protected CalibratorEstimatorBase(IHostEnvironment env,
TCalibratorTrainer calibratorTrainer,
IPredictor predictor = null,
string labelColumn = DefaultColumnNames.Label,
string featureColumn = DefaultColumnNames.Features,
string weightColumn = null)
{
Host = env;
Predictor = predictor;
CalibratorTrainer = calibratorTrainer;
ScoreColumn = TrainerUtils.MakeR4ScalarColumn(DefaultColumnNames.Score); // Do we fantom this being named anything else (renaming column)? Complete metadata?
LabelColumn = TrainerUtils.MakeBoolScalarLabel(labelColumn);
FeatureColumn = TrainerUtils.MakeR4VecFeature(featureColumn);
PredictedLabel = new SchemaShape.Column(DefaultColumnNames.PredictedLabel,
SchemaShape.Column.VectorKind.Scalar,
BoolType.Instance,
false,
new SchemaShape(MetadataUtils.GetTrainerOutputMetadata()));
if (weightColumn != null)
{
WeightColumn = TrainerUtils.MakeR4ScalarWeightColumn(weightColumn);
}
}
/// <summary>
/// Initializes a new instance of <see cref="OrdinaryLeastSquaresRegressionTrainer"/>
/// </summary>
internal OrdinaryLeastSquaresRegressionTrainer(IHostEnvironment env, Options options)
: base(Contracts.CheckRef(env, nameof(env)).Register(LoadNameValue), TrainerUtils.MakeR4VecFeature(options.FeatureColumnName),
TrainerUtils.MakeR4ScalarColumn(options.LabelColumnName), TrainerUtils.MakeR4ScalarWeightColumn(options.ExampleWeightColumnName))
{
Host.CheckValue(options, nameof(options));
Host.CheckUserArg(options.L2Weight >= 0, nameof(options.L2Weight), "L2 regularization term cannot be negative");
_l2Weight = options.L2Weight;
_perParameterSignificance = options.PerParameterSignificance;
}
/// <summary>
/// Initializes a new instance of <see cref="OlsLinearRegressionTrainer"/>
/// </summary>
internal OlsLinearRegressionTrainer(IHostEnvironment env, Arguments args)
: base(Contracts.CheckRef(env, nameof(env)).Register(LoadNameValue), TrainerUtils.MakeR4VecFeature(args.FeatureColumn),
TrainerUtils.MakeR4ScalarColumn(args.LabelColumn), TrainerUtils.MakeR4ScalarWeightColumn(args.WeightColumn, args.WeightColumn.IsExplicit))
{
Host.CheckValue(args, nameof(args));
Host.CheckUserArg(args.L2Weight >= 0, nameof(args.L2Weight), "L2 regularization term cannot be negative");
_l2Weight = args.L2Weight;
_perParameterSignificance = args.PerParameterSignificance;
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeBinaryClassificationTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumn">The name of the label column.</param>
/// <param name="featureColumn">The name of the feature column.</param>
/// <param name="weightColumn">The name for the column containing the initial weight.</param>
/// <param name="numIterations">The number of iterations to use in learning the features.</param>
/// <param name="learningRate">The learning rate. GAMs work best with a small learning rate.</param>
/// <param name="maxBins">The maximum number of bins to use to approximate features</param>
internal RegressionGamTrainer(IHostEnvironment env,
string labelColumn = DefaultColumnNames.Label,
string featureColumn = DefaultColumnNames.Features,
string weightColumn = null,
int numIterations = GamDefaults.NumIterations,
double learningRate = GamDefaults.LearningRates,
int maxBins = GamDefaults.MaxBins)
: base(env, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(labelColumn), featureColumn, weightColumn, numIterations, learningRate, maxBins)
{
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeBinaryClassificationTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumnName">The name of the label column.</param>
/// <param name="featureColumnName">The name of the feature column.</param>
/// <param name="rowGroupColumnName">The name for the column containing the example weight.</param>
/// <param name="numberOfIterations">The number of iterations to use in learning the features.</param>
/// <param name="learningRate">The learning rate. GAMs work best with a small learning rate.</param>
/// <param name="maximumBinCountPerFeature">The maximum number of bins to use to approximate features</param>
internal RegressionGamTrainer(IHostEnvironment env,
string labelColumnName = DefaultColumnNames.Label,
string featureColumnName = DefaultColumnNames.Features,
string rowGroupColumnName = null,
int numberOfIterations = GamDefaults.NumberOfIterations,
double learningRate = GamDefaults.LearningRate,
int maximumBinCountPerFeature = GamDefaults.MaximumBinCountPerFeature)
: base(env, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(labelColumnName), featureColumnName, rowGroupColumnName, numberOfIterations, learningRate, maximumBinCountPerFeature)
{
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeRegressionTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumn">The name of the label column.</param>
/// <param name="featureColumn">The name of the feature column.</param>
/// <param name="weightColumn">The name for the column containing the initial weight.</param>
/// <param name="learningRate">The learning rate.</param>
/// <param name="minDatapointsInLeaves">The minimal number of documents allowed in a leaf of a regression tree, out of the subsampled data.</param>
/// <param name="numLeaves">The max number of leaves in each regression tree.</param>
/// <param name="numTrees">Total number of decision trees to create in the ensemble.</param>
internal FastTreeRegressionTrainer(IHostEnvironment env,
string labelColumn = DefaultColumnNames.Label,
string featureColumn = DefaultColumnNames.Features,
string weightColumn = null,
int numLeaves = Defaults.NumLeaves,
int numTrees = Defaults.NumTrees,
int minDatapointsInLeaves = Defaults.MinDocumentsInLeaves,
double learningRate = Defaults.LearningRates)
: base(env, TrainerUtils.MakeR4ScalarColumn(labelColumn), featureColumn, weightColumn, null, numLeaves, numTrees, minDatapointsInLeaves, learningRate)
{
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeRegressionTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumnName">The name of the label column.</param>
/// <param name="featureColumnName">The name of the feature column.</param>
/// <param name="exampleWeightColumnName">The name for the column containing the example weight.</param>
/// <param name="learningRate">The learning rate.</param>
/// <param name="minimumExampleCountPerLeaf">The minimal number of examples allowed in a leaf of a regression tree, out of the subsampled data.</param>
/// <param name="numberOfLeaves">The max number of leaves in each regression tree.</param>
/// <param name="numberOfTrees">Total number of decision trees to create in the ensemble.</param>
internal FastTreeRegressionTrainer(IHostEnvironment env,
string labelColumnName = DefaultColumnNames.Label,
string featureColumnName = DefaultColumnNames.Features,
string exampleWeightColumnName = null,
int numberOfLeaves = Defaults.NumberOfLeaves,
int numberOfTrees = Defaults.NumberOfTrees,
int minimumExampleCountPerLeaf = Defaults.MinimumExampleCountPerLeaf,
double learningRate = Defaults.LearningRate)
: base(env, TrainerUtils.MakeR4ScalarColumn(labelColumnName), featureColumnName, exampleWeightColumnName, null, numberOfLeaves, numberOfTrees, minimumExampleCountPerLeaf, learningRate)
{
}
/// <summary>
/// Initializes a new instance of <see cref="LightGbmRegressorTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumn">The name of the label column.</param>
/// <param name="featureColumn">The name of the feature column.</param>
/// <param name="weights">The name for the column containing the initial weight.</param>
/// <param name="numLeaves">The number of leaves to use.</param>
/// <param name="numBoostRound">Number of iterations.</param>
/// <param name="minDataPerLeaf">The minimal number of documents allowed in a leaf of the tree, out of the subsampled data.</param>
/// <param name="learningRate">The learning rate.</param>
internal LightGbmRegressorTrainer(IHostEnvironment env,
string labelColumn = DefaultColumnNames.Label,
string featureColumn = DefaultColumnNames.Features,
string weights = null,
int?numLeaves = null,
int?minDataPerLeaf = null,
double?learningRate = null,
int numBoostRound = LightGBM.Options.Defaults.NumBoostRound)
: base(env, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(labelColumn), featureColumn, weights, null, numLeaves, minDataPerLeaf, learningRate, numBoostRound)
{
}
/// <summary>
/// Initializes a new instance of <see cref="LightGbmRegressionTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumnName">The name of the label column.</param>
/// <param name="featureColumnName">The name of the feature column.</param>
/// <param name="exampleWeightColumnName">The name of the example weight column (optional).</param>
/// <param name="numberOfLeaves">The number of leaves to use.</param>
/// <param name="minimumExampleCountPerLeaf">The minimal number of data points allowed in a leaf of the tree, out of the subsampled data.</param>
/// <param name="learningRate">The learning rate.</param>
/// <param name="numberOfIterations">Number of iterations.</param>
internal LightGbmRegressionTrainer(IHostEnvironment env,
string labelColumnName = DefaultColumnNames.Label,
string featureColumnName = DefaultColumnNames.Features,
string exampleWeightColumnName = null,
int?numberOfLeaves = null,
int?minimumExampleCountPerLeaf = null,
double?learningRate = null,
int numberOfIterations = Trainers.LightGbm.Options.Defaults.NumberOfIterations)
: base(env, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(labelColumnName), featureColumnName, exampleWeightColumnName, null, numberOfLeaves, minimumExampleCountPerLeaf, learningRate, numberOfIterations)
{
}
/// <summary>
/// Initializes a new instance of <see cref="LightGbmRegressorTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumn">The name of the label column.</param>
/// <param name="featureColumn">The name of the feature column.</param>
/// <param name="weights">The name for the column containing the initial weight.</param>
/// <param name="numLeaves">The number of leaves to use.</param>
/// <param name="numBoostRound">Number of iterations.</param>
/// <param name="minDataPerLeaf">The minimal number of documents allowed in a leaf of the tree, out of the subsampled data.</param>
/// <param name="learningRate">The learning rate.</param>
/// <param name="advancedSettings">A delegate to set more settings.
/// The settings here will override the ones provided in the direct signature,
/// if both are present and have different values.
/// The columns names, however need to be provided directly, not through the <paramref name="advancedSettings"/>.</param>
public LightGbmRegressorTrainer(IHostEnvironment env,
string labelColumn = DefaultColumnNames.Label,
string featureColumn = DefaultColumnNames.Features,
string weights = null,
int?numLeaves = null,
int?minDataPerLeaf = null,
double?learningRate = null,
int numBoostRound = LightGbmArguments.Defaults.NumBoostRound,
Action <LightGbmArguments> advancedSettings = null)
: base(env, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(labelColumn), featureColumn, weights, null, numLeaves, minDataPerLeaf, learningRate, numBoostRound, advancedSettings)
{
}
/// <summary>
/// Initializes a new instance of <see cref="LightGbmRankingTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumn">The name of the label column.</param>
/// <param name="featureColumn">The name of the feature column.</param>
/// <param name="groupId">The name of the column containing the group ID. </param>
/// <param name="weights">The name of the optional column containing the initial weights.</param>
/// <param name="numLeaves">The number of leaves to use.</param>
/// <param name="numBoostRound">Number of iterations.</param>
/// <param name="minDataPerLeaf">The minimal number of documents allowed in a leaf of the tree, out of the subsampled data.</param>
/// <param name="learningRate">The learning rate.</param>
internal LightGbmRankingTrainer(IHostEnvironment env,
string labelColumn = DefaultColumnNames.Label,
string featureColumn = DefaultColumnNames.Features,
string groupId = DefaultColumnNames.GroupId,
string weights = null,
int?numLeaves = null,
int?minDataPerLeaf = null,
double?learningRate = null,
int numBoostRound = LightGBM.Options.Defaults.NumBoostRound)
: base(env, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(labelColumn), featureColumn, weights, groupId, numLeaves, minDataPerLeaf, learningRate, numBoostRound)
{
Host.CheckNonEmpty(groupId, nameof(groupId));
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeTweedieTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumnName">The name of the label column.</param>
/// <param name="featureColumnName">The name of the feature column.</param>
/// <param name="exampleWeightColumnName">The name for the column containing the example weight.</param>
/// <param name="learningRate">The learning rate.</param>
/// <param name="minimumExampleCountPerLeaf">The minimal number of documents allowed in a leaf of a regression tree, out of the subsampled data.</param>
/// <param name="numberOfLeaves">The max number of leaves in each regression tree.</param>
/// <param name="numberOfTrees">Total number of decision trees to create in the ensemble.</param>
internal FastTreeTweedieTrainer(IHostEnvironment env,
string labelColumnName = DefaultColumnNames.Label,
string featureColumnName = DefaultColumnNames.Features,
string exampleWeightColumnName = null,
int numberOfLeaves = Defaults.NumberOfLeaves,
int numberOfTrees = Defaults.NumberOfTrees,
int minimumExampleCountPerLeaf = Defaults.MinimumExampleCountPerLeaf,
double learningRate = Defaults.LearningRate)
: base(env, TrainerUtils.MakeR4ScalarColumn(labelColumnName), featureColumnName, exampleWeightColumnName, null, numberOfLeaves, numberOfTrees, minimumExampleCountPerLeaf, learningRate)
{
Host.CheckNonEmpty(labelColumnName, nameof(labelColumnName));
Host.CheckNonEmpty(featureColumnName, nameof(featureColumnName));
Initialize();
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeTweedieTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumn">The name of the label column.</param>
/// <param name="featureColumn">The name of the feature column.</param>
/// <param name="weightColumn">The name for the column containing the initial weight.</param>
/// <param name="learningRate">The learning rate.</param>
/// <param name="minDatapointsInLeaves">The minimal number of documents allowed in a leaf of a regression tree, out of the subsampled data.</param>
/// <param name="numLeaves">The max number of leaves in each regression tree.</param>
/// <param name="numTrees">Total number of decision trees to create in the ensemble.</param>
internal FastTreeTweedieTrainer(IHostEnvironment env,
string labelColumn = DefaultColumnNames.Label,
string featureColumn = DefaultColumnNames.Features,
string weightColumn = null,
int numLeaves = Defaults.NumLeaves,
int numTrees = Defaults.NumTrees,
int minDatapointsInLeaves = Defaults.MinDocumentsInLeaves,
double learningRate = Defaults.LearningRates)
: base(env, TrainerUtils.MakeR4ScalarColumn(labelColumn), featureColumn, weightColumn, null, numLeaves, numTrees, minDatapointsInLeaves, learningRate)
{
Host.CheckNonEmpty(labelColumn, nameof(labelColumn));
Host.CheckNonEmpty(featureColumn, nameof(featureColumn));
Initialize();
}
/// <summary>
/// Initializes a new instance of <see cref="LightGbmRankingTrainer"/>
/// </summary>
/// <param name="env">The private instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="labelColumnName">The name of the label column.</param>
/// <param name="featureColumnName">The name of the feature column.</param>
/// <param name="rowGroupdColumnName">The name of the column containing the group ID. </param>
/// <param name="weightsColumnName">The name of the optional column containing the initial weights.</param>
/// <param name="numberOfLeaves">The number of leaves to use.</param>
/// <param name="learningRate">The learning rate.</param>
/// <param name="minimumExampleCountPerLeaf">The minimal number of data points allowed in a leaf of the tree, out of the subsampled data.</param>
/// <param name="numberOfIterations">The number of iterations to use.</param>
internal LightGbmRankingTrainer(IHostEnvironment env,
string labelColumnName = DefaultColumnNames.Label,
string featureColumnName = DefaultColumnNames.Features,
string rowGroupdColumnName = DefaultColumnNames.GroupId,
string weightsColumnName = null,
int?numberOfLeaves = null,
int?minimumExampleCountPerLeaf = null,
double?learningRate = null,
int numberOfIterations = LightGBM.Options.Defaults.NumberOfIterations)
: base(env, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(labelColumnName),
featureColumnName, weightsColumnName, rowGroupdColumnName, numberOfLeaves,
minimumExampleCountPerLeaf, learningRate, numberOfIterations)
{
Host.CheckNonEmpty(rowGroupdColumnName, nameof(rowGroupdColumnName));
}
private protected CalibratorEstimatorBase(IHostEnvironment env,
ICalibratorTrainer calibratorTrainer, string labelColumn, string scoreColumn, string weightColumn)
{
Host = env;
_calibratorTrainer = calibratorTrainer;
if (!string.IsNullOrWhiteSpace(labelColumn))
{
LabelColumn = TrainerUtils.MakeBoolScalarLabel(labelColumn);
}
else
{
env.CheckParam(!calibratorTrainer.NeedsTraining, nameof(labelColumn), "For trained calibrators, " + nameof(labelColumn) + " must be specified.");
}
ScoreColumn = TrainerUtils.MakeR4ScalarColumn(scoreColumn); // Do we fanthom this being named anything else (renaming column)? Complete metadata?
if (weightColumn != null)
{
WeightColumn = TrainerUtils.MakeR4ScalarWeightColumn(weightColumn);
}
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeTweedieTrainer"/> by using the legacy <see cref="Arguments"/> class.
/// </summary>
internal FastTreeTweedieTrainer(IHostEnvironment env, Arguments args)
: base(env, args, TrainerUtils.MakeR4ScalarColumn(args.LabelColumn))
{
Initialize();
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeTweedieTrainer"/> by using the <see cref="Options"/> class.
/// </summary>
/// <param name="env">The instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="options">Algorithm advanced settings.</param>
internal FastTreeTweedieTrainer(IHostEnvironment env, Options options)
: base(env, options, TrainerUtils.MakeR4ScalarColumn(options.LabelColumn))
{
Initialize();
}
internal LightGbmRankingTrainer(IHostEnvironment env, Options options)
: base(env, LoadNameValue, options, TrainerUtils.MakeR4ScalarColumn(options.LabelColumnName))
{
Contracts.CheckUserArg(options.Sigmoid > 0, nameof(Options.Sigmoid), "must be > 0.");
}
internal OnlineGradientDescentTrainer(IHostEnvironment env, Options options)
: base(options, env, UserNameValue, TrainerUtils.MakeR4ScalarColumn(options.LabelColumn))
{
LossFunction = options.LossFunction.CreateComponent(env);
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeRegressionTrainer"/> by using the legacy <see cref="Arguments"/> class.
/// </summary>
internal FastTreeRegressionTrainer(IHostEnvironment env, Arguments args)
: base(env, args, TrainerUtils.MakeR4ScalarColumn(args.LabelColumn))
{
}
/// <summary>
/// Initializes a new instance of <see cref="FastTreeRegressionTrainer"/> by using the <see cref="Options"/> class.
/// </summary>
/// <param name="env">The instance of <see cref="IHostEnvironment"/>.</param>
/// <param name="options">Algorithm advanced settings.</param>
internal FastTreeRegressionTrainer(IHostEnvironment env, Options options)
: base(env, options, TrainerUtils.MakeR4ScalarColumn(options.LabelColumnName))
{
}
internal RegressionGamTrainer(IHostEnvironment env, Arguments args)
: base(env, args, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(args.LabelColumn))
{
}
internal LightGbmRankingTrainer(IHostEnvironment env, Options options)
: base(env, LoadNameValue, options, TrainerUtils.MakeR4ScalarColumn(options.LabelColumnName))
{
}
internal RegressionGamTrainer(IHostEnvironment env, Options options)
: base(env, options, LoadNameValue, TrainerUtils.MakeR4ScalarColumn(options.LabelColumn))
{
}
internal LightGbmRegressorTrainer(IHostEnvironment env, LightGbmArguments args)
: base(env, LoadNameValue, args, TrainerUtils.MakeR4ScalarColumn(args.LabelColumn))
{
}