org.apache.ignite.ml.dataset.primitive (Ignite 2.7.2)

Class Summary
Class	Description
DatasetWrapper<C extends Serializable,D extends AutoCloseable>	A dataset wrapper that allows to introduce new functionality based on common `compute` methods.
FeatureMatrixWithLabelsOnHeapData	A partition `data` of the containing matrix of features and vector of labels stored in heap.
FeatureMatrixWithLabelsOnHeapDataBuilder<K,V,C extends Serializable>	A partition `data` builder that makes `DecisionTreeData`.
SimpleDataset<C extends Serializable>	A simple dataset introduces additional methods based on a matrix of features.
SimpleLabeledDataset<C extends Serializable>	A simple labeled dataset introduces additional methods based on a matrix of features and labels vector.

Package org.apache.ignite.ml.dataset.primitive Description

Package that contains basic primitives build on top of Dataset. Primitives are simple components that can be used in other algorithms based on the dataset infrastructure or for debugging. Primitives include partition context implementations, partition data implementations and extensions of dataset. Partition context and data implementations can be used in other algorithm in case these algorithm doesn't need to keep specific data and can work with standard primitive data or context. Extensions of dataset provides basic most often used functions that can be used for debugging or data analysis.