Herman Sletmoen

Hi. Here is a workaround using an intermediate symbolic array function for evaluating `interp`: ```julia using ModelingToolkit, DataInterpolations, OrdinaryDiffEq using ModelingToolkit: t_nounits as t, D_nounits as D evalinterp(interp::DataInterpolations.AbstractInterpolation, t) =...

For the purpose of just taking up less space in the documentation page, I suppose one could just increase `dtmin` and/or decrease `tspan[2]` to exit after fewer errors.

I suspect what happens is that `KLUFactorization` stores some "null factorization" at construction, and errors when it encounters the true ODE Jacobian for the first time?

Yep, forgot to close.

What's the state of this now @ChrisRackauckas? I find that the saving by `dense` or `saveat` can be an annoying source of allocations and unpredictable GC runtimes when solving many...

It is possible to emulate something like this using callbacks, e.g.: ```julia using OrdinaryDiffEq, BenchmarkTools, Test # 1) allocate inside solver at every time step function f(du, u, p, t)...

> Discuss `complete` semantics Should `complete()` be changed to use `flatten = false` by default? I find it undesireable that the current default `flatten = true` destroys the hierarchical model...

> More importantly, `structural_simplify` using it to figure out what is a parameter and what isn't. ODESystem can use the metadata to do this, but `structural_simplify` should respect whatever bucket...