using ExprOptimization

const grammar = @grammar begin
    Real = x
    Real = Real * Real
    Real = Real + Real
    Real = Real - Real
    Real = |(1:5)
end

const S = SymbolTable(grammar)

ground_truth(x) = x*x + 2x + 1
function loss(tree::RuleNode, grammar::Grammar)
    ex = get_executable(tree, grammar)
    los = 0.0
    for x = -5.0:1.0:5.0
        S[:x] = x
        los += abs2(Core.eval(S,ex) - ground_truth(x))
    end
    los
end

?MonteCarlo

using Random
Random.seed!(10)
p = MonteCarlo(20000, 6)
results_mc = optimize(p, grammar, :Real, loss)
(results_mc.expr, results_mc.loss)

display(results_mc.tree, grammar)

?GeneticProgram

Random.seed!(1)
p = GeneticProgram(1000,20,6,0.3,0.3,0.4)
results_gp = optimize(p, grammar, :Real, loss)
(results_gp.expr, results_gp.loss)

display(results_gp.tree, grammar)

?GrammaticalEvolution

Random.seed!(1)
p = GrammaticalEvolution(grammar,:Real,1000,20,10,10,6,0.2,0.4,0.4; select_method=GrammaticalEvolutions.TruncationSelection(300))
results_ge = optimize(p, grammar, :Real, loss)
(results_ge.expr, results_ge.loss)

display(results_ge.tree, grammar)

?CrossEntropy

Random.seed!(1)
p = CrossEntropy(1000,20,6,500)
results_ce = optimize(p, grammar, :Real, loss)
(results_ce.expr, results_ce.loss)

display(results_ce.tree, grammar)

?PIPE

Random.seed!(3)
p = PIPE(PPT(0.8),1000,20,0.2,0.1,0.05,1,0.2,0.6,0.999,6)
results_pipe = optimize(p, grammar, :Real, loss)
(results_pipe.expr, results_pipe.loss)

display(results_pipe.tree, grammar)