solve the nonlinear equation with only one variable

[GTang]

Dear NLsolve. experts,

Is it possible to solve a nonlinear equation with only one variable?
When I try to implement the following function:

using NLsolve
function fun!(F, x)
F = (x+3)*(x^3-7)+18
end

and solve the nonlinear equation as: nlsolve(fun!, 0.1), it gives me the error message:
"ERROR: MethodError: no method matching nlsolve(::typeof(fun!), ::Float64)
Closest candidates are: nlsolve(::Any, ::Any, ::AbstractArray; inplace, kwargs...) ..."

Thanks!

[Fredrik Ekre]

NLsolve only solve vector problem AFAIK, but you can define a problem of size 1:

julia> using NLsolve

julia> function fun!(F, x)
           F[1] = (x[1] + 3) * (x[1]^3 - 7) + 18
       end
fun! (generic function with 1 method)

julia> nlsolve(fun!, [0.1])
Results of Nonlinear Solver Algorithm
 * Algorithm: Trust-region with dogleg and autoscaling
 * Starting Point: [0.1]
 * Zero: [-0.464978121172224]
 * Inf-norm of residuals: 0.000000
 * Iterations: 6
 * Convergence: true
   * |x - x'| < 0.0e+00: false
   * |f(x)| < 1.0e-08: true
 * Function Calls (f): 7
 * Jacobian Calls (df/dx): 7

[GTang]

NLsolve only solve vector problem AFAIK, but you can define a problem of size 1:

julia> using NLsolve

julia> function fun!(F, x)
           F[1] = (x[1] + 3) * (x[1]^3 - 7) + 18
       end
fun! (generic function with 1 method)

julia> nlsolve(fun!, [0.1])
Results of Nonlinear Solver Algorithm
 * Algorithm: Trust-region with dogleg and autoscaling
 * Starting Point: [0.1]
 * Zero: [-0.464978121172224]
 * Inf-norm of residuals: 0.000000
 * Iterations: 6
 * Convergence: true
   * |x - x'| < 0.0e+00: false
   * |f(x)| < 1.0e-08: true
 * Function Calls (f): 7
 * Jacobian Calls (df/dx): 7

Thanks for help! It works!

[Patrick Kofod Mogensen]

That, or try NonlinearSolve.jl or Roots.jl :) It's probably worth adding such methods here at some point, but it will actually be part of NLSolvers.jl in that case. I'll open an issue there.