/// <summary>
/// Classification model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy on how to combine the models</param>
/// <param name="ensembleSelection">Ensemble selection method used to find the beset subset of models</param>
public ClassificationModelSelectingEnsembleLearner(
IIndexedLearner <ProbabilityPrediction>[] learners,
ICrossValidation <ProbabilityPrediction> crossValidation,
IClassificationEnsembleStrategy ensembleStrategy,
IClassificationEnsembleSelection ensembleSelection)
: this(learners, crossValidation, () => ensembleStrategy, ensembleSelection)
{
}
/// <summary>
/// Regression model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy on how to combine the models</param>
/// <param name="ensembleSelection">Ensemble selection method used to find the beset subset of models</param>
public RegressionModelSelectingEnsembleLearner(
IIndexedLearner <double>[] learners,
ICrossValidation <double> crossValidation,
IRegressionEnsembleStrategy ensembleStrategy,
IRegressionEnsembleSelection ensembleSelection)
: this(learners, crossValidation, () => ensembleStrategy, ensembleSelection)
{
}
/// <summary>
/// Stacking Classification Ensemble Learner.
/// Combines several models into a single ensemble model using a top or meta level model to combine the models.
/// The bottom level models generates output for the top level model using cross validation.
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="metaLearner">Meta learner or top level model for combining the ensemble models</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="includeOriginalFeaturesForMetaLearner">True; the meta learner also receives the original features.
/// False; the meta learner only receives the output of the ensemble models as features. Default is true</param>
public ClassificationStackingEnsembleLearner(
IIndexedLearner <ProbabilityPrediction>[] learners,
ILearner <ProbabilityPrediction> metaLearner,
ICrossValidation <ProbabilityPrediction> crossValidation,
bool includeOriginalFeaturesForMetaLearner = true)
: this(learners, (obs, targets) => metaLearner.Learn(obs, targets),
crossValidation, includeOriginalFeaturesForMetaLearner)
{
}
/// <summary>
/// Stacking Regression Ensemble Learner.
/// Combines several models into a single ensemble model using a top or meta level model to combine the models.
/// The bottom level models generates output for the top level model using cross validation.
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="metaLearner">Meta learner or top level model for combining the ensemble models</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="includeOriginalFeaturesForMetaLearner">True; the meta learner also receives the original features.
/// False; the meta learner only receives the output of the ensemble models as features. Default is true</param>
public RegressionStackingEnsembleLearner(
IIndexedLearner <double>[] learners,
ILearner <double> metaLearner,
ICrossValidation <double> crossValidation,
bool includeOriginalFeaturesForMetaLearner = true)
: this(learners, (obs, targets) => metaLearner.Learn(obs, targets),
crossValidation, includeOriginalFeaturesForMetaLearner)
{
}
/// <summary>
/// Classification model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble using greedy backward elimination.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="numberOfModelsToSelect">Number of models to select</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy for ensembling models</param>
/// <param name="metric">Metric to minimize</param>
public ClassificationBackwardEliminationModelSelectingEnsembleLearner(
IIndexedLearner <ProbabilityPrediction>[] learners,
int numberOfModelsToSelect,
ICrossValidation <ProbabilityPrediction> crossValidation,
IClassificationEnsembleStrategy ensembleStrategy,
IMetric <double, ProbabilityPrediction> metric)
: base(learners, crossValidation, ensembleStrategy,
new BackwardEliminationClassificationEnsembleSelection(
metric, ensembleStrategy, numberOfModelsToSelect))
{
}
/// <summary>
/// Stacking Classification Ensemble Learner.
/// Combines several models into a single ensemble model using a top or meta level model to combine the models.
/// Combines several models into a single ensemble model using a top or meta level model to combine the models.
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="metaLearner">Meta learner or top level model for combining the ensemble models</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="includeOriginalFeaturesForMetaLearner">True; the meta learner also receives the original features.
/// False; the meta learner only receives the output of the ensemble models as features</param>
public ClassificationStackingEnsembleLearner(
IIndexedLearner <ProbabilityPrediction>[] learners,
Func <F64Matrix, double[], IPredictorModel <ProbabilityPrediction> > metaLearner,
ICrossValidation <ProbabilityPrediction> crossValidation,
bool includeOriginalFeaturesForMetaLearner = true)
{
m_learners = learners ?? throw new ArgumentException(nameof(learners));
m_crossValidation = crossValidation ?? throw new ArgumentException(nameof(crossValidation));
m_metaLearner = metaLearner ?? throw new ArgumentException(nameof(metaLearner));
m_includeOriginalFeaturesForMetaLearner = includeOriginalFeaturesForMetaLearner;
}
/// <summary>
/// Regression model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble using backwards elimination.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="numberOfModelsToSelect">Number of models to select</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy for ensembling models</param>
/// <param name="metric">Metric to minimize</param>
public RegressionBackwardEliminationModelSelectingEnsembleLearner(
IIndexedLearner <double>[] learners,
int numberOfModelsToSelect,
ICrossValidation <double> crossValidation,
IRegressionEnsembleStrategy ensembleStrategy,
IMetric <double, double> metric)
: base(learners, crossValidation, ensembleStrategy,
new BackwardEliminationRegressionEnsembleSelection(metric,
ensembleStrategy, numberOfModelsToSelect))
{
}
/// <summary>
/// Classification model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy on how to combine the models</param>
/// <param name="ensembleSelection">Ensemble selection method used to find the beset subset of models</param>
public ClassificationModelSelectingEnsembleLearner(
IIndexedLearner <ProbabilityPrediction>[] learners,
ICrossValidation <ProbabilityPrediction> crossValidation,
Func <IClassificationEnsembleStrategy> ensembleStrategy,
IClassificationEnsembleSelection ensembleSelection)
{
m_learners = learners ?? throw new ArgumentNullException(nameof(learners));
m_crossValidation = crossValidation ?? throw new ArgumentNullException(nameof(crossValidation));
m_ensembleStrategy = ensembleStrategy ?? throw new ArgumentNullException(nameof(ensembleStrategy));
m_ensembleSelection = ensembleSelection ?? throw new ArgumentNullException(nameof(ensembleSelection));
}
/// <summary>
/// Regression model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble.
/// The selection of the best set of models is based on cross validation.
/// Trains several models and selects the best subset of models for the ensemble.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy on how to combine the models</param>
/// <param name="ensembleSelection">Ensemble selection method used to find the beset subset of models</param>
public RegressionModelSelectingEnsembleLearner(
IIndexedLearner <double>[] learners,
ICrossValidation <double> crossValidation,
Func <IRegressionEnsembleStrategy> ensembleStrategy,
IRegressionEnsembleSelection ensembleSelection)
{
m_learners = learners ?? throw new ArgumentNullException(nameof(learners));
m_crossValidation = crossValidation ?? throw new ArgumentNullException(nameof(crossValidation));
m_ensembleStrategy = ensembleStrategy ?? throw new ArgumentNullException(nameof(ensembleStrategy));
m_ensembleSelection = ensembleSelection ?? throw new ArgumentNullException(nameof(ensembleSelection));
}
/// <summary>
/// Regression model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble using greedy forward selection.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="numberOfModelsToSelect">Number of models to select</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy for ensembling models</param>
/// <param name="metric">Metric to minimize</param>
/// <param name="numberOfModelsFromStart">Number of models from start of the search.
/// The top n models will be selected based in their solo performance</param>
/// <param name="selectWithReplacement">If true the same model can be selected multiple times.
/// This will correspond to weighting the models. If false each model can only be selected once. Default is true</param>
public RegressionForwardSearchModelSelectingEnsembleLearner(
IIndexedLearner <double>[] learners,
int numberOfModelsToSelect,
ICrossValidation <double> crossValidation,
IRegressionEnsembleStrategy ensembleStrategy,
IMetric <double, double> metric,
int numberOfModelsFromStart = 1,
bool selectWithReplacement = true)
: base(learners, crossValidation, ensembleStrategy,
new ForwardSearchRegressionEnsembleSelection(metric, ensembleStrategy, numberOfModelsToSelect,
numberOfModelsFromStart, selectWithReplacement))
{
}
/// <summary>
/// Classification model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble using greedy forward selection.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="numberOfModelsToSelect">Number of models to select</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy for ensembling models</param>
/// <param name="metric">Metric to minimize</param>
/// <param name="numberOfModelsFromStart">Number of models from start of the search.
/// The top n models will be selected based in their solo performance</param>
/// <param name="selectWithReplacement">If true the same model can be selected multiple times.
/// This will correspond to weighting the models. If false each model can only be selected once. Default is true</param>
public ClassificationForwardSearchModelSelectingEnsembleLearner(
IIndexedLearner <ProbabilityPrediction>[] learners,
int numberOfModelsToSelect,
ICrossValidation <ProbabilityPrediction> crossValidation,
IClassificationEnsembleStrategy ensembleStrategy,
IMetric <double, ProbabilityPrediction> metric,
int numberOfModelsFromStart = 1,
bool selectWithReplacement = true)
: base(learners, crossValidation, ensembleStrategy,
new ForwardSearchClassificationEnsembleSelection(
metric, ensembleStrategy, numberOfModelsToSelect,
numberOfModelsFromStart, selectWithReplacement))
{
}
/// <summary>
/// Regression model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble using iterative random selection.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="numberOfModelsToSelect">Number of models to select</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy for ensembling models</param>
/// <param name="metric">Metric to minimize</param>
/// <param name="iterations">Number of iterations to random select model combinations.</param>
/// <param name="selectWithReplacement">If true the same model can be selected multiple times.</param>
/// <param name="seed"></param>
public RegressionRandomModelSelectingEnsembleLearner(
IIndexedLearner <double>[] learners,
int numberOfModelsToSelect,
ICrossValidation <double> crossValidation,
IRegressionEnsembleStrategy ensembleStrategy,
IMetric <double, double> metric,
int iterations = 50,
bool selectWithReplacement = true,
int seed = 42)
: base(learners, crossValidation, ensembleStrategy,
new RandomRegressionEnsembleSelection(metric, ensembleStrategy, numberOfModelsToSelect,
iterations, selectWithReplacement, seed))
{
}
/// <summary>
/// Classification model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble using iterative random selection.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="numberOfModelsToSelect">Number of models to select</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy for ensembling models</param>
/// <param name="metric">Metric to minimize</param>
/// <param name="iterations">Number of iterations to random select model combinations.</param>
/// <param name="selectWithReplacement">If true the same model can be selected multiple times.</param>
/// <param name="seed"></param>
public ClassificationRandomModelSelectingEnsembleLearner(
IIndexedLearner <ProbabilityPrediction>[] learners,
int numberOfModelsToSelect,
ICrossValidation <ProbabilityPrediction> crossValidation,
IClassificationEnsembleStrategy ensembleStrategy,
IMetric <double, ProbabilityPrediction> metric,
int iterations = 50,
bool selectWithReplacement = true,
int seed = 42)
: base(learners, crossValidation, ensembleStrategy,
new RandomClassificationEnsembleSelection(
metric, ensembleStrategy, numberOfModelsToSelect,
iterations, selectWithReplacement, seed))
{
}
/// <summary>
/// Stacking Regression Ensemble Learner.
/// Combines several models into a single ensemble model using a top or meta level model to combine the models.
/// The bottom level models generates output for the top level model using cross validation.
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="metaLearner">Meta learner or top level model for combining the ensemble models</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="includeOriginalFeaturesForMetaLearner">True; the meta learner also recieves the original features.
/// False; the meta learner only recieves the output of the ensemble models as features</param>
public RegressionStackingEnsembleLearner(IIndexedLearner <double>[] learners, Func <F64Matrix, double[], IPredictorModel <double> > metaLearner,
ICrossValidation <double> crossValidation, bool includeOriginalFeaturesForMetaLearner = true)
{
if (learners == null)
{
throw new ArgumentException("learners");
}
if (crossValidation == null)
{
throw new ArgumentException("crossValidation");
}
if (metaLearner == null)
{
throw new ArgumentException("metaLearner");
}
m_learners = learners;
m_crossValidation = crossValidation;
m_metaLearner = metaLearner;
m_includeOriginalFeaturesForMetaLearner = includeOriginalFeaturesForMetaLearner;
}
/// <summary>
/// Regression model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble.
/// The selection of the best set of models is based on cross validation.
/// Trains several models and selects the best subset of models for the ensemble.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy on how to combine the models</param>
/// <param name="ensembleSelection">Enemble selection method used to find the beset subset of models</param>
public RegressionModelSelectingEnsembleLearner(IIndexedLearner <double>[] learners, ICrossValidation <double> crossValidation,
Func <IRegressionEnsembleStrategy> ensembleStrategy, IRegressionEnsembleSelection ensembleSelection)
{
if (learners == null)
{
throw new ArgumentNullException("learners");
}
if (crossValidation == null)
{
throw new ArgumentNullException("crossValidation");
}
if (ensembleStrategy == null)
{
throw new ArgumentNullException("ensembleStrategy");
}
if (ensembleSelection == null)
{
throw new ArgumentNullException("ensembleSelection");
}
m_learners = learners;
m_crossValidation = crossValidation;
m_ensembleStrategy = ensembleStrategy;
m_ensembleSelection = ensembleSelection;
}
/// <summary>
/// Classification model selecting EnsembleLearner.
/// Trains several models and selects the best subset of models for the ensemble.
/// The selection of the best set of models is based on cross validation.
/// http://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf
/// </summary>
/// <param name="learners">Learners in the ensemble</param>
/// <param name="crossValidation">Cross validation method</param>
/// <param name="ensembleStrategy">Strategy on how to combine the models</param>
/// <param name="ensembleSelection">Ensemble selection method used to find the beset subset of models</param>
public ClassificationModelSelectingEnsembleLearner(IIndexedLearner <ProbabilityPrediction>[] learners, ICrossValidation <ProbabilityPrediction> crossValidation,
Func <IClassificationEnsembleStrategy> ensembleStrategy, IClassificationEnsembleSelection ensembleSelection)
{
if (learners == null)
{
throw new ArgumentNullException("learners");
}
if (crossValidation == null)
{
throw new ArgumentNullException("crossValidation");
}
if (ensembleStrategy == null)
{
throw new ArgumentNullException("ensembleStrategy");
}
if (ensembleSelection == null)
{
throw new ArgumentNullException("ensembleSelection");
}
m_learners = learners;
m_crossValidation = crossValidation;
m_ensembleStrategy = ensembleStrategy;
m_ensembleSelection = ensembleSelection;
}