Package | Description |
---|---|
org.apache.ignite.ml |
Root ML package.
|
org.apache.ignite.ml.clustering.gmm |
Contains Gauss Mixture Model clustering algorithm (see
GmmModel ). |
org.apache.ignite.ml.clustering.kmeans |
Contains kMeans clustering algorithm.
|
org.apache.ignite.ml.composition |
Contains classes for ensemble of models implementation.
|
org.apache.ignite.ml.composition.bagging |
Contains bootstrap aggregation (bagging) trainer allowing to combine some other trainers and
return a bagged version of them.
|
org.apache.ignite.ml.composition.boosting |
Contains Gradient Boosting regression and classification abstract classes
allowing regressor type selecting in child classes.
|
org.apache.ignite.ml.composition.combinators.parallel |
Contains parallel combinators of trainers and models.
|
org.apache.ignite.ml.composition.combinators.sequential |
Contains sequential combinators of trainers and models.
|
org.apache.ignite.ml.composition.stacking |
Contains classes used for training with stacking technique.
|
org.apache.ignite.ml.environment.logging |
Package contains several logging strategy realisations.
|
org.apache.ignite.ml.inference |
Root package for model inference functionality.
|
org.apache.ignite.ml.inference.parser |
Root package for model inference parsers.
|
org.apache.ignite.ml.knn |
Contains main APIs for kNN algorithms.
|
org.apache.ignite.ml.knn.ann |
Contains main APIs for ANN classification algorithms.
|
org.apache.ignite.ml.knn.classification |
Contains main APIs for kNN classification algorithms.
|
org.apache.ignite.ml.knn.regression |
Contains helper classes for kNN regression algorithms.
|
org.apache.ignite.ml.multiclass |
Contains various multi-classifier models and trainers.
|
org.apache.ignite.ml.naivebayes.discrete |
Contains Bernoulli naive Bayes classifier.
|
org.apache.ignite.ml.naivebayes.gaussian |
Contains Gaussian naive Bayes classifier.
|
org.apache.ignite.ml.nn |
Contains neural networks and related classes.
|
org.apache.ignite.ml.optimization |
Contains implementations of optimization algorithms and related classes.
|
org.apache.ignite.ml.pipeline |
Contains Pipeline API.
|
org.apache.ignite.ml.recommendation |
Contains recommendation system framework.
|
org.apache.ignite.ml.regressions.linear |
Contains various linear regressions.
|
org.apache.ignite.ml.regressions.logistic |
Contains various logistic regressions.
|
org.apache.ignite.ml.selection.cv |
Root package for cross-validation algorithms.
|
org.apache.ignite.ml.selection.scoring.cursor |
Util classes used for score calculation.
|
org.apache.ignite.ml.selection.scoring.evaluator |
Package for model evaluator classes.
|
org.apache.ignite.ml.svm |
Contains main APIs for SVM(support vector machines) algorithms.
|
org.apache.ignite.ml.trainers |
Contains model trainers.
|
org.apache.ignite.ml.tree |
Root package for decision trees.
|
org.apache.ignite.ml.tree.boosting |
Contains implementation of gradient boosting on trees.
|
org.apache.ignite.ml.tree.randomforest.data |
Package contains helper data structures for random forest implementation.
|
org.apache.ignite.ml.xgboost |
Base package for XGBoost model parser, correspondent DTOs and util classes.
|
Modifier and Type | Method and Description |
---|---|
default <V1> IgniteModel<V1,V> |
IgniteModel.andBefore(IgniteFunction<V1,T> before)
Get a composition model of the form
x -> mdl(before(x)) . |
default <V1> IgniteModel<T,V1> |
IgniteModel.andThen(IgniteFunction<V,V1> after)
Get a composition model of the form
x -> after(mdl(x)) . |
default <V1> IgniteModel<T,V1> |
IgniteModel.andThen(IgniteModel<V,V1> after)
Get a composition model of the form
x -> after(mdl(x)) . |
default <X,W> IgniteModel<T,X> |
IgniteModel.combine(IgniteModel<T,W> other,
BiFunction<V,W,X> combiner)
Combines this model with other model via specified combiner
|
Modifier and Type | Method and Description |
---|---|
default <V1> IgniteModel<T,V1> |
IgniteModel.andThen(IgniteModel<V,V1> after)
Get a composition model of the form
x -> after(mdl(x)) . |
default <X,W> IgniteModel<T,X> |
IgniteModel.combine(IgniteModel<T,W> other,
BiFunction<V,W,X> combiner)
Combines this model with other model via specified combiner
|
Modifier and Type | Class and Description |
---|---|
class |
GmmModel
Gaussian Mixture Model.
|
Modifier and Type | Interface and Description |
---|---|
interface |
Clusterer<P,M extends IgniteModel>
Base interface for clusterers.
|
Modifier and Type | Interface and Description |
---|---|
interface |
ClusterizationModel<P,V>
Base interface for all clusterization models.
|
Modifier and Type | Class and Description |
---|---|
class |
KMeansModel
This class encapsulates result of clusterization by KMeans algorithm.
|
Modifier and Type | Class and Description |
---|---|
class |
ModelOnFeaturesSubspace
Model trained on a features subspace with mapping from original features space to subspace.
|
class |
ModelsComposition
Model consisting of several models and prediction aggregation strategy.
|
Modifier and Type | Method and Description |
---|---|
static <I,O,M extends IgniteModel<I,O>,L> |
CompositionUtils.unsafeCoerce(DatasetTrainer<? extends M,L> trainer)
Perform blurring of model type of given trainer to
IgniteModel<I, O> , where I, O are input and output
types of original model. |
Modifier and Type | Method and Description |
---|---|
IgniteModel<Vector,Double> |
ModelOnFeaturesSubspace.getMdl()
Returns model.
|
Modifier and Type | Method and Description |
---|---|
List<IgniteModel<Vector,Double>> |
ModelsComposition.getModels()
Returns containing models.
|
List<IgniteModel<Vector,Double>> |
ModelsCompositionFormat.models() |
static <I,O,M extends IgniteModel<I,O>,L> |
CompositionUtils.unsafeCoerce(DatasetTrainer<? extends M,L> trainer)
Perform blurring of model type of given trainer to
IgniteModel<I, O> , where I, O are input and output
types of original model. |
Constructor and Description |
---|
ModelsComposition(List<? extends IgniteModel<Vector,Double>> models,
PredictionsAggregator predictionsAggregator)
Constructs a new instance of composition of models.
|
ModelsCompositionFormat(List<IgniteModel<Vector,Double>> models,
PredictionsAggregator predictionsAggregator)
Creates an instance of ModelsCompositionFormat.
|
Modifier and Type | Class and Description |
---|---|
class |
BaggedModel
This class represents model produced by
BaggedTrainer . |
Constructor and Description |
---|
BaggedTrainer(DatasetTrainer<? extends IgniteModel,L> tr,
PredictionsAggregator aggregator,
int ensembleSize,
double subsampleRatio,
int featuresVectorSize,
int featureSubspaceDim)
Construct instance of this class with given parameters.
|
Modifier and Type | Class and Description |
---|---|
static class |
GDBTrainer.GDBModel
GDB model.
|
Modifier and Type | Field and Description |
---|---|
protected IgniteSupplier<DatasetTrainer<? extends IgniteModel<Vector,Double>,Double>> |
GDBLearningStrategy.baseMdlTrainerBuilder
Base model trainer builder.
|
Modifier and Type | Method and Description |
---|---|
<K,V> List<IgniteModel<Vector,Double>> |
GDBLearningStrategy.learnModels(DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> preprocessor)
Implementation of gradient boosting iterations.
|
<K,V> List<IgniteModel<Vector,Double>> |
GDBLearningStrategy.update(GDBTrainer.GDBModel mdlToUpdate,
DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> preprocessor)
Gets state of model in arguments, compare it with training parameters of trainer and if they are fit then trainer
updates model in according to new data and return new model.
|
Modifier and Type | Method and Description |
---|---|
GDBLearningStrategy |
GDBLearningStrategy.withBaseModelTrainerBuilder(IgniteSupplier<DatasetTrainer<? extends IgniteModel<Vector,Double>,Double>> buildBaseMdlTrainer)
Sets base model builder.
|
Constructor and Description |
---|
GDBModel(List<? extends IgniteModel<Vector,Double>> models,
WeightedPredictionsAggregator predictionsAggregator,
IgniteFunction<Double,Double> internalToExternalLblMapping)
Creates an instance of GDBModel.
|
Modifier and Type | Class and Description |
---|---|
class |
ModelsParallelComposition<I,O>
Parallel composition of models.
|
Modifier and Type | Method and Description |
---|---|
static <I,O,M extends IgniteModel<I,O>,T extends DatasetTrainer<M,L>,L> |
TrainersParallelComposition.of(List<T> trainers)
Create parallel composition of trainers contained in a given list.
|
Modifier and Type | Method and Description |
---|---|
<K,V> IgniteModel<I,List<O>> |
TrainersParallelComposition.fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> preprocessor)
Trains model based on the specified data.
|
<K,V> IgniteModel<I,List<O>> |
TrainersParallelComposition.update(IgniteModel<I,List<O>> mdl,
DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> preprocessor)
Gets state of model in arguments, compare it with training parameters of trainer and if they are fit then trainer
updates model in according to new data and return new model.
|
protected <K,V> IgniteModel<I,List<O>> |
TrainersParallelComposition.updateModel(IgniteModel<I,List<O>> mdl,
DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> preprocessor)
This method is never called, instead of constructing logic of update from
DatasetTrainer.isUpdateable(IgniteModel) and DatasetTrainer.updateModel(IgniteModel, DatasetBuilder, Preprocessor)
in this class we explicitly override update method. |
Modifier and Type | Method and Description |
---|---|
List<IgniteModel<I,O>> |
ModelsParallelComposition.submodels()
List of submodels constituting this model.
|
Modifier and Type | Method and Description |
---|---|
boolean |
TrainersParallelComposition.isUpdateable(IgniteModel<I,List<O>> mdl)
This method is never called, instead of constructing logic of update from
DatasetTrainer.isUpdateable(M) and
DatasetTrainer.updateModel(M, org.apache.ignite.ml.dataset.DatasetBuilder<K, V>, org.apache.ignite.ml.preprocessing.Preprocessor<K, V>) in this class we explicitly override update method. |
<K,V> IgniteModel<I,List<O>> |
TrainersParallelComposition.update(IgniteModel<I,List<O>> mdl,
DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> preprocessor)
Gets state of model in arguments, compare it with training parameters of trainer and if they are fit then trainer
updates model in according to new data and return new model.
|
protected <K,V> IgniteModel<I,List<O>> |
TrainersParallelComposition.updateModel(IgniteModel<I,List<O>> mdl,
DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> preprocessor)
This method is never called, instead of constructing logic of update from
DatasetTrainer.isUpdateable(IgniteModel) and DatasetTrainer.updateModel(IgniteModel, DatasetBuilder, Preprocessor)
in this class we explicitly override update method. |
Constructor and Description |
---|
ModelsParallelComposition(List<IgniteModel<I,O>> submodels)
Construc an instance of this class from list of submodels.
|
Modifier and Type | Class and Description |
---|---|
class |
ModelsSequentialComposition<I,O1,O2>
Sequential composition of models.
|
Modifier and Type | Field and Description |
---|---|
protected IgniteBiFunction<Integer,? super IgniteModel<I,O1>,IgniteFunction<LabeledVector<L>,LabeledVector<L>>> |
TrainersSequentialComposition.datasetMapping
Dataset mapping.
|
Modifier and Type | Method and Description |
---|---|
IgniteModel<I,O1> |
ModelsSequentialComposition.firstModel()
Get first model.
|
IgniteModel<O1,O2> |
ModelsSequentialComposition.secondModel()
Get second model.
|
Modifier and Type | Method and Description |
---|---|
DatasetTrainer<IgniteModel<I,O2>,L> |
TrainersSequentialComposition.unsafeSimplyTyped()
Performs coersion of this trainer to
DatasetTrainer<IgniteModel<I, O2>, L> . |
Modifier and Type | Method and Description |
---|---|
static <I,O,L> TrainersSequentialComposition<I,O,O,L> |
TrainersSequentialComposition.ofSame(DatasetTrainer<? extends IgniteModel<I,O>,L> tr,
IgniteBiFunction<Integer,? super IgniteModel<I,O>,IgniteFunction<LabeledVector<L>,LabeledVector<L>>> datasetMapping,
IgniteBiPredicate<Integer,IgniteModel<I,O>> shouldStop,
IgniteFunction<O,I> out2In)
Construct sequential composition of same trainers.
|
static <I,O,L> TrainersSequentialComposition<I,O,O,L> |
TrainersSequentialComposition.ofSame(DatasetTrainer<? extends IgniteModel<I,O>,L> tr,
IgniteBiFunction<Integer,? super IgniteModel<I,O>,IgniteFunction<LabeledVector<L>,LabeledVector<L>>> datasetMapping,
IgniteBiPredicate<Integer,IgniteModel<I,O>> shouldStop,
IgniteFunction<O,I> out2In)
Construct sequential composition of same trainers.
|
static <I,O,L> TrainersSequentialComposition<I,O,O,L> |
TrainersSequentialComposition.ofSame(DatasetTrainer<? extends IgniteModel<I,O>,L> tr,
IgniteBiFunction<Integer,? super IgniteModel<I,O>,IgniteFunction<LabeledVector<L>,LabeledVector<L>>> datasetMapping,
IgniteBiPredicate<Integer,IgniteModel<I,O>> shouldStop,
IgniteFunction<O,I> out2In)
Construct sequential composition of same trainers.
|
static <I,O> ModelsSequentialComposition<I,O,O> |
ModelsSequentialComposition.ofSame(List<? extends IgniteModel<I,O>> lst,
IgniteFunction<O,I> output2Input)
Get sequential composition of submodels with same type.
|
Constructor and Description |
---|
ModelsSequentialComposition(IgniteModel<I,O1> mdl1,
IgniteModel<O1,O2> mdl2)
Construct instance of this class from two given models.
|
ModelsSequentialComposition(IgniteModel<I,O1> mdl1,
IgniteModel<O1,O2> mdl2)
Construct instance of this class from two given models.
|
Modifier and Type | Class and Description |
---|---|
class |
SimpleStackedDatasetTrainer<I,O,AM extends IgniteModel<I,O>,L>
DatasetTrainer with same type of input and output of submodels. |
class |
StackedDatasetTrainer<IS,IA,O,AM extends IgniteModel<IA,O>,L>
DatasetTrainer encapsulating stacking technique for model training. |
class |
StackedModel<IS,IA,O,AM extends IgniteModel<IA,O>>
This is a wrapper for model produced by
StackedDatasetTrainer . |
class |
StackedVectorDatasetTrainer<O,AM extends IgniteModel<Vector,O>,L>
StackedDatasetTrainer with Vector as submodels input and output. |
Modifier and Type | Class and Description |
---|---|
class |
StackedModel<IS,IA,O,AM extends IgniteModel<IA,O>>
This is a wrapper for model produced by
StackedDatasetTrainer . |
Modifier and Type | Method and Description |
---|---|
<M1 extends IgniteModel<Matrix,Matrix>> |
StackedVectorDatasetTrainer.addMatrix2MatrixTrainer(DatasetTrainer<M1,L> trainer)
|
<M1 extends IgniteModel<I,I>> |
SimpleStackedDatasetTrainer.addTrainer(DatasetTrainer<M1,L> trainer)
Adds submodel trainer along with converters needed on training and inference stages.
|
<M1 extends IgniteModel<IS,IA>> |
StackedDatasetTrainer.addTrainer(DatasetTrainer<M1,L> trainer)
Adds submodel trainer along with converters needed on training and inference stages.
|
<M1 extends IgniteModel<Vector,Vector>> |
StackedVectorDatasetTrainer.addTrainer(DatasetTrainer<M1,L> trainer)
Adds submodel trainer along with converters needed on training and inference stages.
|
<M1 extends IgniteModel<Vector,Double>> |
StackedVectorDatasetTrainer.addTrainerWithDoubleOutput(DatasetTrainer<M1,L> trainer)
Shortcut for adding trainer
Vector -> Double where this trainer is treated as Vector -> Vector , where
output Vector is constructed by wrapping double value. |
Constructor and Description |
---|
StackedDatasetTrainer(DatasetTrainer<AM,L> aggregatorTrainer,
IgniteBinaryOperator<IA> aggregatingInputMerger,
IgniteFunction<IS,IA> submodelInput2AggregatingInputConverter,
List<DatasetTrainer<IgniteModel<IS,IA>,L>> submodelsTrainers,
IgniteFunction<Vector,IS> vector2SubmodelInputConverter,
IgniteFunction<IA,Vector> submodelOutput2VectorConverter)
Create instance of this class.
|
Modifier and Type | Method and Description |
---|---|
<K,V> IgniteModel<K,V> |
NoOpLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
<K,V> IgniteModel<K,V> |
CustomMLLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
<K,V> IgniteModel<K,V> |
MLLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
<K,V> IgniteModel<K,V> |
ConsoleLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
Modifier and Type | Method and Description |
---|---|
<K,V> IgniteModel<K,V> |
NoOpLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
<K,V> IgniteModel<K,V> |
CustomMLLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
<K,V> IgniteModel<K,V> |
MLLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
<K,V> IgniteModel<K,V> |
ConsoleLogger.log(MLLogger.VerboseLevel verboseLevel,
IgniteModel<K,V> mdl)
Log model according to toString method.
|
Modifier and Type | Method and Description |
---|---|
static <I extends Serializable,O extends Serializable> |
IgniteModelStorageUtil.saveModel(Ignite ignite,
IgniteModel<I,O> mdl,
String name)
Saved specified model with specified name.
|
Modifier and Type | Method and Description |
---|---|
IgniteModel<I,O> |
IgniteModelParser.parse(byte[] mdl)
Accepts serialized model represented by byte array, parses it and returns
Model . |
Modifier and Type | Class and Description |
---|---|
class |
KNNModel<L>
KNN model build on top of distribtued spatial indices.
|
class |
NNClassificationModel
Common methods and fields for all kNN and aNN models
to predict label based on neighbours' labels.
|
Modifier and Type | Class and Description |
---|---|
class |
ANNClassificationModel
ANN model to predict labels in multi-class classification task.
|
Modifier and Type | Class and Description |
---|---|
class |
KNNClassificationModel
KNN classification model.
|
Modifier and Type | Class and Description |
---|---|
class |
KNNRegressionModel
KNN regression model.
|
Modifier and Type | Class and Description |
---|---|
class |
MultiClassModel<M extends IgniteModel<Vector,Double>>
Base class for multi-classification model for set of classifiers.
|
class |
OneVsRestTrainer<M extends IgniteModel<Vector,Double>>
This is a common heuristic trainer for multi-class labeled models.
|
Modifier and Type | Class and Description |
---|---|
class |
MultiClassModel<M extends IgniteModel<Vector,Double>>
Base class for multi-classification model for set of classifiers.
|
Modifier and Type | Class and Description |
---|---|
class |
DiscreteNaiveBayesModel
Discrete naive Bayes model which predicts result value
y belongs to a class C_k, k in [0..K] as
{@code p(C_k,y) =x_1*p_k1^x *... |
Modifier and Type | Class and Description |
---|---|
class |
GaussianNaiveBayesModel
Simple naive Bayes model which predicts result value
y belongs to a class C_k, k in [0..K] as {@code
p(C_k,y) = p(C_k)*p(y_1,C_k) *... |
Modifier and Type | Class and Description |
---|---|
class |
MultilayerPerceptron
Class encapsulating logic of multilayer perceptron.
|
Modifier and Type | Interface and Description |
---|---|
interface |
SmoothParametrized<M extends Parametrized<M>>
Interface for models which are smooth functions of their parameters.
|
Modifier and Type | Class and Description |
---|---|
class |
PipelineMdl<K,V>
Wraps the model produced by
Pipeline . |
Modifier and Type | Method and Description |
---|---|
IgniteModel<Vector,Double> |
PipelineMdl.getInternalMdl() |
Modifier and Type | Method and Description |
---|---|
PipelineMdl<K,V> |
PipelineMdl.withInternalMdl(IgniteModel<Vector,Double> internalMdl) |
Modifier and Type | Class and Description |
---|---|
class |
RecommendationModel<O extends Serializable,S extends Serializable>
Recommendation model that predicts rating for
ObjectSubjectPair . |
Modifier and Type | Class and Description |
---|---|
class |
LinearRegressionModel
Simple linear regression model which predicts result value Y as a linear combination of input variables:
Y = weights * X + intercept.
|
Modifier and Type | Class and Description |
---|---|
class |
LogisticRegressionModel
Logistic regression (logit model) is a generalized linear model used for binomial regression.
|
Modifier and Type | Class and Description |
---|---|
class |
AbstractCrossValidation<M extends IgniteModel<Vector,L>,L,K,V>
Cross validation score calculator.
|
class |
CrossValidation<M extends IgniteModel<Vector,L>,L,K,V>
Cross validation score calculator.
|
class |
DebugCrossValidation<M extends IgniteModel<Vector,L>,L,K,V>
Cross validation score calculator.
|
Constructor and Description |
---|
CacheBasedLabelPairCursor(IgniteCache<K,V> upstreamCache,
IgniteBiPredicate<K,V> filter,
Preprocessor<K,V> preprocessor,
IgniteModel<Vector,L> mdl)
Constructs a new instance of cache based truth with prediction cursor.
|
CacheBasedLabelPairCursor(IgniteCache<K,V> upstreamCache,
Preprocessor<K,V> preprocessor,
IgniteModel<Vector,L> mdl)
Constructs a new instance of cache based truth with prediction cursor.
|
LocalLabelPairCursor(Map<K,V> upstreamMap,
IgniteBiPredicate<K,V> filter,
Preprocessor<K,V> preprocessor,
IgniteModel<Vector,L> mdl)
Constructs a new instance of local truth with prediction cursor.
|
Modifier and Type | Method and Description |
---|---|
static <K,V> BinaryClassificationMetricValues |
Evaluator.evaluate(IgniteCache<K,V> dataCache,
IgniteBiPredicate<K,V> filter,
IgniteModel<Vector,Double> mdl,
Preprocessor<K,V> preprocessor)
Computes the given metrics on the given cache.
|
static <L,K,V> double |
Evaluator.evaluate(IgniteCache<K,V> dataCache,
IgniteBiPredicate<K,V> filter,
IgniteModel<Vector,L> mdl,
Preprocessor<K,V> preprocessor,
Metric<L> metric)
Computes the given metric on the given cache.
|
static <K,V> BinaryClassificationMetricValues |
Evaluator.evaluate(IgniteCache<K,V> dataCache,
IgniteModel<Vector,Double> mdl,
Preprocessor<K,V> preprocessor)
Computes the given metrics on the given cache.
|
static <L,K,V> double |
Evaluator.evaluate(IgniteCache<K,V> dataCache,
IgniteModel<Vector,L> mdl,
Preprocessor<K,V> preprocessor,
Metric<L> metric)
Computes the given metric on the given cache.
|
static <K,V> BinaryClassificationMetricValues |
Evaluator.evaluate(Map<K,V> dataCache,
IgniteBiPredicate<K,V> filter,
IgniteModel<Vector,Double> mdl,
Preprocessor<K,V> preprocessor)
Computes the given metrics on the given cache.
|
static <L,K,V> double |
Evaluator.evaluate(Map<K,V> dataCache,
IgniteBiPredicate<K,V> filter,
IgniteModel<Vector,L> mdl,
Preprocessor<K,V> preprocessor,
Metric<L> metric)
Computes the given metric on the given cache.
|
static <K,V> BinaryClassificationMetricValues |
Evaluator.evaluate(Map<K,V> dataCache,
IgniteModel<Vector,Double> mdl,
Preprocessor<K,V> preprocessor)
Computes the given metrics on the given cache.
|
static <L,K,V> double |
Evaluator.evaluate(Map<K,V> dataCache,
IgniteModel<Vector,L> mdl,
Preprocessor<K,V> preprocessor,
Metric<L> metric)
Computes the given metric on the given cache.
|
static <K,V> RegressionMetricValues |
Evaluator.evaluateRegression(IgniteCache<K,V> dataCache,
IgniteBiPredicate<K,V> filter,
IgniteModel<Vector,Double> mdl,
Preprocessor<K,V> preprocessor)
Computes the regression metrics on the given cache.
|
Modifier and Type | Class and Description |
---|---|
class |
SVMLinearClassificationModel
Base class for SVM linear classification model.
|
Modifier and Type | Class and Description |
---|---|
class |
AdaptableDatasetModel<I,O,IW,OW,M extends IgniteModel<IW,OW>>
Model which is composition of form
before `andThen` inner Mdl `andThen` after . |
class |
AdaptableDatasetTrainer<I,O,IW,OW,M extends IgniteModel<IW,OW>,L>
Type used to adapt input and output types of wrapped
DatasetTrainer . |
class |
DatasetTrainer<M extends IgniteModel,L>
Interface for trainers.
|
class |
MultiLabelDatasetTrainer<M extends IgniteModel>
Interface for trainers that trains on dataset with multiple label per object.
|
class |
SingleLabelDatasetTrainer<M extends IgniteModel>
Interface for trainers that trains on dataset with singe label per object.
|
Modifier and Type | Class and Description |
---|---|
class |
AdaptableDatasetModel<I,O,IW,OW,M extends IgniteModel<IW,OW>>
Model which is composition of form
before `andThen` inner Mdl `andThen` after . |
Modifier and Type | Method and Description |
---|---|
<O1,M1 extends IgniteModel<O,O1>> |
AdaptableDatasetTrainer.andThen(DatasetTrainer<M1,L> tr,
IgniteFunction<AdaptableDatasetModel<I,O,IW,OW,M>,IgniteFunction<LabeledVector<L>,LabeledVector<L>>> datasetMappingProducer)
Create a
TrainersSequentialComposition of whis trainer and specified trainer. |
static <M extends IgniteModel<Vector,Double>,L> |
TrainerTransformers.makeBagged(DatasetTrainer<M,L> trainer,
int ensembleSize,
double subsampleRatio,
int featureVectorSize,
int featuresSubspaceDim,
PredictionsAggregator aggregator)
Add bagging logic to a given trainer.
|
static <I,O,M extends IgniteModel<I,O>,L> |
AdaptableDatasetTrainer.of(DatasetTrainer<M,L> wrapped)
Construct instance of this class from a given
DatasetTrainer . |
<M1 extends IgniteModel<IW,OW>> |
AdaptableDatasetModel.withInnerModel(M1 mdl)
Create new instance of this class with changed inner model.
|
Modifier and Type | Method and Description |
---|---|
static <I,L> DatasetTrainer<IgniteModel<I,I>,L> |
DatasetTrainer.identityTrainer()
Returns the trainer which returns identity model.
|
Modifier and Type | Method and Description |
---|---|
<O1> AdaptableDatasetModel<I,O1,IW,OW,M> |
AdaptableDatasetModel.andThen(IgniteModel<O,O1> after)
Get a composition model of the form
x -> after(mdl(x)) . |
Modifier and Type | Method and Description |
---|---|
static <L> BaggedTrainer<L> |
TrainerTransformers.makeBagged(DatasetTrainer<? extends IgniteModel,L> trainer,
int ensembleSize,
double subsampleRatio,
PredictionsAggregator aggregator)
Add bagging logic to a given trainer.
|
Modifier and Type | Interface and Description |
---|---|
interface |
DecisionTreeNode
Base interface for decision tree nodes.
|
Modifier and Type | Class and Description |
---|---|
class |
DecisionTreeConditionalNode
Decision tree conditional (non-leaf) node.
|
class |
DecisionTreeLeafNode
Decision tree leaf node which contains value.
|
Modifier and Type | Method and Description |
---|---|
<K,V> List<IgniteModel<Vector,Double>> |
GDBOnTreesLearningStrategy.update(GDBTrainer.GDBModel mdlToUpdate,
DatasetBuilder<K,V> datasetBuilder,
Preprocessor<K,V> vectorizer)
Gets state of model in arguments, compare it with training parameters of trainer and if they are fit then trainer
updates model in according to new data and return new model.
|
Modifier and Type | Class and Description |
---|---|
class |
TreeNode
Decision tree node class.
|
class |
TreeRoot
Tree root class.
|
Modifier and Type | Class and Description |
---|---|
class |
XGModelComposition
XGBoost model composition.
|
GridGain In-Memory Computing Platform : ver. 8.9.14 Release Date : November 5 2024