Kirill Zubov

> Thanks. Could you help me with the writeup? LaTeX isn't compiling in the markdown, and I don't know where to upload an image of the solution. documenter.jl compiles in...

the same in `main` branch https://github.com/SciML/NeuralPDE.jl/blob/master/docs/src/pinn/2D.md but in docs it is already compiled

@mohsenhos Interesting, it should not depend on the order, most likely a typo in the script. Could you share the full script? what it is? u(0:1,0)

it is not supported grammar ```julia julia> u(range(0, step=0.05, stop=1),0) u(0.0:0.05:1.0, 0) ``` there is a bug and parsing like `u(x, 0)`

oh no , it is weird I'm surprised but it really creates a set of the point corresponding to StepRangeLen `u(range(0, step=0.05, stop=1),0)` for GridTraining. Although this was not planned...

```julia @parameters t, x @variables u(..) ∂t, ∂x = Differential(t), Differential(x) eq = ∂t(u(t, x)) + ∂x(u(t, x)) ~ 0 domains = [t ∈ Interval(0.0, 1.0), x ∈ Interval(0.0, 1.0)]...

```julia @parameters t, x @variables u(..) ∂t, ∂x = Differential(t), Differential(x) eq = ∂t(u(t, x)) + ∂x(u(t, x)) ~ 0 domains = [t ∈ Interval(0.0, 1.0), x ∈ Interval(0.0, 1.0)]...

it's typo in bound cond ```julia bcs = [u(0, x) ~ cos(x), Dx(u(t, 0)) ~ 0, Dx(u(t, 1)) ~ exp(-t) * -sin(1)] analytic_sol_func(t, x) = exp.(-t) * cos.(x) ```

and here `remake(prob, p=res.minimizer)` correct `remake(prob, u0=res.minimizer)`

```julia # Parameters, variables, and derivatives @parameters t x @variables u(..) Dt = Differential(t) Dx = Differential(x) Dxx = Differential(x)^2 # 1D PDE and boundary conditions eq = Dt(u(t, x))...