org.apache.ignite.ml.tree.randomforest.data.statistics

Class NormalDistributionStatisticsComputer

java.lang.Object

org.apache.ignite.ml.tree.randomforest.data.statistics.NormalDistributionStatisticsComputer

All Implemented Interfaces:

Serializable

public class NormalDistributionStatisticsComputer
extends Object
implements Serializable

Normal distribution parameters computer logic.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
NormalDistributionStatisticsComputer()

Method Summary

Modifier and Type	Method and Description
List<NormalDistributionStatistics>	computeStatistics(List<FeatureMeta> meta, Dataset<EmptyContext,BootstrappedDatasetPartition> dataset) Computes statistics of normal distribution on features in dataset.
List<NormalDistributionStatistics>	computeStatsOnPartition(BootstrappedDatasetPartition part, List<FeatureMeta> meta) Aggregates normal distribution statistics for continual features in dataset partition.
List<NormalDistributionStatistics>	reduceStats(List<NormalDistributionStatistics> left, List<NormalDistributionStatistics> right, List<FeatureMeta> meta) Merges statistics on features from two partitions.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

NormalDistributionStatisticsComputer

public NormalDistributionStatisticsComputer()

Method Detail

computeStatistics

public List<NormalDistributionStatistics> computeStatistics(List<FeatureMeta> meta,
                                                            Dataset<EmptyContext,BootstrappedDatasetPartition> dataset)

Computes statistics of normal distribution on features in dataset.

Parameters:

meta - Meta.

dataset - Dataset.

computeStatsOnPartition

public List<NormalDistributionStatistics> computeStatsOnPartition(BootstrappedDatasetPartition part,
                                                                  List<FeatureMeta> meta)

Aggregates normal distribution statistics for continual features in dataset partition.

Parameters:

part - Partition.

meta - Meta.

Returns:

Statistics for each feature.

reduceStats

public List<NormalDistributionStatistics> reduceStats(List<NormalDistributionStatistics> left,
                                                      List<NormalDistributionStatistics> right,
                                                      List<FeatureMeta> meta)

Merges statistics on features from two partitions.

Parameters:

left - Left.

right - Right.

meta - Features meta.

Returns:

Plus of statistics for each features.