Hi All,
I am trying to blindly use JLBoost on a big dataset with 2381 features and 600 000 samples. I have automatically converted the feature matrix to the DataFrame using df = DataFrame(x, :auto)
, add labels as df.y = y
, and then I started to fit the model with default settings
julia> xgtreemodel = jlboost(df, :y)
this branch has too few records
this branch has too few records
ERROR: KeyError: key not found
Stacktrace:
[1] getindex(::Dict{Any,Any}, ::JLBoostTree) at ./dict.jl:477
[2] _fit_tree!(::LogitLogLoss, ::DataFrame, ::Symbol, ::Array{Symbol,1}, ::Array{Float64,1}, ::JLBoostTree, ::typeof(depth_wise), ::JLBoost.TreeFitStoppingCriterion.var"#1#2"{Int64}; lambda::Int64, gamma::Int64, verbose::Bool, kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /home/tomas.pevny/.julia/packages/JLBoost/hbacI/src/fit_tree.jl:156
[3] jlboost(::DataFrame, ::Symbol, ::Array{Symbol,1}, ::Array{Float64,1}, ::LogitLogLoss, ::JLBoost.var"#17#21"{JLBoost.var"#17#18#22"}, ::JLBoost.var"#9#13"{JLBoost.var"#9#10#14"}, ::typeof(depth_wise), ::JLBoost.TreeFitStoppingCriterion.var"#1#2"{Int64}; nrounds::Int64, eta::Float64, verbose::Bool, kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /home/tomas.pevny/.julia/packages/JLBoost/hbacI/src/jlboost-fit.jl:130
[4] jlboost(::DataFrame, ::Symbol, ::Array{Symbol,1}, ::Array{Float64,1}, ::LogitLogLoss, ::Function, ::Function, ::Function, ::JLBoost.TreeFitStoppingCriterion.var"#1#2"{Int64}) at /home/tomas.pevny/.julia/packages/JLBoost/hbacI/src/jlboost-fit.jl:102
[5] jlboost(::DataFrame, ::Symbol, ::Array{Symbol,1}, ::Array{Float64,1}, ::LogitLogLoss; subsample::Int64, colsample_bytree::Int64, max_depth::Int64, max_leaves::Int64, kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /home/tomas.pevny/.julia/packages/JLBoost/hbacI/src/jlboost-fit.jl:84
[6] jlboost at /home/tomas.pevny/.julia/packages/JLBoost/hbacI/src/jlboost-fit.jl:58 [inlined] (repeats 2 times)
[7] jlboost(::DataFrame, ::Symbol, ::Array{String,1}; kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /home/tomas.pevny/.julia/packages/JLBoost/hbacI/src/jlboost-fit.jl:51
[8] jlboost(::DataFrame, ::Symbol, ::Array{String,1}) at /home/tomas.pevny/.julia/packages/JLBoost/hbacI/src/jlboost-fit.jl:49
[9] top-level scope at REPL[114]:1
Can anyone give me hint, what might went wrong?
Thanks a lot,
Tomas