Symbolic regression examples not working in version 0.11.1

[froucoux]

None of the examples provided in the Jupyter notebook entitled "Symbolic Regression with Genetic Programming" seem to work.

I used the last available version (0.11.1) of the package and tested it with Julia 1.7.2 and 1.6.5 LTS under Windows and Linux.

In the following example (the first one) :

`
using Evolutionary
using Random
using Plots
Plots.gr()
default(fmt = :png)

Random.seed!(42);
d, n = 1, 20
Nguyen1(x) = x * x * x + x * x + x
xs = sort!(2 * rand(n) .- 1)

syms = [:x]

funcs = Function[+, -, *, /]

fitobj(expr) = sum(abs2.(Nguyen1.(xs) - Evolutionary.Expression(expr).(xs))) / length(xs) |> sqrt

Random.seed!(987498737423);
res = Evolutionary.optimize(fitobj,
TreeGP(50, Terminal[syms...], funcs,
mindepth = 1,
maxdepth = 4,
optimizer = GA(
selection = uniformranking(5),
ɛ = 0.1,
mutationRate = 0.95,
crossoverRate = 0.05,
),
)
)
`

It returns :

ERROR: LoadError: MethodError: no method matching TreeGP(; populationSize=50, terminals=Dict(:x => 1), functions=Dict{Function, Int64}((-) => 2, (/) => 2, (*) => 2, (+) => 2), mindepth=1, maxdepth=4, optimizer=GA[P=50,x=0.05,μ=0.95,ɛ=0.1]) Closest candidates are: TreeGP(; populationSize, terminals, functions, mindepth, maxdepth, crossover, mutation, selection, crossoverRate, mutationRate, initialization, simplify, metrics) at /usr/local/julia/share/julia/base/util.jl:478 got unsupported keyword argument "optimizer" TreeGP(::Integer, ::Vector{Union{Function, Real, Symbol}}, ::Vector{Function}; kwargs...) at ~/.julia/packages/Evolutionary/65hL6/src/gp.jl:40 TreeGP(::Integer, ::Dict{Union{Function, Real, Symbol}, Int64}, ::Dict{Function, Int64}, ::Int64, ::Int64, ::Function, ::Function, ::Function, ::Real, ::Real, ::Symbol, ::Union{Nothing, Function}, ::Vector{ConvergenceMetric}) at ~/.julia/packages/Evolutionary/65hL6/src/gp.jl:26 got unsupported keyword arguments "populationSize", "terminals", "functions", "mindepth", "maxdepth", "optimizer" ... Stacktrace: [1] kwerr(kw::NamedTuple{(:populationSize, :terminals, :functions, :mindepth, :maxdepth, :optimizer), Tuple{Int64, Dict{Symbol, Int64}, Dict{Function, Int64}, Int64, Int64, GA{Evolutionary.var"#uniformrank#252"{Evolutionary.var"#uniformrank#251#253"{Int64}}, typeof(Evolutionary.genop), typeof(Evolutionary.genop)}}}, args::Type) @ Base ./error.jl:163 [2] TreeGP(pop::Int64, term::Vector{Union{Function, Real, Symbol}}, func::Vector{Function}; kwargs::Base.Pairs{Symbol, Any, Tuple{Symbol, Symbol, Symbol}, NamedTuple{(:mindepth, :maxdepth, :optimizer), Tuple{Int64, Int64, GA{Evolutionary.var"#uniformrank#252"{Evolutionary.var"#uniformrank#251#253"{Int64}}, typeof(Evolutionary.genop), typeof(Evolutionary.genop)}}}}) @ Evolutionary ~/.julia/packages/Evolutionary/65hL6/src/gp.jl:43 [3] top-level scope @ /home/julia/projects/first_project/benchmark.jl:19 in expression starting at /home/julia/projects/first_project/benchmark.jl:19

The correct code seems to be :

`
using Evolutionary
using Random
using Plots
Plots.gr()
default(fmt = :png)

Random.seed!(42);
d, n = 1, 20
Nguyen1(x) = x * x * x + x * x + x
xs = sort!(2 * rand(n) .- 1)

syms = [:x]

funcs = Function[+, -, *, /]

fitobj(expr) = sum(abs2.(Nguyen1.(xs) - Evolutionary.Expression(expr).(xs))) / length(xs) |> sqrt

Random.seed!(987498737423);
res = Evolutionary.optimize(fitobj,
TreeGP(50, Terminal[syms...], funcs,
mindepth = 1,
maxdepth = 4,
selection = uniformranking(5),
mutationRate = 0.95,
crossoverRate = 0.05
)
)
`

Questions:

• how to specify the epsilon parameter of the GA ?
• Could you correct the examples ?
• Could you correct the documentation of the TreeGP constructor? The keyword "optimizer" is not working.

Thanks in advance and keep doing the good work on the library!

[LauraBMo]

I have the same problem. This is the relevant commit e05171f. There is no comment, so it is hard to guess the goal and how to specify any parameter of the GA(). Moreover, it seems that both GA and TreeGP use the same crossover and mutate functions, see gp.jl#L126C1-L135C4..

Later, there is 640c1ea modifying the doc string (it is not included in the last release yet, install with ]add Evulutionary#master). Maybe, an author of it (@agctute, @DilumAluthge, @timholy , @shindig7 , @Nagefire) could update the example so it shows the new intended way to pass the parameters to GA .