src: add AutoSparseDI option for jacobian based on DI - #22
Open
rveltz wants to merge 4 commits into
Open
Conversation
gdalle
suggested changes
Mar 22, 2026
Comment on lines
+18
to
+21
| # we use the following instead of TracerSparsityDetector | ||
| # because of searchsortedfirst in POSolution | ||
| # sparsity_detector = TracerLocalSparsityDetector(), | ||
| sparsity_detector = DI.DenseSparsityDetector(backend, atol = 1e-8), |
There was a problem hiding this comment.
Note that this makes the sparsity pattern dependent on the actual values inside the vector, which may not be a good idea?
Member
Author
There was a problem hiding this comment.
I am mostly trying to get this to work for now. Nothing of the above works...
The corresponding test is in test_DI.jl
There was a problem hiding this comment.
Can you paste the error messages you get?
Member
Author
There was a problem hiding this comment.
When using TracerLocalSparsityDetector in SparseDIExt, I get
ERROR: PreparationMismatchError (inconsistent types between preparation and execution):
- f!: ✅
- y: ✅
- backend: ✅
- x: ✅
- contexts: ❌
- prep: Tuple{Constant{PeriodicOrbitOCollProblem{ConstantDDEBifProblem{BifFunction{typeof(Hutchinson), Nothing, Nothing, Nothing, typeof(JacHutchinson), Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Float64, BifurcationKit.Jet{Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing}}, typeof(delaysF), Vector{Float64}, Vector{Float64}, @NamedTuple{a::Float64, d::Int64, τ::Float64, Δ::SparseMatrixCSC{Float64, Int64}, N::Int64}, PropertyLens{:a}, typeof(BifurcationKit.plot_default), typeof(BifurcationKit.record_sol_default), typeof(BifurcationKit.save_solution_default), Float64}, DDEBifurcationKit.AutoSparseDI, Float64, Vector{Float64}, Matrix{Float64}, Nothing}}, Constant{@NamedTuple{a::Float64, d::Int64, τ::Float64, Δ::SparseMatrixCSC{Float64, Int64}, N::Int64}}}
- exec: Tuple{Constant{@NamedTuple{a::Float64, d::Int64, τ::Float64, Δ::SparseMatrixCSC{Float64, Int64}, N::Int64}}}
If you are confident that this check is superfluous, you can disable it by running preparation with the keyword argument `strict=Val(false)` inside DifferentiationInterface.
There was a problem hiding this comment.
Sounds like you're trying to detect sparsity through another differentiation operation?
Member
Author
There was a problem hiding this comment.
No, the function is not.
And for TracerSparsityDetector, I get:
ERROR: TypeError: non-boolean (SparseConnectivityTracer.GradientTracer{Int64, BitSet}) used in boolean context
Stacktrace:
[1] searchsortedfirst
@ ./sort.jl:177 [inlined]
[2] searchsortedfirst
@ ./sort.jl:290 [inlined]
[3] searchsortedfirst
@ ./sort.jl:292 [inlined]
[4] __interpolate_posolution(coll::PeriodicOrbitOCollProblem{…}, t0::SparseConnectivityTracer.GradientTracer{…}, x::Vector{…}, period::Int64)
@ BifurcationKit ~/work/prog_gd/julia/dev/dev1/bkorg/BifurcationKitGITHUB/src/periodicorbit/PeriodicOrbitCollocation.jl:1173
[5] functional_coll!(coll::PeriodicOrbitOCollProblem{…}, outc::Base.ReshapedArray{…}, uc::Base.ReshapedArray{…}, period::SparseConnectivityTracer.GradientTracer{…}, ::Tuple{…}, pars::@NamedTuple{…}, u::Vector{…})
@ DDEBifurcationKit ~/work/prog_gd/julia/dev/dev1/bkorg/DDEBifurcationKit.jl/src/periodicorbit/PeriodicOrbitCollocation.jl:70
[6] residual!(coll::PeriodicOrbitOCollProblem{…}, result::Vector{…}, u::Vector{…}, pars::@NamedTuple{…})
@ DDEBifurcationKit ~/work/prog_gd/julia/dev/dev1/bkorg/DDEBifurcationKit.jl/src/periodicorbit/PeriodicOrbitCollocation.jl:10
[7] coll_residual_for_di!
@ ~/work/prog_gd/julia/dev/dev1/bkorg/DDEBifurcationKit.jl/ext/SparseDIExt.jl:8 [inlined]
There was a problem hiding this comment.
Comment on lines
+100
to
+102
| function BK._generate_jacobian(coll::PeriodicOrbitOCollProblem{Tprob}, ::AutoSparseDI, orbitguess, pars; k...) where {Tprob <: AbstractDDEBifurcationProblem} | ||
| error("You need to import `DifferentiationInterface, SparseConnectivityTracer, SparseMatrixColorings` in order to use this jacobian") | ||
| end |
There was a problem hiding this comment.
I find it more idiomatic to throw a MethodError instead, perhaps with an error hint like this one https://github.com/JuliaDiff/DifferentiationInterface.jl/blob/main/DifferentiationInterface/src/init.jl
Comment on lines
+18
to
+20
| DifferentiationInterface = "a0c0ee7d-e4b9-4e03-894e-1c5f64a51d63" | ||
| SparseConnectivityTracer = "9f842d2f-2579-4b1d-911e-f412cf18a3f5" | ||
| SparseMatrixColorings = "0a514795-09f3-496d-8182-132a7b665d35" |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.