lower runtime requirements

Project.toml

@@ -15,7 +15,9 @@ JSON = "682c06a0-de6a-54ab-a142-c8b1cf79cde6"
 LineSearches = "d3d80556-e9d4-5f37-9878-2ab0fcc64255"
 Logging = "56ddb016-857b-54e1-b83d-db4d58db5568"
 LossFunctions = "30fc2ffe-d236-52d8-8643-a9d8f7c094a7"
+MLJBase = "a7f614a8-145f-11e9-1d2a-a57a1082229d"
 MLJModelInterface = "e80e1ace-859a-464e-9ed9-23947d8ae3ea"
+MLJTestInterface = "72560011-54dd-4dc2-94f3-c5de45b75ecd"
 MacroTools = "1914dd2f-81c6-5fcd-8719-6d5c9610ff09"
 Optim = "429524aa-4258-5aef-a3af-852621145aeb"
 PackageExtensionCompat = "65ce6f38-6b18-4e1d-a461-8949797d7930"
@@ -27,6 +29,7 @@ Reexport = "189a3867-3050-52da-a836-e630ba90ab69"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
 SymbolicRegression = "8254be44-1295-4e6a-a16d-46603ac705cb"
 TOML = "fa267f1f-6049-4f14-aa54-33bafae1ed76"
+Zygote = "e88e6eb3-aa80-5325-afca-941959d7151f"
 
 [compat]
 ADTypes = "^1.4.0"
@@ -40,7 +43,9 @@ JSON = "0.21"
 LineSearches = "7"
 Logging = "1"
 LossFunctions = "0.10, 0.11, 1"
+MLJBase = "1.7.0"
 MLJModelInterface = "~1.5, ~1.6, ~1.7, ~1.8, ~1.9, ~1.10, ~1.11"
+MLJTestInterface = "0.2.8"
 MacroTools = "0.4, 0.5"
 Optim = "~1.8, ~1.9, ~1.10"
 PackageExtensionCompat = "1.0.2"
@@ -52,4 +57,5 @@ Reexport = "1"
 StatsBase = "0.33, 0.34"
 SymbolicRegression = ">=1.6.0"
 TOML = "<0.0.1, 1"
+Zygote = "0.6.75"
 julia = "1.10"

examples/parameterized_function.jl

@@ -6,7 +6,7 @@ using Test
 
 rng = MersenneTwister(0)
 X = NamedTuple{(:x1, :x2, :x3, :x4, :x5)}(ntuple(_ -> randn(rng, Float32, 30), Val(5)))
-X = (; X..., classes=rand(rng, 1:2, 30))
+X = (; X..., class=rand(rng, 1:2, 30))
 p1 = [0.0f0, 3.2f0]
 p2 = [1.5f0, 0.5f0]
 

test/test_params.jl

@@ -23,7 +23,7 @@ const default_params = (
     topn=10,
     parsimony=0.000100f0,
     alpha=0.100000f0,
-    maxsize=20,
+    maxsize=10,
     maxdepth=nothing,
     fast_cycle=false,
     migration=true,
@@ -47,8 +47,8 @@ const default_params = (
     crossover_probability=0.0f0,
     warmup_maxsize_by=0.0f0,
     use_frequency=false,
-    population_size=1000,
-    ncycles_per_iteration=300,
+    population_size=500,
+    ncycles_per_iteration=150,
     fraction_replaced=0.1f0,
     verbosity=convert(Int, 1e9),
     probability_negate_constant=0.01f0,
@@ -60,7 +60,7 @@ const default_params = (
     optimizer_algorithm=Optim.NelderMead(; linesearch=LineSearches.BackTracking()),
     optimizer_nrestarts=3,
     optimizer_probability=0.1f0,
-    optimizer_iterations=100,
+    optimizer_iterations=50,
     use_recorder=false,
     recorder_file="pysr_recorder.json",
     tournament_selection_p=1.0,