Herman Sletmoen

The `true` is gone now, but ```julia using Symbolics F = [0.0 1.0; -1.0-10.0*cos(t) -24.0-19.0*sin(t)] inv(F) ``` outputs mixed floats and integers: ```julia 2×2 Matrix{Num}: (-24.0 - 19.0sin(t)) / (1.0...

Now ```julia using SymbolicUtils @syms a simplify(a/a) ``` gives ```julia 1 ```

All three examples given here work now. Maybe this can be closed?

After JuliaAlgebra/MultivariatePolynomials.jl/pull/296 ```julia using SymbolicUtils @syms x simplify_fractions(1/x + 1/(x^2 - x)), simplify_fractions(1/x + 1/(x^2 - 2x)) ``` returns only integer/rational coefficients: ```julia (1 / (-(1//1) + x), (-(1//1) +...

After JuliaAlgebra/MultivariatePolynomials.jl/pull/296 ```julia using Symbolics @variables t B = (((t^2)^3)*(((800//1)*(t^3) / ((800//1)*(t^3) - (14//1)*t))^2)) / ((( (60//1)*(t^2))^2) * ( (14//1)*t)^3) B = Symbolics.simplify(B) ``` now gives rational coefficients: ```julia (50(t^3))...

If possible, it would be great to test that everything works seamlessly through autodiff as well, using recent improvements like #2686. In case it can be of use, here is...

But the constraint `s => 1.0` is not respected in the latter case. See also #2747.

I am really looking forward to this 😀 Thanks!

I think the reason for the inconsistencies is how [`check_parameters()`](https://github.com/SciML/ModelingToolkit.jl/blob/c723155c8b1c8e72f95e889d2b63beb9e8007d9e/src/parameters.jl#L7) are determining parameters based on 1. if they are callable, like `r(t)` 2. if they are arrays `r(t)[1:3]` I guess...

This was fixed by JuliaAlgebra/MultivariatePolynomials.jl/pull/296 😃 ```julia using Symbolics @variables t a(t) D(expr) = expand_derivatives(Differential(t)(expr)) simplify_fractions(D(D(a)/a) + D(D(a)/a)) ``` now gives rational coefficients: ```julia ((2//1)*a(t)*Differential(t)(Differential(t)(a(t))) - (2//1)*(Differential(t)(a(t))^2)) / (a(t)^2) ```...