Avoid specialization on search options

src/MLJInterface.jl

@@ -32,7 +32,7 @@ abstract type AbstractSRRegressor <: MMI.Deterministic end
 """Generate an `SRRegressor` struct containing all the fields in `Options`."""
 function modelexpr(model_name::Symbol)
     struct_def = :(Base.@kwdef mutable struct $(model_name){
-        D<:AbstractDimensions,L,use_recorder,N<:AbstractExpressionNode
+        D<:AbstractDimensions,L,N<:AbstractExpressionNode
     } <: AbstractSRRegressor
         niterations::Int = 10
         node_type::Type{N} = Node

src/Options.jl

@@ -419,7 +419,7 @@ function Options end
     optimizer_probability::Real=0.14,
     optimizer_iterations::Union{Nothing,Integer}=nothing,
     optimizer_options::Union{Dict,NamedTuple,Optim.Options,Nothing}=nothing,
-    val_recorder::Val{use_recorder}=Val(false),
+    use_recorder::Bool=false,
     recorder_file::AbstractString="pysr_recorder.json",
     early_stop_condition::Union{Function,Real,Nothing}=nothing,
     timeout_in_seconds::Union{Nothing,Real}=nothing,
@@ -435,7 +435,7 @@ function Options end
     npopulations::Union{Nothing,Integer}=nothing,
     npop::Union{Nothing,Integer}=nothing,
     kws...,
-) where {use_recorder}
+)
     for k in keys(kws)
         !haskey(deprecated_options_mapping, k) && error("Unknown keyword argument: $k")
         new_key = deprecated_options_mapping[k]
@@ -753,14 +753,7 @@ function Options end
     @assert print_precision > 0
 
     options = Options{
-        eltype(complexity_mapping),
-        typeof(operators),
-        use_recorder,
-        typeof(optimizer_options),
-        typeof(optimizer_algorithm),
-        turbo,
-        bumper,
-        typeof(tournament_selection_weights),
+        eltype(complexity_mapping),turbo,bumper,typeof(tournament_selection_weights)
     }(
         operators,
         bin_constraints,
@@ -822,6 +815,7 @@ function Options end
         nested_constraints,
         deterministic,
         define_helper_functions,
+        use_recorder,
     )
 
     return options

src/OptionsStruct.jl

@@ -1,7 +1,7 @@
 module OptionsStructModule
 
 using Optim: Optim
-using DynamicExpressions: AbstractOperatorEnum
+using DynamicExpressions: AbstractOperatorEnum, OperatorEnum
 using LossFunctions: SupervisedLoss
 
 import ..MutationWeightsModule: MutationWeights
@@ -38,17 +38,8 @@ function ComplexityMapping(;
     )
 end
 
-struct Options{
-    CT,
-    OP<:AbstractOperatorEnum,
-    use_recorder,
-    OPT<:Optim.Options,
-    OPT_A<:Optim.AbstractOptimizer,
-    _turbo,
-    _bumper,
-    W,
-}
-    operators::OP
+struct Options{CT,_turbo,_bumper,W}
+    operators::AbstractOperatorEnum
     bin_constraints::Vector{Tuple{Int,Int}}
     una_constraints::Vector{Int}
     complexity_mapping::ComplexityMapping{CT}
@@ -94,10 +85,10 @@ struct Options{
     loss_function::Union{Nothing,Function}
     progress::Union{Bool,Nothing}
     terminal_width::Union{Int,Nothing}
-    optimizer_algorithm::OPT_A
+    optimizer_algorithm::Optim.AbstractOptimizer
     optimizer_probability::Float32
     optimizer_nrestarts::Int
-    optimizer_options::OPT
+    optimizer_options::Optim.Options
     recorder_file::String
     prob_pick_first::Float32
     early_stop_condition::Union{Function,Nothing}
@@ -108,6 +99,7 @@ struct Options{
     nested_constraints::Union{Vector{Tuple{Int,Int,Vector{Tuple{Int,Int,Int}}}},Nothing}
     deterministic::Bool
     define_helper_functions::Bool
+    use_recorder::Bool
 end
 
 function Base.print(io::IO, options::Options)

src/Recorder.jl

@@ -2,12 +2,10 @@ module RecorderModule
 
 using ..CoreModule: RecordType, Options
 
-is_recording(::Options{<:Any,<:Any,use_recorder}) where {use_recorder} = use_recorder
-
 "Assumes that `options` holds the user options::Options"
 macro recorder(ex)
     quote
-        if is_recording($(esc(:options)))
+        if $(esc(:options)).use_recorder
             $(esc(ex))
         end
     end

src/SymbolicRegression.jl

@@ -223,7 +223,7 @@ using .HallOfFameModule:
     HallOfFame, calculate_pareto_frontier, string_dominating_pareto_curve
 using .SingleIterationModule: s_r_cycle, optimize_and_simplify_population
 using .ProgressBarsModule: WrappedProgressBar
-using .RecorderModule: @recorder, is_recording, find_iteration_from_record
+using .RecorderModule: @recorder, find_iteration_from_record
 using .MigrationModule: migrate!
 using .SearchUtilsModule:
     DefaultWorkerOutputType,
@@ -354,7 +354,7 @@ function equation_search(
     y::AbstractMatrix{T};
     niterations::Int=10,
     weights::Union{AbstractMatrix{T},AbstractVector{T},Nothing}=nothing,
-    options::Options=Options(),
+    @nospecialize(options::Options = Options()),
     node_type::Type{N}=Node,
     variable_names::Union{AbstractVector{String},Nothing}=nothing,
     display_variable_names::Union{AbstractVector{String},Nothing}=variable_names,
@@ -446,7 +446,7 @@ function equation_search(
     datasets::Vector{D},
     ::Type{N}=Node;
     niterations::Int=10,
-    options::Options=Options(),
+    @nospecialize(options::Options = Options()),
     parallelism=:multithreading,
     numprocs::Union{Int,Nothing}=nothing,
     procs::Union{Vector{Int},Nothing}=nothing,
@@ -597,7 +597,7 @@ function _equation_search(
     ::Val{DIM_OUT},
     datasets::Vector{D},
     niterations::Int,
-    options::Options,
+    @nospecialize(options::Options),
     ::Type{N},
     numprocs::Int,
     procs::Union{Vector{Int},Nothing},
@@ -629,7 +629,9 @@ function _equation_search(
 
     example_dataset = datasets[1]
     nout = size(datasets, 1)
+    @assert nout >= 1
     @assert (nout == 1 || DIM_OUT == 2)
+    @assert options.populations >= 1
 
     if runtests
         test_option_configuration(PARALLELISM, datasets, saved_state, options, verbosity)
@@ -701,23 +703,23 @@ function _equation_search(
     # Get the next worker process to give a job:
     worker_assignment = initialize_worker_assignment()
 
-    hallOfFame = load_saved_hall_of_fame(saved_state)
-    hallOfFame = if hallOfFame === nothing
-        [HallOfFame(options, T, L, N) for j in 1:nout]
+    init_hall_of_fame = load_saved_hall_of_fame(saved_state)
+    hallOfFame = if init_hall_of_fame === nothing
+        HallOfFameType[HallOfFame(options, T, L, N) for j in 1:nout]
     else
         # Recompute losses for the hall of fame, in
         # case the dataset changed:
-        for (hof, dataset) in zip(hallOfFame, datasets)
+        for (hof, dataset) in zip(init_hall_of_fame, datasets)
             for member in hof.members[hof.exists]
                 score, result_loss = score_func(dataset, member, options)
                 member.score = score
                 member.loss = result_loss
             end
         end
-        hallOfFame
+        init_hall_of_fame
     end
-    @assert length(hallOfFame) == nout
     hallOfFame::Vector{HallOfFameType}
+    @assert length(hallOfFame) == nout
 
     for j in 1:nout, i in 1:(options.populations)
         worker_idx = assign_next_worker!(
@@ -944,7 +946,7 @@ function _equation_search(
             worker_idx = assign_next_worker!(
                 worker_assignment; out=j, pop=i, parallelism=PARALLELISM, procs
             )
-            iteration = if is_recording(options)
+            iteration = if options.use_recorder
                 key = "out$(j)_pop$(i)"
                 find_iteration_from_record(key, record) + 1
             else
@@ -1076,7 +1078,7 @@ end
 function _dispatch_s_r_cycle(
     in_pop::Population,
     dataset::Dataset,
-    options::Options;
+    @nospecialize(options::Options);
     pop::Int,
     out::Int,
     iteration::Int,

test/test_params.jl

@@ -54,7 +54,7 @@ default_params = (
     optimizer_nrestarts=3,
     optimizer_probability=0.1f0,
     optimizer_iterations=100,
-    val_recorder=Val(false),
+    use_recorder=false,
     recorder_file="pysr_recorder.json",
     tournament_selection_p=1.0,
     early_stop_condition=nothing,

test/test_recorder.jl

@@ -9,7 +9,7 @@ y = 3 * cos.(X[2, :]) + X[1, :] .^ 2 .- 2
 options = SymbolicRegression.Options(;
     binary_operators=(+, *, /, -),
     unary_operators=(cos,),
-    val_recorder=Val(true),
+    use_recorder=true,
     recorder_file="pysr_recorder.json",
     crossover_probability=0.0,  # required for recording, as not set up to track crossovers.
     populations=2,