GitHub

LGBMClassifier

LGBMClassifier

A model type for constructing a LightGBM classifier, based on LightGBM.jl, and implementing the MLJ model interface.

From MLJ, the type can be imported using

LGBMClassifier = @load LGBMClassifier pkg=LightGBM

Do model = LGBMClassifier() to construct an instance with default hyper-parameters. Provide keyword arguments to override hyper-parameter defaults, as in LGBMClassifier(objective=...).

`LightGBM, short for light gradient-boosting machine, is a framework for gradient boosting based on decision tree algorithms and used for classification and other machine learning tasks, with a focus on performance and scalability. This model in particular is used for various types of classification tasks.

Training data In MLJ or MLJBase, bind an instance model to data with

mach = machine(model, X, y)

Here:

  • X is any table of input features (eg, a DataFrame) whose columns are of scitype Continuous; check the column scitypes with schema(X); alternatively, X is any AbstractMatrix with Continuous elements; check the scitype with scitype(X).
  • y is a vector of targets whose items are of scitype Continuous. Check the scitype with scitype(y).

Train the machine using fit!(mach, rows=...).

Operations

  • predict(mach, Xnew): return predictions of the target given new features Xnew, which should have the same scitype as X above.

Hyper-parameters

See https://lightgbm.readthedocs.io/en/v3.3.5/Parameters.html.

Fitted parameters

The fields of fitted_params(mach) are:

  • fitresult: Fitted model information, contains a LGBMClassification object, a CategoricalArray of the input class names, and the classifier with all its parameters

Report

The fields of report(mach) are:

  • training_metrics: A dictionary containing all training metrics.

  • importance: A namedtuple containing:

    • gain: The total gain of each split used by the model
    • split: The number of times each feature is used by the model.

Examples


using DataFrames
using MLJ

## load the model
LGBMClassifier = @load LGBMClassifier pkg=LightGBM 

X, y = @load_iris 
X = DataFrame(X)
train, test = partition(collect(eachindex(y)), 0.70, shuffle=true)

first(X, 3)
lgb = LGBMClassifier() ## initialise a model with default params
mach = machine(lgb, X[train, :], y[train]) |> fit!

predict(mach, X[test, :])

## access feature importances
model_report = report(mach)
gain_importance = model_report.importance.gain
split_importance = model_report.importance.split

See also LightGBM.jl and the unwrapped model type LightGBM.LGBMClassification