ai.libs.jaicore.ml.core.dataset.sampling.inmemory.stratified.sampling

Class AttributeBasedStratiAmountSelectorAndAssigner<I extends INumericLabeledAttributeArrayInstance<?>,D extends IOrderedLabeledAttributeArrayDataset<I,?>>

java.lang.Object

ai.libs.jaicore.ml.core.dataset.sampling.inmemory.stratified.sampling.AttributeBasedStratiAmountSelectorAndAssigner<I,D>

All Implemented Interfaces:

IStratiAmountSelector<D>, IStratiAssigner<I,D>

public class AttributeBasedStratiAmountSelectorAndAssigner<I extends INumericLabeledAttributeArrayInstance<?>,D extends IOrderedLabeledAttributeArrayDataset<I,?>>
extends java.lang.Object
implements IStratiAmountSelector<D>, IStratiAssigner<I,D>

This class is responsible for computing the amount of strati in attribute-based stratified sampling and assigning elements to the strati.

Constructor Summary

Constructors

Constructor and Description
AttributeBasedStratiAmountSelectorAndAssigner() SCALE-54: Explicitly allow to not provide an attribute list
AttributeBasedStratiAmountSelectorAndAssigner(java.util.List<java.lang.Integer> attributeIndices)
AttributeBasedStratiAmountSelectorAndAssigner(java.util.List<java.lang.Integer> attributeIndices, DiscretizationHelper.DiscretizationStrategy discretizationStrategy, int numberOfCategories)
AttributeBasedStratiAmountSelectorAndAssigner(java.util.List<java.lang.Integer> attributeIndices, java.util.Map<java.lang.Integer,AttributeDiscretizationPolicy> discretizationPolicies)

Method Summary

Modifier and Type	Method and Description
int	assignToStrati(I datapoint) Custom logic for assigning datapoints into strati.
int	getNumCPUs()
void	init(D dataset) Initializes the algorithm for stratum assignment.
void	init(D dataset, int stratiAmount) Initialize custom assigner if necessary.
int	selectStratiAmount(D dataset) Select a suitable amount of strati for a Dataset.
void	setNumCPUs(int numberOfCPUs) Sets the number of CPU cores that can be used for parallel computation

Methods inherited from class java.lang.Object

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

Constructor Detail

AttributeBasedStratiAmountSelectorAndAssigner

public AttributeBasedStratiAmountSelectorAndAssigner()

SCALE-54: Explicitly allow to not provide an attribute list

AttributeBasedStratiAmountSelectorAndAssigner

public AttributeBasedStratiAmountSelectorAndAssigner(java.util.List<java.lang.Integer> attributeIndices)

AttributeBasedStratiAmountSelectorAndAssigner

public AttributeBasedStratiAmountSelectorAndAssigner(java.util.List<java.lang.Integer> attributeIndices,
                                                     DiscretizationHelper.DiscretizationStrategy discretizationStrategy,
                                                     int numberOfCategories)

AttributeBasedStratiAmountSelectorAndAssigner

public AttributeBasedStratiAmountSelectorAndAssigner(java.util.List<java.lang.Integer> attributeIndices,
                                                     java.util.Map<java.lang.Integer,AttributeDiscretizationPolicy> discretizationPolicies)

Method Detail

selectStratiAmount

public int selectStratiAmount(D dataset)

Description copied from interface: IStratiAmountSelector

Select a suitable amount of strati for a Dataset.

Specified by:

selectStratiAmount in interface IStratiAmountSelector<D extends IOrderedLabeledAttributeArrayDataset<I,?>>

Parameters:

dataset - The dataset that will be stratified.

Returns:

The determined amount of strati.

setNumCPUs

public void setNumCPUs(int numberOfCPUs)

Description copied from interface: IStratiAmountSelector

Sets the number of CPU cores that can be used for parallel computation

Specified by:

setNumCPUs in interface IStratiAmountSelector<D extends IOrderedLabeledAttributeArrayDataset<I,?>>

Specified by:

setNumCPUs in interface IStratiAssigner<I extends INumericLabeledAttributeArrayInstance<?>,D extends IOrderedLabeledAttributeArrayDataset<I,?>>

getNumCPUs

public int getNumCPUs()

Specified by:

getNumCPUs in interface IStratiAmountSelector<D extends IOrderedLabeledAttributeArrayDataset<I,?>>

Specified by:

getNumCPUs in interface IStratiAssigner<I extends INumericLabeledAttributeArrayInstance<?>,D extends IOrderedLabeledAttributeArrayDataset<I,?>>

init

public void init(D dataset,
                 int stratiAmount)

Description copied from interface: IStratiAssigner

Initialize custom assigner if necessary.

Specified by:

init in interface IStratiAssigner<I extends INumericLabeledAttributeArrayInstance<?>,D extends IOrderedLabeledAttributeArrayDataset<I,?>>

Parameters:

dataset - The dataset the datapoints will be sampled from.

stratiAmount - The predetermined amount of strati the dataset will be stratified into.

init

public void init(D dataset)

Initializes the algorithm for stratum assignment.

Parameters:

dataset -

assignToStrati

public int assignToStrati(I datapoint)

Description copied from interface: IStratiAssigner

Custom logic for assigning datapoints into strati.

Specified by:

assignToStrati in interface IStratiAssigner<I extends INumericLabeledAttributeArrayInstance<?>,D extends IOrderedLabeledAttributeArrayDataset<I,?>>

Parameters:

datapoint - The datapoint that has to be assigned.

Returns:

The index of the strati the datapoint will be assigned into.