Tomáš Votroubek

Rigth, then I guess what I'm asking for is a fallback implementation for objects without a strided memory layout. Tensor operations are still well defined even on general views, right?...

Is this still broken? For the QCQP this can be quite important too -- along with propagating the start values of monomials to the lifted variables.

Are you sure? Compare Gurobi's output on the projection done by hand ```jl using JuMP, PolyJuMP, Gurobi m = Model(Gurobi.Optimizer) @variable(m, x >= 0, start=1/sqrt(3)) @variable(m, y >= 0, start=1-start_value(x))...

I'll give you the context, in case it helps your decision. I'm using HC in a randomized iterative algorithm as a fallback method to find Nash equilibria. Edge cases are...

Sure, I'll give it a shot if you want; though I doubt my ability to meaningfully contribute to the project as I have no experience with implementing homotopy methods.

BTW @saschatimme, what did you mean by "at least two variables?" This works `res = solve(System([-x+y, -y+x]))`, but this `res = solve(System([x, y]))` does not. I tried, but failed. It...

While looking into the issue i found broken interpolation in [utils.jl at line 152](https://github.com/JuliaHomotopyContinuation/HomotopyContinuation.jl/blob/e1f39643e7e4129244f695a712079793cc0cc3d3/src/utils.jl#L152) Should be: ```jl function Base.showerror(io::IO, E::KeywordArgumentException) print(io, "Invalid argument $(E.given) for $(E.key). ", E.msg) end ```...