LightGbm(this SweepableMultiClassificationTrainers trainer, string labelColumnName = "Label", string featureColumnName = "Features", SweepableOption <LightGbmMulticlassTrainer.Options> optionBuilder = null, LightGbmMulticlassTrainer.Options defaultOption = null)
{
var context = trainer.Context;
if (optionBuilder == null)
{
optionBuilder = LightGbmMulticlassTrainerSweepableOptions.Default;
}
optionBuilder.SetDefaultOption(defaultOption);
return(context.AutoML().CreateSweepableEstimator(
(context, option) =>
{
if (defaultOption != null)
{
Util.CopyFieldsTo(defaultOption, option);
}
option.LabelColumnName = labelColumnName;
option.FeatureColumnName = featureColumnName;
return context.MulticlassClassification.Trainers.LightGbm(option);
},
optionBuilder,
trainerName: nameof(LightGbmMulticlassTrainer),
inputs: new string[] { featureColumnName },
outputs: new string[] { PredictedLabel }));
}
NaiveBayes(this SweepableMultiClassificationTrainers trainer, string labelColumnName = "Label", string featureColumnName = "Features")
{
var context = trainer.Context;
var instance = context.MulticlassClassification.Trainers.NaiveBayes(labelColumnName, featureColumnName);
return(context.AutoML().CreateUnsweepableEstimator(
instance,
estimatorName: nameof(NaiveBayesMulticlassTrainer),
inputs: new string[] { featureColumnName },
outputs: new string[] { PredictedLabel }));
}
OneVersusAll <TModel, TOption>(this SweepableMultiClassificationTrainers trainer, ISweepableEstimator <ITrainerEstimator <BinaryPredictionTransformer <TModel>, TModel>, TOption> node, string labelColumnName = "Label", bool imputeMissingLabelsAsNegative = false, Microsoft.ML.IEstimator <Microsoft.ML.ISingleFeaturePredictionTransformer <Microsoft.ML.Calibrators.ICalibrator> > calibrator = default, int maximumCalibrationExampleCount = 1000000000, bool useProbabilities = true)
where TModel : class
where TOption : class
{
var context = trainer.Context;
return(context.AutoML().CreateSweepableEstimator(
(context, option) =>
{
var estimator = node.EstimatorFactory(option);
return context.MulticlassClassification.Trainers.OneVersusAll(estimator, labelColumnName, imputeMissingLabelsAsNegative, calibrator, maximumCalibrationExampleCount, useProbabilities);
},
node.OptionBuilder,
inputs: node.InputColumns,
outputs: node.OutputColumns,
trainerName: node.EstimatorName + "Ova"));
}
SdcaNonCalibreated(this SweepableMultiClassificationTrainers trainer, string labelColumnName = "Label", string featureColumnName = "Features", SweepableOption <SdcaNonCalibratedMulticlassTrainer.Options> optionBuilder = null, SdcaNonCalibratedMulticlassTrainer.Options defaultOption = null)
{
var context = trainer.Context;
if (optionBuilder == null)
{
optionBuilder = SdcaNonCalibratedMulticlassTrainerSweepableOptions.Default;
}
optionBuilder.SetDefaultOption(defaultOption);
return(context.AutoML().CreateSweepableEstimator(
(context, option) =>
{
option.LabelColumnName = labelColumnName;
option.FeatureColumnName = featureColumnName;
return context.MulticlassClassification.Trainers.SdcaNonCalibrated(option);
},
optionBuilder,
trainerName: nameof(SdcaNonCalibratedMulticlassTrainer),
inputs: new string[] { featureColumnName },
outputs: new string[] { PredictedLabel }));
}
