatrixAssembler(V,V)
assemble_vector(f,a,V)
end

function assemble_vector!(b,f,V::FESpace)
Copy link
Member

Choose a reason for hiding this comment

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

and here

a = SparseMatrixAssembler(V,V)
assemble_vector!(b,f,a,V)
end

function assemble_matrix_and_vector(f,b,U::FESpace,V::FESpace)
a = SparseMatrixAssembler(U,V)
assemble_matrix_and_vector(f,b,a,U,V)
end

function assemble_matrix_and_vector!(M,r,f,b,U::FESpace,V::FESpace)
Copy link
Member

Choose a reason for hiding this comment

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

and here

a = SparseMatrixAssembler(U,V)
assemble_matrix_and_vector!(M,r,f,b,a,U,V)
end

# Abstract interface for computing the data to be sent to the assembler

function collect_cell_matrix(trial::FESpace,test::FESpace,mat_contributions)
Expand Down
41 changes: 34 additions & 7 deletions test/FESpacesTests/AssemblersTests.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,13 +65,40 @@ A,b = assemble_matrix_and_vector(assem,data)
x = A\b
uh = FEFunction(U,x)

A = assemble_matrix(a,U,V)
b = assemble_vector(ℓ,V)
A,b = assemble_matrix_and_vector(a,ℓ,U,V)

A = assemble_matrix(a(du,dv),U,V)
b = assemble_vector(ℓ(dv),V)
A,b = assemble_matrix_and_vector(a(du,dv),ℓ(dv),U,V)
A1 = assemble_matrix(a,U,V)
b1 = assemble_vector(ℓ,V)
A2,b2 = assemble_matrix_and_vector(a,ℓ,U,V)

A12 = copy(A1); A12[1,1]=rand()
b12 = copy(b1); b12[1]=rand()
A22 = copy(A2); A22[1,1]=rand()
b22 = copy(b2); b22[1]=rand()

tol = 1.e-14
assemble_matrix!(A12,a,U,V)
assemble_vector!(b12,ℓ,V)
assemble_matrix_and_vector!(A22,b22,a,ℓ,U,V)
@test norm(A12-A1) < tol
@test norm(b12-b1) < tol
@test norm(A22-A2) < tol
@test norm(b22-b2) < tol

A1 = assemble_matrix(a(du,dv),U,V)
b1 = assemble_vector(ℓ(dv),V)
A2,b2 = assemble_matrix_and_vector(a(du,dv),ℓ(dv),U,V)

A12 = copy(A1); A12[1,1]=rand()
b12 = copy(b1); b12[1]=rand()
A22 = copy(A2); A22[1,1]=rand()
b22 = copy(b2); b22[1]=rand()

assemble_matrix!(A12,a(du,dv),U,V)
assemble_vector!(b12,ℓ(dv),V)
assemble_matrix_and_vector!(A22,b22,a(du,dv),ℓ(dv),U,V)
@test norm(A12-A1) < tol
@test norm(b12-b1) < tol
@test norm(A22-A2) < tol
@test norm(b22-b2) < tol

V = TestFESpace(
model,
Expand Down