search

JuliaAI / MLJTuning.jl

Hyperparameter optimization algorithms for use in the MLJ machine learning framework
MIT License
67 stars 12 forks source link

`learning_curve` throwing nested task error #155

Open ablaom opened 2 years ago

ablaom commented 2 years ago 
X, y = make_blobs()

model = (@load RandomForestClassifier pkg=DecisionTree)()
mach = machine(model, X, y)

r = range(model, :n_trees, lower=10, upper=70, scale=:log10)
many_curves = learning_curve(mach,
                             range=r,
                             resampling=Holdout(),
                             measure=cross_entropy,
                             rng_name=:rng,
                             rngs=1)

Evaluating Learning curve with 1 rngs:   0%[>                 ]  ETA: N/A┌ Error: Problem fi
tting the machine Machine{RandomForestClassifier,…}.                              
└ @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:533
[ Info: Running type checks... 
[ Info: Type checks okay. 
┌ Error: Problem fitting the machine Machine{Resampler{Holdout},…}. 
└ @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:533
[ Info: Running type checks... 
[ Info: Type checks okay. 
┌ Error: Problem fitting the machine Machine{ProbabilisticTunedModel{Grid,…},…}. 
└ @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:533
[ Info: Running type checks... 
[ Info: Type checks okay. 
ERROR: TaskFailedException
Stacktrace:
  [1] wait
    @ ./task.jl:322 [inlined]
  [2] threading_run(func::Function)
    @ Base.Threads ./threadingconstructs.jl:34
  [3] macro expansion
    @ ./threadingconstructs.jl:93 [inlined]
  [4] build_forest(labels::Vector{UInt32}, features::Matrix{Float64}, n_subfeatures::Int64, 
n_trees::Int64, partial_sampling::Float64, max_depth::Int64, min_samples_leaf::Int64, min_samples_split::Int64, min_purity_increase::Float64; rng::Random.MersenneTwister)        
    @ DecisionTree ~/.julia/packages/DecisionTree/iWCbW/src/classification/main.jl:223
  [5] fit(m::MLJDecisionTreeInterface.RandomForestClassifier, verbosity::Int64, X::DataFrames.DataFrame, y::CategoricalVector{Int64, UInt32, Int64, CategoricalValue{Int64, UInt32}, Union{}})                           
    @ MLJDecisionTreeInterface ~/.julia/packages/MLJDecisionTreeInterface/RZmUr/src/MLJDecisionTreeInterface.jl:200                                                             
  [6] fit_only!(mach::Machine{MLJDecisionTreeInterface.RandomForestClassifier, true}; rows::
Vector{Int64}, verbosity::Int64, force::Bool)                                              
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:531
  [7] #fit!#103
    @ ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:598 [inlined]
  [8] fit_and_extract_on_fold
    @ ~/.julia/packages/MLJBase/HZmTU/src/resampling.jl:1088 [inlined]
  [9] (::MLJBase.var"#276#277"{MLJBase.var"#fit_and_extract_on_fold#299"{Vector{Tuple{Vector{Int64}, Vector{Int64}}}, Nothing, Nothing, Int64, Vector{LogLoss{Float64}}, Vector{typeof(predict)}, Bool, Bool, CategoricalVector{Int64, UInt32, Int64, CategoricalValue{Int64, UInt32}, Union{}}, DataFrames.DataFrame}, Machine{MLJDecisionTreeInterface.RandomForestClassifier, true}, Int64, ProgressMeter.Progress})(k::Int64)
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/resampling.jl:932
 [10] mapreduce_first
    @ ./reduce.jl:392 [inlined]
 [11] _mapreduce(f::MLJBase.var"#276#277"{MLJBase.var"#fit_and_extract_on_fold#299"{Vector{Tuple{Vector{Int64}, Vector{Int64}}}, Nothing, Nothing, Int64, Vector{LogLoss{Float64}}, Vector{typeof(predict)}, Bool, Bool, CategoricalVector{Int64, UInt32, Int64, CategoricalValue{Int64, UInt32}, Union{}}, DataFrames.DataFrame}, Machine{MLJDecisionTreeInterface.RandomForestClassifier, true}, Int64, ProgressMeter.Progress}, op::typeof(vcat), #unused#::IndexLinear, 
A::UnitRange{Int64})                                                                       
    @ Base ./reduce.jl:403
 [12] _mapreduce_dim
    @ ./reducedim.jl:318 [inlined]
 [13] #mapreduce#672
    @ ./reducedim.jl:310 [inlined]
 [14] mapreduce
    @ ./reducedim.jl:310 [inlined]
 [15] _evaluate!(func::MLJBase.var"#fit_and_extract_on_fold#299"{Vector{Tuple{Vector{Int64}, Vector{Int64}}}, Nothing, Nothing, Int64, Vector{LogLoss{Float64}}, Vector{typeof(predict)}, Bool, Bool, CategoricalVector{Int64, UInt32, Int64, CategoricalValue{Int64, UInt32}, Union{}}, DataFrames.DataFrame}, mach::Machine{MLJDecisionTreeInterface.RandomForestClassifier, true}, #unused#::CPU1{Nothing}, nfolds::Int64, verbosity::Int64)
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/resampling.jl:931
 [16] evaluate!(mach::Machine{MLJDecisionTreeInterface.RandomForestClassifier, true}, resampling::Vector{Tuple{Vector{Int64}, Vector{Int64}}}, weights::Nothing, class_weights::Nothing, rows::Nothing, verbosity::Int64, repeats::Int64, measures::Vector{LogLoss{Float64}}, operations::Vector{typeof(predict)}, acceleration::CPU1{Nothing}, force::Bool)              
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/resampling.jl:1126
 [17] evaluate!(::Machine{MLJDecisionTreeInterface.RandomForestClassifier, true}, ::Holdout, ::Nothing, ::Nothing, ::Nothing, ::Int64, ::Int64, ::Vector{LogLoss{Float64}}, ::Vector{typeof(predict)}, ::CPU1{Nothing}, ::Bool)                                                 
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/resampling.jl:1193
 [18] fit(::Resampler{Holdout}, ::Int64, ::DataFrames.DataFrame, ::CategoricalVector{Int64, UInt32, Int64, CategoricalValue{Int64, UInt32}, Union{}})                           
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/resampling.jl:1337
 [19] fit_only!(mach::Machine{Resampler{Holdout}, false}; rows::Nothing, verbosity::Int64, force::Bool)                                                                                
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:531
 [20] #fit!#103
    @ ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:598 [inlined]
 [21] event!(metamodel::MLJDecisionTreeInterface.RandomForestClassifier, resampling_machine::Machine{Resampler{Holdout}, false}, verbosity::Int64, tuning::Grid, history::Nothing, state
::NamedTuple{(:models, :fields, :parameter_scales, :models_delivered), Tuple{Vector{MLJDecisionTreeInterface.RandomForestClassifier}, Vector{Symbol}, Vector{Symbol}, Bool}})
    @ MLJTuning ~/.julia/packages/MLJTuning/efiDR/src/tuned_models.jl:395
 [22] #35
    @ ~/.julia/packages/MLJTuning/efiDR/src/tuned_models.jl:433 [inlined]
 [23] iterate
    @ ./generator.jl:47 [inlined]
 [24] _collect(c::Vector{MLJDecisionTreeInterface.RandomForestClassifier}, itr::Base.Generator{Vector{MLJDecisionTreeInterface.RandomForestClassifier}, MLJTuning.var"#35#36"{Machine{Resampler{Holdout}, false}, Int64, Grid, Nothing, NamedTuple{(:models, :fields, :parameter_scales, :models_delivered), Tuple{Vector{MLJDecisionTreeInterface.RandomForestClassifier}, Vector{Symbol}, Vector{Symbol}, Bool}}, ProgressMeter.Progress}}, #unused#::Base.EltypeUnknown, 
isz::Base.HasShape{1})                                                                     
    @ Base ./array.jl:695
 [25] collect_similar
    @ ./array.jl:606 [inlined]
 [26] map
    @ ./abstractarray.jl:2294 [inlined]
 [27] assemble_events!(metamodels::Vector{MLJDecisionTreeInterface.RandomForestClassifier}, 
resampling_machine::Machine{Resampler{Holdout}, false}, verbosity::Int64, tuning::Grid, history::Nothing, state::NamedTuple{(:models, :fields, :parameter_scales, :models_delivered), Tuple{Vector{MLJDecisionTreeInterface.RandomForestClassifier}, Vector{Symbol}, Vector{Symbol}, Bool}}, acceleration::CPU1{Nothing})
    @ MLJTuning ~/.julia/packages/MLJTuning/efiDR/src/tuned_models.jl:432
 [28] build!(history::Nothing, n::Int64, tuning::Grid, model::MLJDecisionTreeInterface.RandomForestClassifier, model_buffer::Channel{Any}, state::NamedTuple{(:models, :fields, :parameter_scales, :models_delivered), Tuple{Vector{MLJDecisionTreeInterface.RandomForestClassifier}, Vector{Symbol}, Vector{Symbol}, Bool}}, verbosity::Int64, acceleration::CPU1{Nothing}, resampling_machine::Machine{Resampler{Holdout}, false})                                     
    @ MLJTuning ~/.julia/packages/MLJTuning/efiDR/src/tuned_models.jl:625
 [29] fit(::MLJTuning.ProbabilisticTunedModel{Grid, MLJDecisionTreeInterface.RandomForestClassifier}, ::Int64, ::DataFrames.DataFrame, ::CategoricalVector{Int64, UInt32, Int64, CategoricalValue{Int64, UInt32}, Union{}})                           
    @ MLJTuning ~/.julia/packages/MLJTuning/efiDR/src/tuned_models.jl:704
 [30] fit_only!(mach::Machine{MLJTuning.ProbabilisticTunedModel{Grid, MLJDecisionTreeInterface.RandomForestClassifier}, true}; rows::Nothing, verbosity::Int64, force::Bool)
    @ MLJBase ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:531
 [31] #fit!#103
    @ ~/.julia/packages/MLJBase/HZmTU/src/machines.jl:598 [inlined]
 [32] (::MLJTuning.var"#61#62"{Machine{MLJTuning.ProbabilisticTunedModel{Grid, MLJDecisionTreeInterface.RandomForestClassifier}, true}, Nothing, Symbol, Int64, ProgressMeter.Progress})
(rng::Random.MersenneTwister)                                                              
    @ MLJTuning ~/.julia/packages/MLJTuning/efiDR/src/learning_curves.jl:231
 [33] mapreduce_first
    @ ./reduce.jl:392 [inlined]
 [34] _mapreduce(f::MLJTuning.var"#61#62"{Machine{MLJTuning.ProbabilisticTunedModel{Grid, MLJDecisionTreeInterface.RandomForestClassifier}, true}, Nothing, Symbol, Int64, ProgressMeter.Progress}, op::typeof(MLJTuning._collate), #unused#::IndexLinear, A::Vector{Random.MersenneTwister})                                                                   
    @ Base ./reduce.jl:403
 [35] _mapreduce_dim
    @ ./reducedim.jl:318 [inlined]
 [36] #mapreduce#672
    @ ./reducedim.jl:310 [inlined]
 [37] mapreduce
    @ ./reducedim.jl:310 [inlined]
 [38] _tuning_results(rngs::Vector{Random.MersenneTwister}, acceleration::CPU1{Nothing}, tuned::Machine{MLJTuning.ProbabilisticTunedModel{Grid, MLJDecisionTreeInterface.RandomForestClassifier}, true}, rows::Nothing, rng_name::Symbol, verbosity::Int64)
    @ MLJTuning ~/.julia/packages/MLJTuning/efiDR/src/learning_curves.jl:229
 [39] learning_curve(::MLJDecisionTreeInterface.RandomForestClassifier, ::MLJBase.Source, ::
Vararg{MLJBase.Source, N} where N; resolution::Int64, resampling::Holdout, weights::Nothing, measures::Nothing, measure::LogLoss{Float64}, rows::Nothing, operation::Nothing, ranges::Nothing, range::MLJBase.NumericRange{Int64, MLJBase.Bounded, Symbol}, repeats::Int64, acceleration::CPU1{Nothing}, acceleration_grid::CPU1{Nothing}, verbosity::Int64, rngs::Int64, rng_name::Symbol, check_measure::Bool)                                                      
    @ MLJTuning ~/.julia/packages/MLJTuning/efiDR/src/learning_curves.jl:173
 [40] #learning_curve#58
    @ ~/.julia/packages/MLJTuning/efiDR/src/learning_curves.jl:92 [inlined]
 [41] top-level scope
    @ REPL[44]:1

    nested task error: AssertionError: length(ints) == 501
    Stacktrace:
      [1] mt_setfull!(r::Random.MersenneTwister, #unused#::Type{UInt64})
        @ Random /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/RNGs.jl:260
      [2] reserve1
        @ /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/RNGs.jl:291 [inlined]
      [3] mt_pop!
        @ /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/RNGs.jl:296 [inlined]
      [4] rand
        @ /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/RNGs.jl:464 [inlined]
      [5] rand
        @ /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/Random.jl:256 [inlined]
      [6] rand(rng::Random.MersenneTwister, sp::Random.SamplerRangeNDL{UInt64, Int64})
        @ Random /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/generation.jl:332
      [7] rand!
        @ /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/Random.jl:271 [inlined]                              
      [8] rand!(rng::Random.MersenneTwister, A::Vector{Int64}, X::UnitRange{Int64})
        @ Random /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/Random.jl:266
      [9] rand
        @ /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/Random.jl:279 [inlined]
     [10] rand
        @ /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.6/Random/src/Random.jl:282 [inlined]
     [11] macro expansion
        @ ~/.julia/packages/DecisionTree/iWCbW/src/classification/main.jl:224 [inlined]
     [12] (::DecisionTree.var"#62#threadsfor_fun#22"{Random.MersenneTwister, Vector{UInt32}, Matrix{Float64}, Int64, Int64, Int64, Float64, DecisionTree.var"#20#21"{Vector{Float64}}, Vector{Union{DecisionTree.Leaf{UInt32}, DecisionTree.Node{Float64, UInt32}}}, Int64, Int64, UnitRange{Int64}})(onethread::Bool)
        @ DecisionTree ./threadingconstructs.jl:81
     [13] (::DecisionTree.var"#62#threadsfor_fun#22"{Random.MersenneTwister, Vector{UInt32}, Matrix{Float64}, Int64, Int64, Int64, Float64, DecisionTree.var"#20#21"{Vector{Float64}}, Vector{Union{DecisionTree.Leaf{UInt32}, DecisionTree.Node{Float64, UInt32}}}, Int64, Int64, UnitRange{Int64}})()
        @ DecisionTree ./threadingconstructs.jl:48
(MachineLearningInJulia2020) pkg> status
      Status `~/Google Drive/Julia/MLJ/MachineLearningInJulia2020/Project.toml`
  [336ed68f] CSV v0.9.6
  [324d7699] CategoricalArrays v0.10.1
  [ed09eef8] ComputationalResources v0.3.2
  [a93c6f00] DataFrames v1.2.2
  [7806a523] DecisionTree v0.10.11
  [31c24e10] Distributions v0.25.18
  [f6006082] EvoTrees v0.8.4
  [98b081ad] Literate v2.9.3
  [add582a8] MLJ v0.16.9
  [a7f614a8] MLJBase v0.18.23
  [d354fa79] MLJClusteringInterface v0.1.4
  [094fc8d1] MLJFlux v0.2.5
  [6ee0df7b] MLJLinearModels v0.5.6
  [d491faf4] MLJModels v0.14.12
  [1b6a4a23] MLJMultivariateStatsInterface v0.2.2
  [5ae90465] MLJScikitLearnInterface v0.1.10
  [b8a86587] NearestNeighbors v0.4.9
  [a03496cd] PlotlyBase v0.8.18
  [91a5bcdd] Plots v1.22.4
  [321657f4] ScientificTypes v2.3.0
  [2913bbd2] StatsBase v0.33.10
  [bd369af6] Tables v1.6.0
  [b8865327] UnicodePlots v2.4.6
  [9a3f8284] Random

Julia 1.6.3