Herman Sletmoen

Maybe it should simplify like this? However, this assumes `x ≠ 0`. ``` 0 ~ ((x - x*ρ)*β) / x + x^2 ---> 0 ~ (1 - ρ)*β + x^2...

I'm gonna close this for now, because I think https://github.com/JuliaSymbolics/Symbolics.jl/pull/1247 was a better way of fixing the issue. Not substituting in non-symbolic expressions would also be breaking. For example, it...

Yep, thanks! I am preparing a PR now.

This was fixed by https://github.com/JuliaSymbolics/Symbolics.jl/pull/1247.

~But here it's the other way around, though.~ EDIT: stupid me, scc is faster.

To me it makes most sense that the plotting (continues to) show a minimal representation of the solution, i.e. only the unknowns, by default. But I do think it sounds...

See also the workaround(s) in https://discourse.julialang.org/t/solving-ensembleproblem-efficiently-for-large-systems-memory-issues/116146/.

> Is the array of things to set all matching in type? It is matching in the sense that all the updated concrete parameter types are subtypes of the declared...

By accident I discovered this. If I `remake` the problem in this _particular_ way, `solve` is type-stable: ```julia newu0, newp = setsym_oop(prob, [sys.I])(prob, Any[Idummy]) newprob = remake(prob; u0 = newu0,...

Thanks for the tip! Those look very relevant for me. Noted for a future refactor :)