Yingbo Ma comments

Results 223 comments of


                                            Yingbo Ma

Julia 1.9.0 doesn't drop zeros when assigning to a sparse array only when the elements are Big

Yeah, this solution sounds good.

Julia 1.9.0 doesn't drop zeros when assigning to a sparse array only when the elements are Big

In MTK, we had to add an explicit `dropzero!` because of this change https://github.com/SciML/ModelingToolkit.jl/blob/05525b540820569c01d2249096dbbc5880d943d0/src/structural_transformation/bareiss.jl#L146-L148 ```julia julia> using ModelingToolkit, BenchmarkTools, SparseArrays julia> A = BigInt.(sprand(Int, 100, 100, 0.01)) + I; julia>...

More structural transforms

Fine Dulmage-Mendelsohn decompositions are just another name for BLT or SCC.

Creating variables that are vectors of symbolic variables is not documented

```julia @named fol_separate = ODESystem([ RHS ~ (1 - x)/τ, D(x) ~ RHS ]) prob = ODEProblem(structural_simplify(fol_separate), [x=>0.0, RHS=>0.0], (0.0,10.0), [τ => 3.0]) ```

ParentScope() of parameters half works

Is there an MWE?

(Possibly) move `wrap` and `unwrap` here

Yeah. I think it's fine. It won't be breaking either.

Exception: EXCEPTION_ACCESS_VIOLATION

I am not sure about the cause, and I don't have a Windows machine. Does it crash Julia every time when you run it?

macro for `@.. thread=true`

You can make a PR if you want this feature.

Speedup _observed_ with dynamic broadcasting

Should we fix this inaccuracy by inserting a sleep call in the dynamic broadcasting branch?

Differentiating concretely typed function signatures

Yes, Shashi and I are looking into IRTools.jl and source-to-source transformation. But before that, we want to solve #33 first to make ForwardDiff2 functional.