Serialized Form

Package weka.associations

Class weka.associations.AbstractAssociator extends java.lang.Object implements Serializable

serialVersionUID: -3017644543382432070L

Class weka.associations.Apriori extends AbstractAssociator implements Serializable

serialVersionUID: 3277498842319212687L

Serialized Fields

m_minSupport

double m_minSupport

The minimum support.

m_upperBoundMinSupport

double m_upperBoundMinSupport

The upper bound on the support

m_lowerBoundMinSupport

double m_lowerBoundMinSupport

The lower bound for the minimum support.

m_metricType

int m_metricType

The selected metric type.

m_minMetric

double m_minMetric

The minimum metric score.

m_numRules

int m_numRules

The maximum number of rules that are output.

m_delta

double m_delta

Delta by which m_minSupport is decreased in each iteration.

m_significanceLevel

double m_significanceLevel

Significance level for optional significance test.

m_cycles

int m_cycles

Number of cycles used before required number of rules was one.

m_Ls

FastVector m_Ls

The set of all sets of itemsets L.

m_hashtables

FastVector m_hashtables

The same information stored in hash tables.

m_allTheRules

FastVector[] m_allTheRules

The list of all generated rules.

m_instances

Instances m_instances

The instances (transactions) to be used for generating the association rules.

m_outputItemSets

boolean m_outputItemSets

Output itemsets found?

m_removeMissingCols

boolean m_removeMissingCols

Remove columns with all missing values

m_verbose

boolean m_verbose

Report progress iteratively

m_onlyClass

Instances m_onlyClass

Only the class attribute of all Instances.

m_classIndex

int m_classIndex

The class index.

m_car

boolean m_car

Flag indicating whether class association rules are mined.

m_treatZeroAsMissing

boolean m_treatZeroAsMissing

Treat zeros as missing (rather than a value in their own right)

Class weka.associations.AprioriItemSet extends ItemSet implements Serializable

serialVersionUID: 7684467755712672058L

Class weka.associations.CaRuleGeneration extends RuleGeneration implements Serializable

serialVersionUID: 3065752149646517703L

Class weka.associations.FilteredAssociator extends SingleAssociatorEnhancer implements Serializable

serialVersionUID: -4523450618538717400L

Serialized Fields

m_Filter

Filter m_Filter

The filter

m_FilteredInstances

Instances m_FilteredInstances

The instance structure of the filtered instances

m_ClassIndex

int m_ClassIndex

The class index.

Class weka.associations.GeneralizedSequentialPatterns extends AbstractAssociator implements Serializable

serialVersionUID: -4119691320812254676L

Serialized Fields

m_MinSupport

double m_MinSupport

the minimum support threshold

m_DataSeqID

int m_DataSeqID

number indicating the attribute holding the data sequence ID

m_OriginalDataSet

Instances m_OriginalDataSet

original sequential data set to be used for sequential patterns extraction

m_AllSequentialPatterns

FastVector m_AllSequentialPatterns

all generated frequent sequences, i.e. sequential patterns

m_Cycles

int m_Cycles

number of cycles performed until termination

m_CycleStart

java.lang.String m_CycleStart

String indicating the starting time of an cycle.

m_CycleEnd

java.lang.String m_CycleEnd

String indicating the ending time of an cycle.

m_AlgorithmStart

java.lang.String m_AlgorithmStart

String indicating the starting time of the algorithm.

m_FilterAttributes

java.lang.String m_FilterAttributes

String containing the attribute numbers that are used for result filtering; -1 means no filtering

m_FilterAttrVector

FastVector m_FilterAttrVector

Vector containing the attribute numbers that are used for result filtering; -1 means no filtering

m_Debug

boolean m_Debug

Whether the classifier is run in debug mode.

Class weka.associations.HotSpot extends java.lang.Object implements Serializable

serialVersionUID: 42972325096347677L

Serialized Fields

m_targetSI

SingleIndex m_targetSI

index of the target attribute

m_target

int m_target

m_support

double m_support

Support as a fraction of the total training set

m_supportCount

int m_supportCount

Support as an instance count

m_globalTarget

double m_globalTarget

The global value of the attribute of interest (mean or probability)

m_minImprovement

double m_minImprovement

The minimum improvement necessary to justify adding a test

m_globalSupport

int m_globalSupport

Actual global support of the target value (discrete target only)

m_targetIndexSI

SingleIndex m_targetIndexSI

For discrete target, the index of the value of interest

m_targetIndex

int m_targetIndex

m_maxBranchingFactor

int m_maxBranchingFactor

At each level of the tree consider at most this number extensions

m_numInstances

int m_numInstances

Number of instances in the full data

m_head

weka.associations.HotSpot.HotNode m_head

The head of the tree

m_header

Instances m_header

Header of the training data

m_lookups

int m_lookups

Debugging stuff

m_insertions

int m_insertions

m_hits

int m_hits

m_debug

boolean m_debug

m_minimize

boolean m_minimize

Minimize, rather than maximize the target

m_errorMessage

java.lang.String m_errorMessage

Error messages relating to too large/small support values

m_ruleLookup

java.util.HashMap<K,V> m_ruleLookup

Rule lookup table

Class weka.associations.HotSpot.HotNode extends java.lang.Object implements Serializable

Serialized Fields

m_insts

Instances m_insts

m_targetValue

double m_targetValue

m_children

weka.associations.HotSpot.HotNode[] m_children

m_testDetails

weka.associations.HotSpot.HotNode.HotTestDetails[] m_testDetails

m_id

int m_id

Class weka.associations.HotSpot.HotNode.HotTestDetails extends java.lang.Object implements Serializable

Serialized Fields

m_merit

double m_merit

m_support

int m_support

m_subsetSize

int m_subsetSize

m_splitAttIndex

int m_splitAttIndex

m_splitValue

double m_splitValue

m_lessThan

boolean m_lessThan

Class weka.associations.ItemSet extends java.lang.Object implements Serializable

serialVersionUID: 2724000045282835791L

Serialized Fields

m_items

int[] m_items

The items stored as an array of of ints.

m_counter

int m_counter

Counter for how many transactions contain this item set.

m_totalTransactions

int m_totalTransactions

The total number of transactions

m_treatZeroAsMissing

boolean m_treatZeroAsMissing

Treat zeros as missing (rather than a value in their own right)

Class weka.associations.LabeledItemSet extends ItemSet implements Serializable

serialVersionUID: 4158771925518299903L

Serialized Fields

m_classLabel

int m_classLabel

The class label.

m_ruleSupCounter

int m_ruleSupCounter

The support of the rule.

Class weka.associations.PredictiveApriori extends AbstractAssociator implements Serializable

serialVersionUID: 8109088846865075341L

Serialized Fields

m_premiseCount

int m_premiseCount

The minimum support.

m_numRules

int m_numRules

The maximum number of rules that are output.

m_Ls

FastVector m_Ls

The set of all sets of itemsets.

m_hashtables

FastVector m_hashtables

The same information stored in hash tables.

m_allTheRules

FastVector[] m_allTheRules

The list of all generated rules.

m_instances

Instances m_instances

The instances (transactions) to be used for generating the association rules.

m_priors

java.util.Hashtable<K,V> m_priors

The hashtable containing the prior probabilities.

m_midPoints

double[] m_midPoints

The mid points of the intervals used for the prior estimation.

m_expectation

double m_expectation

The expected predictive accuracy a rule needs to be a candidate for the output.

m_best

java.util.TreeSet<E> m_best

The n best rules.

m_bestChanged

boolean m_bestChanged

Flag keeping track if the list of the n best rules has changed.

m_count

int m_count

Counter for the time of generation for an association rule.

m_priorEstimator

PriorEstimation m_priorEstimator

The prior estimator.

m_classIndex

int m_classIndex

The class index.

m_car

boolean m_car

Flag indicating whether class association rules are mined.

Class weka.associations.PriorEstimation extends java.lang.Object implements Serializable

serialVersionUID: 5570863216522496271L

Serialized Fields

m_numRandRules

int m_numRandRules

The number of rnadom rules.

m_numIntervals

int m_numIntervals

The number of intervals.

m_randNum

java.util.Random m_randNum

The random number generator.

m_instances

Instances m_instances

The instances for which association rules are mined.

m_CARs

boolean m_CARs

Flag indicating whether standard association rules or class association rules are mined.

m_distribution

java.util.Hashtable<K,V> m_distribution

Hashtable to store the confidence values of randomly generated rules.

m_priors

java.util.Hashtable<K,V> m_priors

Hashtable containing the estimated prior probabilities.

m_sum

double m_sum

Sums up the confidences of all rules with a certain length.

m_midPoints

double[] m_midPoints

The mid points of the discrete intervals in which the interval [0,1] is divided.

Class weka.associations.RuleGeneration extends java.lang.Object implements Serializable

serialVersionUID: -8927041669872491432L

Serialized Fields

m_items

int[] m_items

The items stored as an array of of integer.

m_counter

int m_counter

Counter for how many transactions contain this item set.

m_totalTransactions

int m_totalTransactions

The total number of transactions

m_change

boolean m_change

Flag indicating whether the list fo the best rules has changed.

m_expectation

double m_expectation

The minimum expected predictive accuracy that is needed to be a candidate for the list of the best rules.

m_minRuleCount

int m_minRuleCount

The minimum support a rule needs to be a candidate for the list of the best rules.

m_midPoints

double[] m_midPoints

Sorted array of the mied points of the intervals used for prior estimation.

m_priors

java.util.Hashtable<K,V> m_priors

Hashtable conatining the estimated prior probabilities.

m_best

java.util.TreeSet<E> m_best

The list of the actual n best rules.

m_count

int m_count

Integer indicating the generation time of a rule.

m_instances

Instances m_instances

The instances.

Class weka.associations.RuleItem extends java.lang.Object implements Serializable

serialVersionUID: -3761299128347476534L

Serialized Fields

m_premise

ItemSet m_premise

The premise of a rule.

m_consequence

ItemSet m_consequence

The consequence of a rule.

m_accuracy

double m_accuracy

The expected predictive accuracy of a rule.

m_genTime

int m_genTime

The generation time of a rule.

Class weka.associations.SingleAssociatorEnhancer extends AbstractAssociator implements Serializable

serialVersionUID: -3665885256363525164L

Serialized Fields

m_Associator

Associator m_Associator

The base associator to use

Class weka.associations.Tertius extends AbstractAssociator implements Serializable

serialVersionUID: 5556726848380738179L

Serialized Fields

m_results

SimpleLinkedList m_results

The results.

m_hypotheses

int m_hypotheses

Number of hypotheses considered.

m_explored

int m_explored

Number of hypotheses explored.

m_time

java.util.Date m_time

Time needed for the search.

m_valuesText

java.awt.TextField m_valuesText

Field to output the current values.

m_instances

Instances m_instances

Instances used for the search.

m_predicates

java.util.ArrayList<E> m_predicates

Predicates used in the rules.

m_status

int m_status

Status of the search.

m_best

int m_best

Number of best confirmation values to search.

m_frequencyThreshold

double m_frequencyThreshold

Frequency threshold for the body and the negation of the head.

m_confirmationThreshold

double m_confirmationThreshold

Confirmation threshold for the rules.

m_noiseThreshold

double m_noiseThreshold

Maximal number of counter-instances.

m_repeat

boolean m_repeat

Repeat attributes ?

m_numLiterals

int m_numLiterals

Number of literals in a rule.

m_negation

int m_negation

Type of negation used in the rules.

m_classification

boolean m_classification

Classification bias.

m_classIndex

int m_classIndex

Index of class attribute.

m_horn

boolean m_horn

Horn clauses bias.

m_equivalent

boolean m_equivalent

Perform test on equivalent rules ?

m_sameClause

boolean m_sameClause

Perform test on same clauses ?

m_subsumption

boolean m_subsumption

Perform subsumption test ?

m_missing

int m_missing

Way of handling missing values in the search.

m_roc

boolean m_roc

Perform ROC analysis ?

m_partsString

java.lang.String m_partsString

Name of the file containing the parts for individual-based learning.

m_parts

Instances m_parts

Part instances for individual-based learning.

m_printValues

int m_printValues

Type of values output.

Package weka.associations.gsp

Class weka.associations.gsp.Element extends java.lang.Object implements Serializable

serialVersionUID: -7900701276019516371L

Serialized Fields

m_Events

int[] m_Events

events/items stored as an array of ints

Class weka.associations.gsp.Sequence extends java.lang.Object implements Serializable

serialVersionUID: -5001018056339156390L

Serialized Fields

m_SupportCount

int m_SupportCount

the support count of the Sequence

m_Elements

FastVector m_Elements

ordered list of the comprised elements/itemsets

Package weka.associations.tertius

Class weka.associations.tertius.AttributeValueLiteral extends Literal implements Serializable

serialVersionUID: 4077436297281456239L

Serialized Fields

m_value

java.lang.String m_value

m_index

int m_index

Class weka.associations.tertius.Body extends LiteralSet implements Serializable

serialVersionUID: 4870689270432218016L

Class weka.associations.tertius.Head extends LiteralSet implements Serializable

serialVersionUID: 5068076274253706199L

Class weka.associations.tertius.IndividualInstance extends Instance implements Serializable

serialVersionUID: -7903938733476585114L

Serialized Fields

m_parts

Instances m_parts

Class weka.associations.tertius.IndividualInstances extends Instances implements Serializable

serialVersionUID: -7355054814895636733L

Class weka.associations.tertius.IndividualLiteral extends AttributeValueLiteral implements Serializable

serialVersionUID: 4712404824517887435L

Serialized Fields

m_type

int m_type

Class weka.associations.tertius.Literal extends java.lang.Object implements Serializable

serialVersionUID: 2675363669503575771L

Serialized Fields

m_predicate

Predicate m_predicate

m_sign

int m_sign

m_negation

Literal m_negation

m_missing

int m_missing

Class weka.associations.tertius.LiteralSet extends java.lang.Object implements Serializable

serialVersionUID: 6094536488654503152L

Serialized Fields

m_literals

java.util.ArrayList<E> m_literals

Literals contained in this set.

m_lastLiteral

Literal m_lastLiteral

Last literal added to this set.

m_numInstances

int m_numInstances

Number of instances in the data the set deals with.

m_counterInstances

java.util.ArrayList<E> m_counterInstances

Set of counter-instances of this part of the rule.

m_counter

int m_counter

Counter for the number of counter-instances.

m_type

int m_type

Type of properties expressed in this set (individual or parts properties).

Class weka.associations.tertius.Predicate extends java.lang.Object implements Serializable

serialVersionUID: -8374702481965026640L

Serialized Fields

m_literals

java.util.ArrayList<E> m_literals

m_name

java.lang.String m_name

m_index

int m_index

m_isClass

boolean m_isClass

Class weka.associations.tertius.Rule extends java.lang.Object implements Serializable

serialVersionUID: -7763378359090435505L

Serialized Fields

m_body

Body m_body

The body of the rule.

m_head

Head m_head

The head of the rule.

m_repeatPredicate

boolean m_repeatPredicate

Can repeat predicates in the rule ?

m_maxLiterals

int m_maxLiterals

Maximal number of literals in the rule.

m_negBody

boolean m_negBody

Can there be negations in the body ?

m_negHead

boolean m_negHead

Can there be negations in the head ?

m_classRule

boolean m_classRule

Is this rule a classification rule ?

m_singleHead

boolean m_singleHead

Can there be only one literal in the head ?

m_numInstances

int m_numInstances

Number of instances in the data this rule deals with.

m_counterInstances

java.util.ArrayList<E> m_counterInstances

Set of counter-instances of this rule.

m_counter

int m_counter

Counter for the counter-instances of this rule.

m_confirmation

double m_confirmation

Confirmation of this rule.

m_optimistic

double m_optimistic

Optimistic estimate of this rule.

Class weka.associations.tertius.SimpleLinkedList extends java.lang.Object implements Serializable

serialVersionUID: -1491148276509976299L

Serialization Methods

readObject

private void readObject(java.io.ObjectInputStream s)
                 throws java.io.IOException,
                        java.lang.ClassNotFoundException

Reconstitute this LinkedList instance from a stream (that is deserialize it).

Throws:

java.io.IOException

java.lang.ClassNotFoundException

writeObject

private void writeObject(java.io.ObjectOutputStream s)
                  throws java.io.IOException

Save the state of this LinkedList instance to a stream (that is, serialize it).

Serial Data:

The size of the list (the number of elements it contains) is emitted (int), followed by all of its elements (each an Object) in the proper order.

Throws:

java.io.IOException

Serialized Fields

first

weka.associations.tertius.SimpleLinkedList.Entry first

last

weka.associations.tertius.SimpleLinkedList.Entry last

Class weka.associations.tertius.SimpleLinkedList.LinkedListInverseIterator extends java.lang.Object implements Serializable

serialVersionUID: 6290379064027832108L

Serialized Fields

current

weka.associations.tertius.SimpleLinkedList.Entry current

lastReturned

weka.associations.tertius.SimpleLinkedList.Entry lastReturned

Class weka.associations.tertius.SimpleLinkedList.LinkedListIterator extends java.lang.Object implements Serializable

serialVersionUID: -2448555236100426759L

Serialized Fields

current

weka.associations.tertius.SimpleLinkedList.Entry current

lastReturned

weka.associations.tertius.SimpleLinkedList.Entry lastReturned

Package weka.attributeSelection

Class weka.attributeSelection.ASEvaluation extends java.lang.Object implements Serializable

serialVersionUID: 2091705669885950849L

Class weka.attributeSelection.ASSearch extends java.lang.Object implements Serializable

serialVersionUID: 7591673350342236548L

Class weka.attributeSelection.AttributeSelection extends java.lang.Object implements Serializable

serialVersionUID: 4170171824147584330L

Serialized Fields

m_trainInstances

Instances m_trainInstances

the instances to select attributes from

m_ASEvaluator

ASEvaluation m_ASEvaluator

the attribute/subset evaluator

m_searchMethod

ASSearch m_searchMethod

the search method

m_numFolds

int m_numFolds

the number of folds to use for cross validation

m_selectionResults

java.lang.StringBuffer m_selectionResults

holds a string describing the results of the attribute selection

m_doRank

boolean m_doRank

rank features (if allowed by the search method)

m_doXval

boolean m_doXval

do cross validation

m_seed

int m_seed

seed used to randomly shuffle instances for cross validation

m_numToSelect

int m_numToSelect

number of attributes requested from ranked results

m_selectedAttributeSet

int[] m_selectedAttributeSet

the selected attributes

m_attributeRanking

double[][] m_attributeRanking

the attribute indexes and associated merits if a ranking is produced

m_transformer

AttributeTransformer m_transformer

if a feature selection run involves an attribute transformer

m_attributeFilter

Remove m_attributeFilter

the attribute filter for processing instances with respect to the most recent feature selection run

m_rankResults

double[][] m_rankResults

hold statistics for repeated feature selection, such as under cross validation

m_subsetResults

double[] m_subsetResults

m_trials

int m_trials

Class weka.attributeSelection.AttributeSetEvaluator extends ASEvaluation implements Serializable

serialVersionUID: -5744881009422257389L

Class weka.attributeSelection.BestFirst extends ASSearch implements Serializable

serialVersionUID: 7841338689536821867L

Serialized Fields

m_maxStale

int m_maxStale

maximum number of stale nodes before terminating search

m_searchDirection

int m_searchDirection

0 == backward search, 1 == forward search, 2 == bidirectional

m_starting

int[] m_starting

holds an array of starting attributes

m_startRange

Range m_startRange

holds the start set for the search as a Range

m_hasClass

boolean m_hasClass

does the data have a class

m_classIndex

int m_classIndex

holds the class index

m_numAttribs

int m_numAttribs

number of attributes in the data

m_totalEvals

int m_totalEvals

total number of subsets evaluated during a search

m_debug

boolean m_debug

for debugging

m_bestMerit

double m_bestMerit

holds the merit of the best subset found

m_cacheSize

int m_cacheSize

holds the maximum size of the lookup cache for evaluated subsets

Class weka.attributeSelection.BestFirst.Link2 extends java.lang.Object implements Serializable

serialVersionUID: -8236598311516351420L

Serialized Fields

m_data

java.lang.Object[] m_data

m_merit

double m_merit

Class weka.attributeSelection.BestFirst.LinkedList2 extends FastVector implements Serializable

serialVersionUID: 3250538292330398929L

Serialized Fields

m_MaxSize

int m_MaxSize

Max number of elements in the list

Class weka.attributeSelection.CfsSubsetEval extends ASEvaluation implements Serializable

serialVersionUID: 747878400813276317L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The training instances

m_disTransform

Discretize m_disTransform

Discretise attributes when class in nominal

m_classIndex

int m_classIndex

The class index

m_isNumeric

boolean m_isNumeric

Is the class numeric

m_numAttribs

int m_numAttribs

Number of attributes in the training data

m_numInstances

int m_numInstances

Number of instances in the training data

m_missingSeperate

boolean m_missingSeperate

Treat missing values as seperate values

m_locallyPredictive

boolean m_locallyPredictive

Include locally predicitive attributes

m_corr_matrix

float[][] m_corr_matrix

Holds the matrix of attribute correlations

m_std_devs

double[] m_std_devs

Standard deviations of attributes (when using pearsons correlation)

m_c_Threshold

double m_c_Threshold

Threshold for admitting locally predictive features

Class weka.attributeSelection.ChiSquaredAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: -8316857822521717692L

Serialized Fields

m_missing_merge

boolean m_missing_merge

Treat missing values as a seperate value

m_Binarize

boolean m_Binarize

Just binarize numeric attributes

m_ChiSquareds

double[] m_ChiSquareds

The chi-squared value for each attribute

Class weka.attributeSelection.ClassifierSubsetEval extends HoldOutSubsetEvaluator implements Serializable

serialVersionUID: 7532217899385278710L

Serialized Fields

m_trainingInstances

Instances m_trainingInstances

training instances

m_classIndex

int m_classIndex

class index

m_numAttribs

int m_numAttribs

number of attributes in the training data

m_numInstances

int m_numInstances

number of training instances

m_Classifier

Classifier m_Classifier

holds the classifier to use for error estimates

m_Evaluation

Evaluation m_Evaluation

holds the evaluation object to use for evaluating the classifier

m_holdOutFile

java.io.File m_holdOutFile

the file that containts hold out/test instances

m_holdOutInstances

Instances m_holdOutInstances

the instances to test on

m_useTraining

boolean m_useTraining

evaluate on training data rather than seperate hold out/test set

Class weka.attributeSelection.ConsistencySubsetEval extends ASEvaluation implements Serializable

serialVersionUID: -2880323763295270402L

Serialized Fields

m_trainInstances

Instances m_trainInstances

training instances

m_classIndex

int m_classIndex

class index

m_numAttribs

int m_numAttribs

number of attributes in the training data

m_numInstances

int m_numInstances

number of instances in the training data

m_disTransform

Discretize m_disTransform

Discretise numeric attributes

m_table

java.util.Hashtable<K,V> m_table

Hash table for evaluating feature subsets

Class weka.attributeSelection.ConsistencySubsetEval.hashKey extends java.lang.Object implements Serializable

serialVersionUID: 6144138512017017408L

Serialized Fields

attributes

double[] attributes

Array of attribute values for an instance

missing

boolean[] missing

True for an index if the corresponding attribute value is missing.

key

int key

The key

Class weka.attributeSelection.CostSensitiveASEvaluation extends ASEvaluation implements Serializable

serialVersionUID: -7045833833363396977L

Serialized Fields

m_MatrixSource

int m_MatrixSource

Indicates the current cost matrix source

m_OnDemandDirectory

java.io.File m_OnDemandDirectory

The directory used when loading cost files on demand, null indicates current directory

m_CostFile

java.lang.String m_CostFile

The name of the cost file, for command line options

m_CostMatrix

CostMatrix m_CostMatrix

The cost matrix

m_evaluator

ASEvaluation m_evaluator

The base evaluator to use

m_seed

int m_seed

random number seed

Class weka.attributeSelection.CostSensitiveAttributeEval extends CostSensitiveASEvaluation implements Serializable

serialVersionUID: 4484876541145458447L

Class weka.attributeSelection.CostSensitiveSubsetEval extends CostSensitiveASEvaluation implements Serializable

serialVersionUID: 2924546096103426700L

Class weka.attributeSelection.ExhaustiveSearch extends ASSearch implements Serializable

serialVersionUID: 5741842861142379712L

Serialized Fields

m_bestGroup

java.util.BitSet m_bestGroup

the best feature set found during the search

m_bestMerit

double m_bestMerit

the merit of the best subset found

m_hasClass

boolean m_hasClass

does the data have a class

m_classIndex

int m_classIndex

holds the class index

m_numAttribs

int m_numAttribs

number of attributes in the data

m_verbose

boolean m_verbose

if true, then ouput new best subsets as the search progresses

m_evaluations

int m_evaluations

the number of subsets evaluated during the search

Class weka.attributeSelection.FCBFSearch extends ASSearch implements Serializable

serialVersionUID: 8209699587428369942L

Serialized Fields

m_starting

int[] m_starting

Holds the starting set as an array of attributes

m_startRange

Range m_startRange

Holds the start set for the search as a range

m_attributeList

int[] m_attributeList

Holds the ordered list of attributes

m_attributeMerit

double[] m_attributeMerit

Holds the list of attribute merit scores

m_hasClass

boolean m_hasClass

Data has class attribute---if unsupervised evaluator then no class

m_classIndex

int m_classIndex

Class index of the data if supervised evaluator

m_numAttribs

int m_numAttribs

The number of attribtes

m_threshold

double m_threshold

A threshold by which to discard attributes---used by the AttributeSelection module

m_numToSelect

int m_numToSelect

The number of attributes to select. -1 indicates that all attributes are to be retained. Has precedence over m_threshold

m_calculatedNumToSelect

int m_calculatedNumToSelect

Used to compute the number to select

m_generateOutput

boolean m_generateOutput

Used to determine whether we create a new dataset according to the selected features

m_asEval

ASEvaluation m_asEval

Used to store the ref of the Evaluator we use

m_rankedFCBF

double[][] m_rankedFCBF

Holds the list of attribute merit scores generated by FCBF

m_selectedFeatures

double[][] m_selectedFeatures

Hold the list of selected features

Class weka.attributeSelection.FilteredAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: 2111121880778327334L

Serialized Fields

m_evaluator

AttributeEvaluator m_evaluator

Base evaluator

m_filter

Filter m_filter

Filter

m_filteredInstances

Instances m_filteredInstances

Filtered instances structure

Class weka.attributeSelection.FilteredSubsetEval extends ASEvaluation implements Serializable

serialVersionUID: 2111121880778327334L

Serialized Fields

m_evaluator

SubsetEvaluator m_evaluator

Base evaluator

m_filter

Filter m_filter

Filter

m_filteredInstances

Instances m_filteredInstances

Filtered instances structure

Class weka.attributeSelection.GainRatioAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: -8504656625598579926L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The training instances

m_classIndex

int m_classIndex

The class index

m_numAttribs

int m_numAttribs

The number of attributes

m_numInstances

int m_numInstances

The number of instances

m_numClasses

int m_numClasses

The number of classes

m_missing_merge

boolean m_missing_merge

Merge missing values

Class weka.attributeSelection.GeneticSearch extends ASSearch implements Serializable

serialVersionUID: -1618264232838472679L

Serialized Fields

m_starting

int[] m_starting

holds a starting set as an array of attributes. Becomes one member of the initial random population

m_startRange

Range m_startRange

holds the start set for the search as a Range

m_hasClass

boolean m_hasClass

does the data have a class

m_classIndex

int m_classIndex

holds the class index

m_numAttribs

int m_numAttribs

number of attributes in the data

m_population

weka.attributeSelection.GeneticSearch.GABitSet[] m_population

the current population

m_popSize

int m_popSize

the number of individual solutions

m_best

weka.attributeSelection.GeneticSearch.GABitSet m_best

the best population member found during the search

m_bestFeatureCount

int m_bestFeatureCount

the number of features in the best population member

m_lookupTableSize

int m_lookupTableSize

the number of entries to cache for lookup

m_lookupTable

java.util.Hashtable<K,V> m_lookupTable

the lookup table

m_random

java.util.Random m_random

random number generation

m_seed

int m_seed

seed for random number generation

m_pCrossover

double m_pCrossover

the probability of crossover occuring

m_pMutation

double m_pMutation

the probability of mutation occuring

m_sumFitness

double m_sumFitness

sum of the current population fitness

m_maxFitness

double m_maxFitness

m_minFitness

double m_minFitness

m_avgFitness

double m_avgFitness

m_maxGenerations

int m_maxGenerations

the maximum number of generations to evaluate

m_reportFrequency

int m_reportFrequency

how often reports are generated

m_generationReports

java.lang.StringBuffer m_generationReports

holds the generation reports

Class weka.attributeSelection.GeneticSearch.GABitSet extends java.lang.Object implements Serializable

serialVersionUID: -2930607837482622224L

Serialized Fields

m_chromosome

java.util.BitSet m_chromosome

the bitset

m_objective

double m_objective

holds raw merit

m_fitness

double m_fitness

the fitness

Class weka.attributeSelection.GreedyStepwise extends ASSearch implements Serializable

serialVersionUID: -6312951970168325471L

Serialized Fields

m_hasClass

boolean m_hasClass

does the data have a class

m_classIndex

int m_classIndex

holds the class index

m_numAttribs

int m_numAttribs

number of attributes in the data

m_rankingRequested

boolean m_rankingRequested

true if the user has requested a ranked list of attributes

m_doRank

boolean m_doRank

go from one side of the search space to the other in order to generate a ranking

m_doneRanking

boolean m_doneRanking

used to indicate whether or not ranking has been performed

m_threshold

double m_threshold

A threshold by which to discard attributes---used by the AttributeSelection module

m_numToSelect

int m_numToSelect

The number of attributes to select. -1 indicates that all attributes are to be retained. Has precedence over m_threshold

m_calculatedNumToSelect

int m_calculatedNumToSelect

m_bestMerit

double m_bestMerit

the merit of the best subset found

m_rankedAtts

double[][] m_rankedAtts

a ranked list of attribute indexes

m_rankedSoFar

int m_rankedSoFar

m_best_group

java.util.BitSet m_best_group

the best subset found

m_ASEval

ASEvaluation m_ASEval

m_Instances

Instances m_Instances

m_startRange

Range m_startRange

holds the start set for the search as a Range

m_starting

int[] m_starting

holds an array of starting attributes

m_backward

boolean m_backward

Use a backwards search instead of a forwards one

m_conservativeSelection

boolean m_conservativeSelection

If set then attributes will continue to be added during a forward search as long as the merit does not degrade

Class weka.attributeSelection.HoldOutSubsetEvaluator extends ASEvaluation implements Serializable

serialVersionUID: 8280529785412054174L

Class weka.attributeSelection.InfoGainAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: -1949849512589218930L

Serialized Fields

m_missing_merge

boolean m_missing_merge

Treat missing values as a seperate value

m_Binarize

boolean m_Binarize

Just binarize numeric attributes

m_InfoGains

double[] m_InfoGains

The info gain for each attribute

Class weka.attributeSelection.LatentSemanticAnalysis extends UnsupervisedAttributeEvaluator implements Serializable

serialVersionUID: -8712112988018106198L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The data to transform analyse/transform

m_trainHeader

Instances m_trainHeader

Keep a copy for the class attribute (if set)

m_transformedFormat

Instances m_transformedFormat

The header for the transformed data format

m_hasClass

boolean m_hasClass

Data has a class set

m_classIndex

int m_classIndex

Class index

m_numAttributes

int m_numAttributes

Number of attributes

m_numInstances

int m_numInstances

Number of instances

m_transpose

boolean m_transpose

Is transpose necessary because numAttributes < numInstances?

m_u

Matrix m_u

Will hold the left singular vectors

m_s

Matrix m_s

Will hold the singular values

m_v

Matrix m_v

Will hold the right singular values

m_transformationMatrix

Matrix m_transformationMatrix

Will hold the matrix used to transform instances to the new feature space

m_replaceMissingFilter

ReplaceMissingValues m_replaceMissingFilter

Filters for original data

m_normalizeFilter

Normalize m_normalizeFilter

m_nominalToBinaryFilter

NominalToBinary m_nominalToBinaryFilter

m_attributeFilter

Remove m_attributeFilter

m_outputNumAttributes

int m_outputNumAttributes

The number of attributes in the LSA transformed data

m_normalize

boolean m_normalize

Normalize the input data?

m_rank

double m_rank

The approximation rank to use (between 0 and 1 means coverage proportion)

m_sumSquaredSingularValues

double m_sumSquaredSingularValues

The sum of the squares of the singular values

m_actualRank

int m_actualRank

The actual rank number to use for computation

m_maxAttributesInName

int m_maxAttributesInName

Maximum number of attributes in the transformed attribute name

Class weka.attributeSelection.LFSMethods.Link2 extends java.lang.Object implements Serializable

serialVersionUID: -7422719407475185086L

Serialized Fields

m_data

java.lang.Object[] m_data

m_merit

double m_merit

Class weka.attributeSelection.LFSMethods.LinkedList2 extends FastVector implements Serializable

serialVersionUID: -7776010892419656105L

Serialized Fields

m_MaxSize

int m_MaxSize

Class weka.attributeSelection.LinearForwardSelection extends ASSearch implements Serializable

Serialized Fields

m_maxStale

int m_maxStale

maximum number of stale nodes before terminating search

m_forwardSearchMethod

int m_forwardSearchMethod

0 == forward selection, 1 == floating forward search

m_performRanking

boolean m_performRanking

perform initial ranking to select top-ranked attributes

m_numUsedAttributes

int m_numUsedAttributes

number of top-ranked attributes that are taken into account for the search

m_linearSelectionType

int m_linearSelectionType

0 == fixed-set, 1 == fixed-width

m_starting

int[] m_starting

holds an array of starting attributes

m_startRange

Range m_startRange

holds the start set for the search as a Range

m_hasClass

boolean m_hasClass

does the data have a class

m_classIndex

int m_classIndex

holds the class index

m_numAttribs

int m_numAttribs

number of attributes in the data

m_totalEvals

int m_totalEvals

total number of subsets evaluated during a search

m_verbose

boolean m_verbose

for debugging

m_bestMerit

double m_bestMerit

holds the merit of the best subset found

m_cacheSize

int m_cacheSize

holds the maximum size of the lookup cache for evaluated subsets

Class weka.attributeSelection.OneRAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: 4386514823886856980L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The training instances

m_classIndex

int m_classIndex

The class index

m_numAttribs

int m_numAttribs

The number of attributes

m_numInstances

int m_numInstances

The number of instances

m_randomSeed

int m_randomSeed

Random number seed

m_folds

int m_folds

Number of folds for cross validation

m_evalUsingTrainingData

boolean m_evalUsingTrainingData

Use training data to evaluate merit rather than x-val

m_minBucketSize

int m_minBucketSize

Passed on to OneR

Class weka.attributeSelection.PrincipalComponents extends UnsupervisedAttributeEvaluator implements Serializable

serialVersionUID: 3310137541055815078L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The data to transform analyse/transform

m_trainHeader

Instances m_trainHeader

Keep a copy for the class attribute (if set)

m_transformedFormat

Instances m_transformedFormat

The header for the transformed data format

m_originalSpaceFormat

Instances m_originalSpaceFormat

The header for data transformed back to the original space

m_hasClass

boolean m_hasClass

Data has a class set

m_classIndex

int m_classIndex

Class index

m_numAttribs

int m_numAttribs

Number of attributes

m_numInstances

int m_numInstances

Number of instances

m_correlation

double[][] m_correlation

Correlation matrix for the original data

m_eigenvectors

double[][] m_eigenvectors

Will hold the unordered linear transformations of the (normalized) original data

m_eigenvalues

double[] m_eigenvalues

Eigenvalues for the corresponding eigenvectors

m_sortedEigens

int[] m_sortedEigens

Sorted eigenvalues

m_sumOfEigenValues

double m_sumOfEigenValues

sum of the eigenvalues

m_replaceMissingFilter

ReplaceMissingValues m_replaceMissingFilter

Filters for original data

m_normalizeFilter

Normalize m_normalizeFilter

m_nominalToBinFilter

NominalToBinary m_nominalToBinFilter

m_attributeFilter

Remove m_attributeFilter

m_attribFilter

Remove m_attribFilter

used to remove the class column if a class column is set

m_outputNumAtts

int m_outputNumAtts

The number of attributes in the pc transformed data

m_normalize

boolean m_normalize

normalize the input data?

m_coverVariance

double m_coverVariance

the amount of varaince to cover in the original data when retaining the best n PC's

m_transBackToOriginal

boolean m_transBackToOriginal

transform the data through the pc space and back to the original space ?

m_maxAttrsInName

int m_maxAttrsInName

maximum number of attributes in the transformed attribute name

m_eTranspose

double[][] m_eTranspose

holds the transposed eigenvectors for converting back to the original space

Class weka.attributeSelection.RaceSearch extends ASSearch implements Serializable

serialVersionUID: 4015453851212985720L

Serialized Fields

m_Instances

Instances m_Instances

the training instances

m_raceType

int m_raceType

the selected search type

m_xvalType

int m_xvalType

the selected xval type

m_classIndex

int m_classIndex

the class index

m_numAttribs

int m_numAttribs

the number of attributes in the data

m_totalEvals

int m_totalEvals

the total number of partially/fully evaluated subsets

m_bestMerit

double m_bestMerit

holds the merit of the best subset found

m_theEvaluator

HoldOutSubsetEvaluator m_theEvaluator

the subset evaluator to use

m_sigLevel

double m_sigLevel

the significance level for comparisons

m_delta

double m_delta

threshold for comparisons

m_samples

int m_samples

the number of samples above which to begin testing for similarity between competing subsets

m_numFolds

int m_numFolds

number of cross validation folds---equal to the number of instances for leave-one-out cv

m_ASEval

ASEvaluation m_ASEval

the attribute evaluator to generate the initial ranking when doing a rank race

m_Ranking

int[] m_Ranking

will hold the attribute ranking produced by the above attribute evaluator if doing a rank search

m_debug

boolean m_debug

verbose output for monitoring the search and debugging

m_rankingRequested

boolean m_rankingRequested

If true then produce a ranked list of attributes by fully traversing a forward hillclimb race

m_rankedAtts

double[][] m_rankedAtts

The ranked list of attributes produced if m_rankingRequested is true

m_rankedSoFar

int m_rankedSoFar

The number of attributes ranked so far (if ranking is requested)

m_numToSelect

int m_numToSelect

The number of attributes to retain if a ranking is requested. -1 indicates that all attributes are to be retained. Has precedence over m_threshold

m_calculatedNumToSelect

int m_calculatedNumToSelect

m_threshold

double m_threshold

the threshold for removing attributes if ranking is requested

Class weka.attributeSelection.RandomSearch extends ASSearch implements Serializable

serialVersionUID: 7479392617377425484L

Serialized Fields

m_starting

int[] m_starting

holds a starting set as an array of attributes.

m_startRange

Range m_startRange

holds the start set as a range

m_bestGroup

java.util.BitSet m_bestGroup

the best feature set found during the search

m_bestMerit

double m_bestMerit

the merit of the best subset found

m_onlyConsiderBetterAndSmaller

boolean m_onlyConsiderBetterAndSmaller

only accept a feature set as being "better" than the best if its merit is better or equal to the best, and it contains fewer features than the best (this allows LVF to be implimented).

m_hasClass

boolean m_hasClass

does the data have a class

m_classIndex

int m_classIndex

holds the class index

m_numAttribs

int m_numAttribs

number of attributes in the data

m_seed

int m_seed

seed for random number generation

m_searchSize

double m_searchSize

percentage of the search space to consider

m_iterations

int m_iterations

the number of iterations performed

m_random

java.util.Random m_random

random number object

m_verbose

boolean m_verbose

output new best subsets as the search progresses

Class weka.attributeSelection.Ranker extends ASSearch implements Serializable

serialVersionUID: -9086714848510751934L

Serialized Fields

m_starting

int[] m_starting

Holds the starting set as an array of attributes

m_startRange

Range m_startRange

Holds the start set for the search as a range

m_attributeList

int[] m_attributeList

Holds the ordered list of attributes

m_attributeMerit

double[] m_attributeMerit

Holds the list of attribute merit scores

m_hasClass

boolean m_hasClass

Data has class attribute---if unsupervised evaluator then no class

m_classIndex

int m_classIndex

Class index of the data if supervised evaluator

m_numAttribs

int m_numAttribs

The number of attribtes

m_threshold

double m_threshold

A threshold by which to discard attributes---used by the AttributeSelection module

m_numToSelect

int m_numToSelect

The number of attributes to select. -1 indicates that all attributes are to be retained. Has precedence over m_threshold

m_calculatedNumToSelect

int m_calculatedNumToSelect

Used to compute the number to select

Class weka.attributeSelection.RankSearch extends ASSearch implements Serializable

serialVersionUID: -7992268736874353755L

Serialized Fields

m_hasClass

boolean m_hasClass

does the data have a class

m_classIndex

int m_classIndex

holds the class index

m_numAttribs

int m_numAttribs

number of attributes in the data

m_best_group

java.util.BitSet m_best_group

the best subset found

m_ASEval

ASEvaluation m_ASEval

the attribute evaluator to use for generating the ranking

m_SubsetEval

ASEvaluation m_SubsetEval

the subset evaluator with which to evaluate the ranking

m_Instances

Instances m_Instances

the training instances

m_bestMerit

double m_bestMerit

the merit of the best subset found

m_Ranking

int[] m_Ranking

will hold the attribute ranking

m_add

int m_add

add this many attributes in each iteration from the ranking

m_startPoint

int m_startPoint

start from this point in the ranking

Class weka.attributeSelection.ReliefFAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: -8422186665795839379L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The training instances

m_classIndex

int m_classIndex

The class index

m_numAttribs

int m_numAttribs

The number of attributes

m_numInstances

int m_numInstances

The number of instances

m_numericClass

boolean m_numericClass

Numeric class

m_numClasses

int m_numClasses

The number of classes if class is nominal

m_ndc

double m_ndc

Used to hold the probability of a different class val given nearest instances (numeric class)

m_nda

double[] m_nda

Used to hold the prob of different value of an attribute given nearest instances (numeric class case)

m_ndcda

double[] m_ndcda

Used to hold the prob of a different class val and different att val given nearest instances (numeric class case)

m_weights

double[] m_weights

Holds the weights that relief assigns to attributes

m_classProbs

double[] m_classProbs

Prior class probabilities (discrete class case)

m_sampleM

int m_sampleM

The number of instances to sample when estimating attributes default == -1, use all instances

m_Knn

int m_Knn

The number of nearest hits/misses

m_karray

double[][][] m_karray

k nearest scores + instance indexes for n classes

m_maxArray

double[] m_maxArray

Upper bound for numeric attributes

m_minArray

double[] m_minArray

Lower bound for numeric attributes

m_worst

double[] m_worst

Keep track of the farthest instance for each class

m_index

int[] m_index

Index in the m_karray of the farthest instance for each class

m_stored

int[] m_stored

Number of nearest neighbours stored of each class

m_seed

int m_seed

Random number seed used for sampling instances

m_weightsByRank

double[] m_weightsByRank

used to (optionally) weight nearest neighbours by their distance from the instance in question. Each entry holds exp(-((rank(r_i, i_j)/sigma)^2)) where rank(r_i,i_j) is the rank of instance i_j in a sequence of instances ordered by the distance from r_i. sigma is a user defined parameter, default=20

m_sigma

int m_sigma

m_weightByDistance

boolean m_weightByDistance

Weight by distance rather than equal weights

Class weka.attributeSelection.ScatterSearchV1 extends ASSearch implements Serializable

serialVersionUID: -8512041420388121326L

Serialized Fields

m_numAttribs

int m_numAttribs

number of attributes in the data

m_classIndex

int m_classIndex

holds the class index

m_treshold

double m_treshold

holds the treshhold that delimits the good attributes

m_initialThreshold

double m_initialThreshold

the initial threshold

m_typeOfCombination

int m_typeOfCombination

the kind of comination betwen parents ((0)greedy combination/(1)reduced greedy combination)

m_random

java.util.Random m_random

random number generation

m_seed

int m_seed

seed for random number generation

m_debug

boolean m_debug

verbose output for monitoring the search and debugging

m_InformationReports

java.lang.StringBuffer m_InformationReports

holds a report of the search

m_totalEvals

int m_totalEvals

total number of subsets evaluated during a search

m_bestMerit

double m_bestMerit

holds the merit of the best subset found

m_processinTime

long m_processinTime

time for procesing the search method

m_population

java.util.List<E> m_population

holds the Initial Population of Subsets

m_popSize

int m_popSize

holds the population size

m_initialPopSize

int m_initialPopSize

holds the user selected initial population size

m_calculatedInitialPopSize

int m_calculatedInitialPopSize

if no initial user pop size, then this holds the initial pop size calculated from the number of attributes in the data (for use in the toString() method)

m_attributeRanking

java.util.List<E> m_attributeRanking

holds the attributes ranked

ASEvaluator

SubsetEvaluator ASEvaluator

Evaluator used to know the significance of a subset (for guiding the search)

Class weka.attributeSelection.ScatterSearchV1.Subset extends java.lang.Object implements Serializable

Serialized Fields

merit

double merit

subset

java.util.BitSet subset

Class weka.attributeSelection.SignificanceAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: -8504656625598579926L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The training instances

m_classIndex

int m_classIndex

The class index

m_numAttribs

int m_numAttribs

The number of attributes

m_numInstances

int m_numInstances

The number of instances

m_numClasses

int m_numClasses

The number of classes

m_missing_merge

boolean m_missing_merge

Merge missing values

Class weka.attributeSelection.SubsetSizeForwardSelection extends ASSearch implements Serializable

Serialized Fields

m_performRanking

boolean m_performRanking

perform initial ranking to select top-ranked attributes

m_numUsedAttributes

int m_numUsedAttributes

number of top-ranked attributes that are taken into account for the search

m_linearSelectionType

int m_linearSelectionType

0 == fixed-set, 1 == fixed-width

m_setSizeEval

ASEvaluation m_setSizeEval

the subset evaluator to use for subset size determination

m_numFolds

int m_numFolds

Number of cross validation folds for subset size determination (default = 5).

m_seed

int m_seed

Seed for cross validation subset size determination. (default = 1)

m_numAttribs

int m_numAttribs

number of attributes in the data

m_totalEvals

int m_totalEvals

total number of subsets evaluated during a search

m_verbose

boolean m_verbose

for debugging

m_bestMerit

double m_bestMerit

holds the merit of the best subset found

m_cacheSize

int m_cacheSize

holds the maximum size of the lookup cache for evaluated subsets

Class weka.attributeSelection.SVMAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: -6489975709033967447L

Serialized Fields

m_attScores

double[] m_attScores

The attribute scores

m_numToEliminate

int m_numToEliminate

Constant rate of attribute elimination per iteration

m_percentToEliminate

int m_percentToEliminate

Percentage rate of attribute elimination, trumps constant rate (above threshold), ignored if = 0

m_percentThreshold

int m_percentThreshold

Threshold below which percent elimination switches to constant elimination

m_smoCParameter

double m_smoCParameter

Complexity parameter to pass on to SMO

m_smoTParameter

double m_smoTParameter

Tolerance parameter to pass on to SMO

m_smoPParameter

double m_smoPParameter

Epsilon parameter to pass on to SMO

m_smoFilterType

int m_smoFilterType

Filter parameter to pass on to SMO

Class weka.attributeSelection.SymmetricalUncertAttributeEval extends ASEvaluation implements Serializable

serialVersionUID: -8096505776132296416L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The training instances

m_classIndex

int m_classIndex

The class index

m_numAttribs

int m_numAttribs

The number of attributes

m_numInstances

int m_numInstances

The number of instances

m_numClasses

int m_numClasses

The number of classes

m_missing_merge

boolean m_missing_merge

Treat missing values as a seperate value

Class weka.attributeSelection.SymmetricalUncertAttributeSetEval extends AttributeSetEvaluator implements Serializable

serialVersionUID: 8351377335495873202L

Serialized Fields

m_trainInstances

Instances m_trainInstances

The training instances

m_classIndex

int m_classIndex

The class index

m_numAttribs

int m_numAttribs

The number of attributes

m_numInstances

int m_numInstances

The number of instances

m_numClasses

int m_numClasses

The number of classes

m_missing_merge

boolean m_missing_merge

Treat missing values as a seperate value

Class weka.attributeSelection.TabuSearch extends ASSearch implements Serializable

serialVersionUID: -8812132617585120414L

Serialized Fields

m_numAttribs

int m_numAttribs

number of attributes in the data

m_classIndex

int m_classIndex

holds the class index

m_random

java.util.Random m_random

random number generation

m_seed

int m_seed

seed for random number generation

m_diversificationProb

double m_diversificationProb

probabilities of diversification

m_numIterations

int m_numIterations

number of iterations for getting the best subset

m_totalEvals

int m_totalEvals

total number of subsets evaluated during a search

m_Sbest

TabuSearch.Subset m_Sbest

holds the best Subset found

m_numNeighborhood

int m_numNeighborhood

holds the number of neighborhood to generate from a Subset

m_processinTime

long m_processinTime

time for procesing the search method

m_initialSize

int m_initialSize

holds the solution size

m_initialSolution

TabuSearch.Subset m_initialSolution

Subset holding the initial solution

m_rankedAttribs

java.util.List<E> m_rankedAttribs

attribute ranking

m_vectorTabu

java.util.List<E> m_vectorTabu

tabu List

ASEvaluator

SubsetEvaluator ASEvaluator

Evaluator used to know the significance of a subset (for guiding the search)

Class weka.attributeSelection.TabuSearch.Subset extends java.lang.Object implements Serializable

Serialized Fields

merit

double merit

subset

java.util.BitSet subset

Class weka.attributeSelection.UnsupervisedAttributeEvaluator extends ASEvaluation implements Serializable

serialVersionUID: -4100897318675336178L

Class weka.attributeSelection.UnsupervisedSubsetEvaluator extends ASEvaluation implements Serializable

serialVersionUID: 627934376267488763L

Class weka.attributeSelection.WrapperSubsetEval extends ASEvaluation implements Serializable

serialVersionUID: -4573057658746728675L

Serialized Fields

m_trainInstances

Instances m_trainInstances

training instances

m_classIndex

int m_classIndex

class index

m_numAttribs

int m_numAttribs

number of attributes in the training data

m_numInstances

int m_numInstances

number of instances in the training data

m_Evaluation

Evaluation m_Evaluation

holds an evaluation object

m_BaseClassifier

Classifier m_BaseClassifier

holds the base classifier object

m_folds

int m_folds

number of folds to use for cross validation

m_seed

int m_seed

random number seed

m_threshold

double m_threshold

the threshold by which to do further cross validations when estimating the accuracy of a subset

m_evaluationMeasure

int m_evaluationMeasure

The evaluation measure to use

Package weka.classifiers

Class weka.classifiers.Classifier extends java.lang.Object implements Serializable

serialVersionUID: 6502780192411755341L

Serialized Fields

m_Debug

boolean m_Debug

Whether the classifier is run in debug mode.

Class weka.classifiers.CostMatrix extends java.lang.Object implements Serializable

serialVersionUID: -1973792250544554965L

Serialized Fields

m_size

int m_size

m_matrix

java.lang.Object[][] m_matrix

[rows][columns]

Class weka.classifiers.EnsembleLibrary extends java.lang.Object implements Serializable

serialVersionUID: -7987178904923706760L

Serialized Fields

m_Models

java.util.TreeSet<E> m_Models

the set of classifiers that constitute the library

Class weka.classifiers.EnsembleLibraryModel extends java.lang.Object implements Serializable

serialVersionUID: 7932816660173443200L

Serialization Methods

readObject

private void readObject(java.io.ObjectInputStream stream)
                 throws java.io.IOException,
                        java.lang.ClassNotFoundException

Custom serialization method

Throws:

java.io.IOException - if something goes wrong IO-wise

java.lang.ClassNotFoundException - if class couldn't be found

writeObject

private void writeObject(java.io.ObjectOutputStream stream)
                  throws java.io.IOException

Custom serialization method

Throws:

java.io.IOException - if something goes wrong IO-wise

Serialized Fields

m_Classifier

Classifier m_Classifier

this is an array of options

m_DescriptionText

java.lang.String m_DescriptionText

the description of this classifier

m_ErrorText

java.lang.String m_ErrorText

this is also saved separately

m_OptionsWereValid

boolean m_OptionsWereValid

a boolean variable tracking whether or not this classifier was able to be built successfully with the given set of options

m_StringRepresentation

java.lang.String m_StringRepresentation

this is stores redundantly to speed up the many operations that frequently need to get the model string representations (like JList renderers)

Class weka.classifiers.EnsembleLibraryModelComparator extends java.lang.Object implements Serializable

serialVersionUID: -6522464740036141188L

Class weka.classifiers.IteratedSingleClassifierEnhancer extends SingleClassifierEnhancer implements Serializable

serialVersionUID: -6217979135443319724L

Serialized Fields

m_Classifiers

Classifier[] m_Classifiers

Array for storing the generated base classifiers.

m_NumIterations

int m_NumIterations

The number of iterations.

Class weka.classifiers.MultipleClassifiersCombiner extends Classifier implements Serializable

serialVersionUID: 2776436621129422119L

Serialized Fields

m_Classifiers

Classifier[] m_Classifiers

Array for storing the generated base classifiers.

Class weka.classifiers.RandomizableClassifier extends Classifier implements Serializable

serialVersionUID: -8816375798262351903L

Serialized Fields

m_Seed

int m_Seed

The random number seed.

Class weka.classifiers.RandomizableIteratedSingleClassifierEnhancer extends IteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: 5063351391524938557L

Serialized Fields

m_Seed

int m_Seed

The random number seed.

Class weka.classifiers.RandomizableMultipleClassifiersCombiner extends MultipleClassifiersCombiner implements Serializable

serialVersionUID: 5057936555724785679L

Serialized Fields

m_Seed

int m_Seed

The random number seed.

Class weka.classifiers.RandomizableSingleClassifierEnhancer extends SingleClassifierEnhancer implements Serializable

serialVersionUID: 558286687096157160L

Serialized Fields

m_Seed

int m_Seed

The random number seed.

Class weka.classifiers.SingleClassifierEnhancer extends Classifier implements Serializable

serialVersionUID: -3665885256363525164L

Serialized Fields

m_Classifier

Classifier m_Classifier

The base classifier to use

Package weka.classifiers.bayes

Class weka.classifiers.bayes.AODE extends Classifier implements Serializable

serialVersionUID: 9197439980415113523L

Serialized Fields

m_CondiCounts

double[][][] m_CondiCounts

3D array (m_NumClasses * m_TotalAttValues * m_TotalAttValues) of attribute counts, i.e., the number of times an attribute value occurs in conjunction with another attribute value and a class value.

m_ClassCounts

double[] m_ClassCounts

The number of times each class value occurs in the dataset

m_SumForCounts

double[][] m_SumForCounts

The sums of attribute-class counts -- if there are no missing values for att, then m_SumForCounts[classVal][att] will be the same as m_ClassCounts[classVal]

m_NumClasses

int m_NumClasses

The number of classes

m_NumAttributes

int m_NumAttributes

The number of attributes in dataset, including class

m_NumInstances

int m_NumInstances

The number of instances in the dataset

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Instances

Instances m_Instances

The dataset

m_TotalAttValues

int m_TotalAttValues

The total number of values (including an extra for each attribute's missing value, which are included in m_CondiCounts) for all attributes (not including class). E.g., for three atts each with two possible values, m_TotalAttValues would be 9 (6 values + 3 missing). This variable is used when allocating space for m_CondiCounts matrix.

m_StartAttIndex

int[] m_StartAttIndex

The starting index (in the m_CondiCounts matrix) of the values for each attribute

m_NumAttValues

int[] m_NumAttValues

The number of values for each attribute

m_Frequencies

double[] m_Frequencies

The frequency of each attribute value for the dataset

m_SumInstances

double m_SumInstances

The number of valid class values observed in dataset -- with no missing classes, this number is the same as m_NumInstances.

m_Limit

int m_Limit

An att's frequency must be this value or more to be a superParent

m_Debug

boolean m_Debug

If true, outputs debugging info

m_MEstimates

boolean m_MEstimates

flag for using m-estimates

m_Weight

int m_Weight

value for m in m-estimate

Class weka.classifiers.bayes.AODEsr extends Classifier implements Serializable

serialVersionUID: 5602143019183068848L

Serialized Fields

m_CondiCounts

double[][][] m_CondiCounts

3D array (m_NumClasses * m_TotalAttValues * m_TotalAttValues) of attribute counts, i.e. the number of times an attribute value occurs in conjunction with another attribute value and a class value.

m_CondiCountsNoClass

double[][] m_CondiCountsNoClass

2D array (m_TotalAttValues * m_TotalAttValues) of attributes counts. similar to m_CondiCounts, but ignoring class value.

m_ClassCounts

double[] m_ClassCounts

The number of times each class value occurs in the dataset

m_SumForCounts

double[][] m_SumForCounts

The sums of attribute-class counts -- if there are no missing values for att, then m_SumForCounts[classVal][att] will be the same as m_ClassCounts[classVal]

m_NumClasses

int m_NumClasses

The number of classes

m_NumAttributes

int m_NumAttributes

The number of attributes in dataset, including class

m_NumInstances

int m_NumInstances

The number of instances in the dataset

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Instances

Instances m_Instances

The dataset

m_TotalAttValues

int m_TotalAttValues

The total number of values (including an extra for each attribute's missing value, which are included in m_CondiCounts) for all attributes (not including class). Eg. for three atts each with two possible values, m_TotalAttValues would be 9 (6 values + 3 missing). This variable is used when allocating space for m_CondiCounts matrix.

m_StartAttIndex

int[] m_StartAttIndex

The starting index (in the m_CondiCounts matrix) of the values for each attribute

m_NumAttValues

int[] m_NumAttValues

The number of values for each attribute

m_Frequencies

double[] m_Frequencies

The frequency of each attribute value for the dataset

m_SumInstances

double m_SumInstances

The number of valid class values observed in dataset -- with no missing classes, this number is the same as m_NumInstances.

m_Limit

int m_Limit

An att's frequency must be this value or more to be a superParent

m_Debug

boolean m_Debug

If true, outputs debugging info

m_MWeight

double m_MWeight

m value for m-estimation

m_Laplace

boolean m_Laplace

Using LapLace estimation or not

m_Critical

int m_Critical

the critical value for the specialization-generalization

Class weka.classifiers.bayes.BayesianLogisticRegression extends Classifier implements Serializable

serialVersionUID: -8013478897911757631L

Serialized Fields

debug

boolean debug

DEBUG Mode

NormalizeData

boolean NormalizeData

Choose whether to normalize data or not

Tolerance

double Tolerance

Tolerance criteria for the stopping criterion.

Threshold

double Threshold

Threshold for binary classification of probabilisitic estimate

PriorClass

int PriorClass

Distribution Prior class

NumFolds

int NumFolds

NumFolds for CV based Hyperparameters selection

HyperparameterSelection

int HyperparameterSelection

Hyperparameter selection method

ClassIndex

int ClassIndex

The class index from the training data

HyperparameterValue

double HyperparameterValue

Best hyperparameter for test phase

HyperparameterRange

java.lang.String HyperparameterRange

CV Hyperparameter Range

maxIterations

int maxIterations

Maximum number of iterations

iterationCounter

int iterationCounter

Iteration counter

BetaVector

double[] BetaVector

Array for storing coefficients of Bayesian regression model.

DeltaBeta

double[] DeltaBeta

Array to store Regression Coefficient updates.

DeltaUpdate

double[] DeltaUpdate

Trust Region Radius Update

Delta

double[] Delta

Trust Region Radius

Hyperparameters

double[] Hyperparameters

Array to store Hyperparameter values for each feature.

R

double[] R

R(i)= BetaVector X x(i) X y(i). This an intermediate value with respect to vector BETA, input values and corresponding class labels

DeltaR

double[] DeltaR

This vector is used to store the increments on the R(i). It is also used to determining the stopping criterion.

Change

double Change

This variable is used to keep track of change in the value of delta summation of r(i).

m_Filter

Filter m_Filter

Filter interface used to point to weka.filters.unsupervised.attribute.Normalize object

m_Instances

Instances m_Instances

Dataset provided to do Training/Test set.

m_PriorUpdate

Prior m_PriorUpdate

Prior class object interface

Class weka.classifiers.bayes.BayesNet extends Classifier implements Serializable

serialVersionUID: 746037443258775954L

Serialized Fields

m_ParentSets

ParentSet[] m_ParentSets

The parent sets.

m_Distributions

Estimator[][] m_Distributions

The attribute estimators containing CPTs.

m_DiscretizeFilter

Discretize m_DiscretizeFilter

filter used to quantize continuous variables, if any

m_nNonDiscreteAttribute

int m_nNonDiscreteAttribute

attribute index of a non-nominal attribute

m_MissingValuesFilter

ReplaceMissingValues m_MissingValuesFilter

filter used to fill in missing values, if any

m_NumClasses

int m_NumClasses

The number of classes

m_Instances

Instances m_Instances

The dataset header for the purposes of printing out a semi-intelligible model

m_ADTree

ADNode m_ADTree

Datastructure containing ADTree representation of the database. This may result in more efficient access to the data.

m_otherBayesNet

BIFReader m_otherBayesNet

Bayes network to compare the structure with.

m_bUseADTree

boolean m_bUseADTree

Use the experimental ADTree datastructure for calculating contingency tables

m_SearchAlgorithm

SearchAlgorithm m_SearchAlgorithm

Search algorithm used for learning the structure of a network.

m_BayesNetEstimator

BayesNetEstimator m_BayesNetEstimator

Search algorithm used for learning the structure of a network.

Class weka.classifiers.bayes.ComplementNaiveBayes extends Classifier implements Serializable

serialVersionUID: 7246302925903086397L

Serialized Fields

wordWeights

double[][] wordWeights

Weight of words for each class. The weight is actually the log of the probability of a word (w) given a class (c) (i.e. log(Pr[w|c])). The format of the matrix is: wordWeights[class][wordAttribute]

smoothingParameter

double smoothingParameter

Holds the smoothing value to avoid word probabilities of zero.
P.S.: According to the paper this is the Alpha i parameter

m_normalizeWordWeights

boolean m_normalizeWordWeights

True if the words weights are to be normalized

numClasses

int numClasses

Holds the number of Class values present in the set of specified instances

header

Instances header

The instances header that'll be used in toString

Class weka.classifiers.bayes.DMNBtext extends Classifier implements Serializable

serialVersionUID: 5932177450183457085L

Serialized Fields

m_NumIterations

int m_NumIterations

The number of iterations.

m_BinaryWord

boolean m_BinaryWord

m_numClasses

int m_numClasses

m_headerInfo

Instances m_headerInfo

m_binaryClassifiers

DMNBtext.DNBBinary[] m_binaryClassifiers

Class weka.classifiers.bayes.DMNBtext.DNBBinary extends java.lang.Object implements Serializable

Serialized Fields

m_perWordPerClass

double[][] m_perWordPerClass

The number of iterations.

m_wordsPerClass

double[] m_wordsPerClass

m_classIndex

int m_classIndex

m_classDistribution

double[] m_classDistribution

m_numAttributes

int m_numAttributes

number of unique words

m_targetClass

int m_targetClass

m_WordLaplace

double m_WordLaplace

m_coefficient

double[] m_coefficient

m_classRatio

double m_classRatio

m_wordRatio

double m_wordRatio

Class weka.classifiers.bayes.HNB extends Classifier implements Serializable

serialVersionUID: -4503874444306113214L

Serialized Fields

m_ClassCounts

double[] m_ClassCounts

The number of each class value occurs in the dataset

m_ClassAttAttCounts

double[][][] m_ClassAttAttCounts

The number of class and two attributes values occurs in the dataset

m_NumAttValues

int[] m_NumAttValues

The number of values for each attribute in the dataset

m_TotalAttValues

int m_TotalAttValues

The number of values for all attributes in the dataset

m_NumClasses

int m_NumClasses

The number of classes in the dataset

m_NumAttributes

int m_NumAttributes

The number of attributes including class in the dataset

m_NumInstances

int m_NumInstances

The number of instances in the dataset

m_ClassIndex

int m_ClassIndex

The index of the class attribute in the dataset

m_StartAttIndex

int[] m_StartAttIndex

The starting index of each attribute in the dataset

m_condiMutualInfo

double[][] m_condiMutualInfo

The 2D array of conditional mutual information of each pair attributes

Class weka.classifiers.bayes.NaiveBayes extends Classifier implements Serializable

serialVersionUID: 5995231201785697655L

Serialized Fields

m_Distributions

Estimator[][] m_Distributions

The attribute estimators.

m_ClassDistribution

Estimator m_ClassDistribution

The class estimator.

m_UseKernelEstimator

boolean m_UseKernelEstimator

Whether to use kernel density estimator rather than normal distribution for numeric attributes

m_UseDiscretization

boolean m_UseDiscretization

Whether to use discretization than normal distribution for numeric attributes

m_NumClasses

int m_NumClasses

The number of classes (or 1 for numeric class)

m_Instances

Instances m_Instances

The dataset header for the purposes of printing out a semi-intelligible model

m_Disc

Discretize m_Disc

The discretization filter.

m_displayModelInOldFormat

boolean m_displayModelInOldFormat

Class weka.classifiers.bayes.NaiveBayesMultinomial extends Classifier implements Serializable

serialVersionUID: 5932177440181257085L

Serialized Fields

m_probOfWordGivenClass

double[][] m_probOfWordGivenClass

probability that a word (w) exists in a class (H) (i.e. Pr[w|H]) The matrix is in the this format: probOfWordGivenClass[class][wordAttribute] NOTE: the values are actually the log of Pr[w|H]

m_probOfClass

double[] m_probOfClass

the probability of a class (i.e. Pr[H])

m_numAttributes

int m_numAttributes

number of unique words

m_numClasses

int m_numClasses

number of class values

m_lnFactorialCache

double[] m_lnFactorialCache

cache lnFactorial computations

m_headerInfo

Instances m_headerInfo

copy of header information for use in toString method

Class weka.classifiers.bayes.NaiveBayesMultinomialUpdateable extends NaiveBayesMultinomial implements Serializable

serialVersionUID: -7204398796974263186L

Serialized Fields

m_wordsPerClass

double[] m_wordsPerClass

the word count per class

Class weka.classifiers.bayes.NaiveBayesSimple extends Classifier implements Serializable

serialVersionUID: -1478242251770381214L

Serialized Fields

m_Counts

double[][][] m_Counts

All the counts for nominal attributes.

m_Means

double[][] m_Means

The means for numeric attributes.

m_Devs

double[][] m_Devs

The standard deviations for numeric attributes.

m_Priors

double[] m_Priors

The prior probabilities of the classes.

m_Instances

Instances m_Instances

The instances used for training.

Class weka.classifiers.bayes.NaiveBayesUpdateable extends NaiveBayes implements Serializable

serialVersionUID: -5354015843807192221L

Class weka.classifiers.bayes.WAODE extends Classifier implements Serializable

serialVersionUID: 2170978824284697882L

Serialized Fields

m_ClassCounts

double[] m_ClassCounts

The number of each class value occurs in the dataset

m_AttCounts

double[] m_AttCounts

The number of each attribute value occurs in the dataset

m_AttAttCounts

double[][] m_AttAttCounts

The number of two attributes values occurs in the dataset

m_ClassAttAttCounts

double[][][] m_ClassAttAttCounts

The number of class and two attributes values occurs in the dataset

m_NumAttValues

int[] m_NumAttValues

The number of values for each attribute in the dataset

m_TotalAttValues

int m_TotalAttValues

The number of values for all attributes in the dataset

m_NumClasses

int m_NumClasses

The number of classes in the dataset

m_NumAttributes

int m_NumAttributes

The number of attributes including class in the dataset

m_NumInstances

int m_NumInstances

The number of instances in the dataset

m_ClassIndex

int m_ClassIndex

The index of the class attribute in the dataset

m_StartAttIndex

int[] m_StartAttIndex

The starting index of each attribute in the dataset

m_mutualInformation

double[] m_mutualInformation

The array of mutual information between each attribute and class

m_Header

Instances m_Header

the header information of the training data

m_Internals

boolean m_Internals

whether to print more internals in the toString method

See Also:

WAODE.toString()

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

Package weka.classifiers.bayes.blr

Class weka.classifiers.bayes.blr.GaussianPriorImpl extends Prior implements Serializable

serialVersionUID: -2995684220141159223L

Class weka.classifiers.bayes.blr.LaplacePriorImpl extends Prior implements Serializable

serialVersionUID: 2353576123257012607L

Serialized Fields

m_Instances

Instances m_Instances

Beta

double Beta

Hyperparameter

double Hyperparameter

DeltaUpdate

double DeltaUpdate

R

double[] R

Delta

double Delta

Class weka.classifiers.bayes.blr.Prior extends java.lang.Object implements Serializable

Serialized Fields

m_Instances

Instances m_Instances

Beta

double Beta

Hyperparameter

double Hyperparameter

DeltaUpdate

double DeltaUpdate

R

double[] R

Delta

double Delta

log_posterior

double log_posterior

log_likelihood

double log_likelihood

penalty

double penalty

Package weka.classifiers.bayes.net

Class weka.classifiers.bayes.net.ADNode extends java.lang.Object implements Serializable

serialVersionUID: 397409728366910204L

Serialized Fields

m_VaryNodes

VaryNode[] m_VaryNodes

list of VaryNode children

m_Instances

Instance[] m_Instances

list of Instance children (either m_Instances or m_VaryNodes is instantiated)

m_nCount

int m_nCount

count

m_nStartNode

int m_nStartNode

first node in VaryNode array

Class weka.classifiers.bayes.net.BayesNetGenerator extends EditableBayesNet implements Serializable

serialVersionUID: -7462571170596157720L

Serialized Fields

m_nSeed

int m_nSeed

the seed value

random

java.util.Random random

the random number generator

m_bGenerateNet

boolean m_bGenerateNet

m_nNrOfNodes

int m_nNrOfNodes

m_nNrOfArcs

int m_nNrOfArcs

m_nNrOfInstances

int m_nNrOfInstances

m_nCardinality

int m_nCardinality

m_sBIFFile

java.lang.String m_sBIFFile

Class weka.classifiers.bayes.net.BIFReader extends BayesNet implements Serializable

serialVersionUID: -8358864680379881429L

Serialized Fields

m_nPositionX

int[] m_nPositionX

m_nPositionY

int[] m_nPositionY

m_order

int[] m_order

m_sFile

java.lang.String m_sFile

the current filename

Class weka.classifiers.bayes.net.EditableBayesNet extends BayesNet implements Serializable

serialVersionUID: 746037443258735954L

Serialized Fields

m_nPositionX

FastVector m_nPositionX

location of nodes, used for graph drawing *

m_nPositionY

FastVector m_nPositionY

m_fMarginP

FastVector m_fMarginP

marginal distributions *

m_nEvidence

FastVector m_nEvidence

evidence values, used for evidence propagation *

m_undoStack

FastVector m_undoStack

undo stack for undoin edit actions, or redo edit actions

m_nCurrentEditAction

int m_nCurrentEditAction

current action in undo stack

m_nSavedPointer

int m_nSavedPointer

action that the network is saved

m_bNeedsUndoAction

boolean m_bNeedsUndoAction

flag to indicate whether an edit action needs to introduce an undo action. This is only false when an undo or redo action is performed.

Class weka.classifiers.bayes.net.GUI extends javax.swing.JPanel implements Serializable

serialVersionUID: -2038911085935515624L

Serialized Fields

m_layoutEngine

LayoutEngine m_layoutEngine

The current LayoutEngine

m_GraphPanel

weka.classifiers.bayes.net.GUI.GraphPanel m_GraphPanel

Panel actually displaying the graph

m_BayesNet

EditableBayesNet m_BayesNet

Container of Bayesian network

m_sFileName

java.lang.String m_sFileName

String containing file name storing current network

m_marginCalculator

MarginCalculator m_marginCalculator

used for calculating marginals in Bayesian netwowrks

m_marginCalculatorWithEvidence

MarginCalculator m_marginCalculatorWithEvidence

used for calculating marginals in Bayesian netwowrks when evidence is present

m_bViewMargins

boolean m_bViewMargins

flag indicating whether marginal distributions of each of the nodes should be shown in display.

m_bViewCliques

boolean m_bViewCliques

m_menuBar

javax.swing.JMenuBar m_menuBar

The menu bar

m_Instances

Instances m_Instances

data selected from file. Used to train a Bayesian network on

m_jTfZoom

javax.swing.JTextField m_jTfZoom

Text field for specifying zoom

m_jTbTools

javax.swing.JToolBar m_jTbTools

toolbar containing buttons at top of window

m_jStatusBar

javax.swing.JLabel m_jStatusBar

status bar at bottom of window

m_jTfNodeWidth

javax.swing.JTextField m_jTfNodeWidth

TextField for node's width

m_jTfNodeHeight

javax.swing.JTextField m_jTfNodeHeight

TextField for nodes height

m_jScrollPane

javax.swing.JScrollPane m_jScrollPane

this contains the m_GraphPanel GraphPanel

ICONPATH

java.lang.String ICONPATH

path for icons

m_fScale

double m_fScale

current zoom value

m_nNodeHeight

int m_nNodeHeight

standard width of node

m_nNodeWidth

int m_nNodeWidth

m_nPaddedNodeWidth

int m_nPaddedNodeWidth

m_nZoomPercents

int[] m_nZoomPercents

used when using zoomIn and zoomOut buttons

a_new

javax.swing.Action a_new

actions triggered by GUI events

a_quit

javax.swing.Action a_quit

a_save

javax.swing.Action a_save

a_export

weka.classifiers.bayes.net.GUI.ActionExport a_export

a_print

weka.classifiers.bayes.net.GUI.ActionPrint a_print

a_load

javax.swing.Action a_load

a_zoomin

javax.swing.Action a_zoomin

a_zoomout

javax.swing.Action a_zoomout

a_layout

javax.swing.Action a_layout

a_saveas

javax.swing.Action a_saveas

a_viewtoolbar

javax.swing.Action a_viewtoolbar

a_viewstatusbar

javax.swing.Action a_viewstatusbar

a_networkgenerator

javax.swing.Action a_networkgenerator

a_datagenerator

javax.swing.Action a_datagenerator

a_datasetter

javax.swing.Action a_datasetter

a_learn

javax.swing.Action a_learn

a_learnCPT

javax.swing.Action a_learnCPT

a_help

javax.swing.Action a_help

a_about

javax.swing.Action a_about

a_addnode

weka.classifiers.bayes.net.GUI.ActionAddNode a_addnode

a_delnode

javax.swing.Action a_delnode

a_cutnode

javax.swing.Action a_cutnode

a_copynode

javax.swing.Action a_copynode

a_pastenode

javax.swing.Action a_pastenode

a_selectall

javax.swing.Action a_selectall

a_addarc

javax.swing.Action a_addarc

a_delarc

javax.swing.Action a_delarc

a_undo

javax.swing.Action a_undo

a_redo

javax.swing.Action a_redo

a_alignleft

javax.swing.Action a_alignleft

a_alignright

javax.swing.Action a_alignright

a_aligntop

javax.swing.Action a_aligntop

a_alignbottom

javax.swing.Action a_alignbottom

a_centerhorizontal

javax.swing.Action a_centerhorizontal

a_centervertical

javax.swing.Action a_centervertical

a_spacehorizontal

javax.swing.Action a_spacehorizontal

a_spacevertical

javax.swing.Action a_spacevertical

m_nCurrentNode

int m_nCurrentNode

node currently selected through right clicking

m_Selection

weka.classifiers.bayes.net.GUI.Selection m_Selection

selection of nodes

m_nSelectedRect

java.awt.Rectangle m_nSelectedRect

selection rectangle drawn through dragging with left mouse button

m_clipboard

weka.classifiers.bayes.net.GUI.ClipBoard m_clipboard

Class weka.classifiers.bayes.net.MarginCalculator extends java.lang.Object implements Serializable

serialVersionUID: 650278019241175534L

Serialized Fields

m_debug

boolean m_debug

m_root

MarginCalculator.JunctionTreeNode m_root

jtNodes

MarginCalculator.JunctionTreeNode[] jtNodes

m_Margins

double[][] m_Margins

Class weka.classifiers.bayes.net.MarginCalculator.JunctionTreeNode extends java.lang.Object implements Serializable

serialVersionUID: 650278019241175536L

Serialized Fields

m_bayesNet

BayesNet m_bayesNet

reference Bayes net for information about variables like name, cardinality, etc. but not for relations between nodes

m_nNodes

int[] m_nNodes

nodes of the Bayes net in this junction node

m_nCardinality

int m_nCardinality

cardinality of the instances of variables in this junction node

m_fi

double[] m_fi

potentials for first network

m_P

double[] m_P

distribution over this junction node according to first Bayes network

m_MarginalP

double[][] m_MarginalP

m_parentSeparator

MarginCalculator.JunctionTreeSeparator m_parentSeparator

m_children

java.util.Vector<E> m_children

Class weka.classifiers.bayes.net.MarginCalculator.JunctionTreeSeparator extends java.lang.Object implements Serializable

serialVersionUID: 6502780192411755343L

Serialized Fields

m_nNodes

int[] m_nNodes

m_nCardinality

int m_nCardinality

m_fiParent

double[] m_fiParent

m_fiChild

double[] m_fiChild

m_parentNode

MarginCalculator.JunctionTreeNode m_parentNode

m_childNode

MarginCalculator.JunctionTreeNode m_childNode

m_bayesNet

BayesNet m_bayesNet

Class weka.classifiers.bayes.net.ParentSet extends java.lang.Object implements Serializable

serialVersionUID: 4155021284407181838L

Serialized Fields

m_nParents

int[] m_nParents

Holds indexes of parents

m_nNrOfParents

int m_nNrOfParents

Holds number of parents

m_nCardinalityOfParents

int m_nCardinalityOfParents

Holds cardinality of parents (= number of instantiations the parents can take)

Class weka.classifiers.bayes.net.VaryNode extends java.lang.Object implements Serializable

serialVersionUID: -6196294370675872424L

Serialized Fields

m_iNode

int m_iNode

index of the node varied

m_nMCV

int m_nMCV

most common value

m_ADNodes

ADNode[] m_ADNodes

list of ADNode children

Package weka.classifiers.bayes.net.estimate

Class weka.classifiers.bayes.net.estimate.BayesNetEstimator extends java.lang.Object implements Serializable

serialVersionUID: 2184330197666253884L

Serialized Fields

m_fAlpha

double m_fAlpha

Holds prior on count

Class weka.classifiers.bayes.net.estimate.BMAEstimator extends SimpleEstimator implements Serializable

serialVersionUID: -1846028304233257309L

Serialized Fields

m_bUseK2Prior

boolean m_bUseK2Prior

whether to use K2 prior

Class weka.classifiers.bayes.net.estimate.DiscreteEstimatorBayes extends Estimator implements Serializable

serialVersionUID: 4215400230843212684L

Serialized Fields

m_Counts

double[] m_Counts

Hold the counts

m_SumOfCounts

double m_SumOfCounts

Hold the sum of counts

m_nSymbols

int m_nSymbols

Holds number of symbols in distribution

m_fPrior

double m_fPrior

Holds the prior probability

Class weka.classifiers.bayes.net.estimate.DiscreteEstimatorFullBayes extends DiscreteEstimatorBayes implements Serializable

serialVersionUID: 6774941981423312133L

Class weka.classifiers.bayes.net.estimate.MultiNomialBMAEstimator extends BayesNetEstimator implements Serializable

serialVersionUID: 8330705772601586313L

Serialized Fields

m_bUseK2Prior

boolean m_bUseK2Prior

whether to use K2 prior

Class weka.classifiers.bayes.net.estimate.SimpleEstimator extends BayesNetEstimator implements Serializable

serialVersionUID: 5874941612331806172L

Package weka.classifiers.bayes.net.search

Class weka.classifiers.bayes.net.search.SearchAlgorithm extends java.lang.Object implements Serializable

serialVersionUID: 6164792240778525312L

Serialized Fields

m_nMaxNrOfParents

int m_nMaxNrOfParents

Holds upper bound on number of parents

m_bInitAsNaiveBayes

boolean m_bInitAsNaiveBayes

determines whether initial structure is an empty graph or a Naive Bayes network

m_bMarkovBlanketClassifier

boolean m_bMarkovBlanketClassifier

Determines whether after structure is found a MarkovBlanketClassifier correction should be applied If this is true, m_bInitAsNaiveBayes is overridden and interpreted as false.

Package weka.classifiers.bayes.net.search.ci

Class weka.classifiers.bayes.net.search.ci.CISearchAlgorithm extends LocalScoreSearchAlgorithm implements Serializable

serialVersionUID: 3165802334119704560L

Serialized Fields

m_BayesNet

BayesNet m_BayesNet

m_instances

Instances m_instances

Class weka.classifiers.bayes.net.search.ci.ICSSearchAlgorithm extends CISearchAlgorithm implements Serializable

serialVersionUID: -2510985917284798576L

Serialized Fields

m_nMaxCardinality

int m_nMaxCardinality

maximum size of separating set

Package weka.classifiers.bayes.net.search.fixed

Class weka.classifiers.bayes.net.search.fixed.FromFile extends SearchAlgorithm implements Serializable

serialVersionUID: 7334358169507619525L

Serialized Fields

m_sBIFFile

java.lang.String m_sBIFFile

name of file to read structure from

Class weka.classifiers.bayes.net.search.fixed.NaiveBayes extends SearchAlgorithm implements Serializable

serialVersionUID: -4808572519709755811L

Package weka.classifiers.bayes.net.search.global

Class weka.classifiers.bayes.net.search.global.GeneticSearch extends GlobalScoreSearchAlgorithm implements Serializable

serialVersionUID: 4236165533882462203L

Serialized Fields

m_nRuns

int m_nRuns

number of runs

m_nPopulationSize

int m_nPopulationSize

size of population

m_nDescendantPopulationSize

int m_nDescendantPopulationSize

size of descendant population

m_bUseCrossOver

boolean m_bUseCrossOver

use cross-over?

m_bUseMutation

boolean m_bUseMutation

use mutation?

m_bUseTournamentSelection

boolean m_bUseTournamentSelection

use tournament selection or take best sub-population

m_nSeed

int m_nSeed

random number seed

m_random

java.util.Random m_random

random number generator

Class weka.classifiers.bayes.net.search.global.GlobalScoreSearchAlgorithm extends SearchAlgorithm implements Serializable

serialVersionUID: 7341389867906199781L

Serialized Fields

m_BayesNet

BayesNet m_BayesNet

points to Bayes network for which a structure is searched for

m_bUseProb

boolean m_bUseProb

toggle between scoring using accuracy = 0-1 loss (when false) or class probabilities (when true)

m_nNrOfFolds

int m_nNrOfFolds

number of folds for k-fold cross validation

m_nCVType

int m_nCVType

Holds the cross validation strategy used to measure quality of network

Class weka.classifiers.bayes.net.search.global.HillClimber extends GlobalScoreSearchAlgorithm implements Serializable

serialVersionUID: -3885042888195820149L

Serialized Fields

m_bUseArcReversal

boolean m_bUseArcReversal

use the arc reversal operator

Class weka.classifiers.bayes.net.search.global.K2 extends GlobalScoreSearchAlgorithm implements Serializable

serialVersionUID: -6626871067466338256L

Serialized Fields

m_bRandomOrder

boolean m_bRandomOrder

Holds flag to indicate ordering should be random

Class weka.classifiers.bayes.net.search.global.RepeatedHillClimber extends HillClimber implements Serializable

serialVersionUID: -7359197180460703069L

Serialized Fields

m_nRuns

int m_nRuns

number of runs

m_nSeed

int m_nSeed

random number seed

m_random

java.util.Random m_random

random number generator

Class weka.classifiers.bayes.net.search.global.SimulatedAnnealing extends GlobalScoreSearchAlgorithm implements Serializable

serialVersionUID: -5482721887881010916L

Serialized Fields

m_fTStart

double m_fTStart

start temperature

m_fDelta

double m_fDelta

change in temperature at every run

m_nRuns

int m_nRuns

number of runs

m_bUseArcReversal

boolean m_bUseArcReversal

use the arc reversal operator

m_nSeed

int m_nSeed

random number seed

m_random

java.util.Random m_random

random number generator

Class weka.classifiers.bayes.net.search.global.TabuSearch extends HillClimber implements Serializable

serialVersionUID: 1176705618756672292L

Serialized Fields

m_nRuns

int m_nRuns

number of runs

m_nTabuList

int m_nTabuList

size of tabu list

m_oTabuList

weka.classifiers.bayes.net.search.global.HillClimber.Operation[] m_oTabuList

the actual tabu list

Class weka.classifiers.bayes.net.search.global.TAN extends GlobalScoreSearchAlgorithm implements Serializable

serialVersionUID: 1715277053980895298L

Package weka.classifiers.bayes.net.search.local

Class weka.classifiers.bayes.net.search.local.GeneticSearch extends LocalScoreSearchAlgorithm implements Serializable

serialVersionUID: -7037070678911459757L

Serialized Fields

m_nRuns

int m_nRuns

number of runs

m_nPopulationSize

int m_nPopulationSize

size of population

m_nDescendantPopulationSize

int m_nDescendantPopulationSize

size of descendant population

m_bUseCrossOver

boolean m_bUseCrossOver

use cross-over?

m_bUseMutation

boolean m_bUseMutation

use mutation?

m_bUseTournamentSelection

boolean m_bUseTournamentSelection

use tournament selection or take best sub-population

m_nSeed

int m_nSeed

random number seed

m_random

java.util.Random m_random

random number generator

Class weka.classifiers.bayes.net.search.local.HillClimber extends LocalScoreSearchAlgorithm implements Serializable

serialVersionUID: 4322783593818122403L

Serialized Fields

m_Cache

weka.classifiers.bayes.net.search.local.HillClimber.Cache m_Cache

cache for storing score differences

m_bUseArcReversal

boolean m_bUseArcReversal

use the arc reversal operator

Class weka.classifiers.bayes.net.search.local.K2 extends LocalScoreSearchAlgorithm implements Serializable

serialVersionUID: 6176545934752116631L

Serialized Fields

m_bRandomOrder

boolean m_bRandomOrder

Holds flag to indicate ordering should be random

Class weka.classifiers.bayes.net.search.local.LAGDHillClimber extends HillClimber implements Serializable

serialVersionUID: 7217437499439184344L

Serialized Fields

m_nNrOfLookAheadSteps

int m_nNrOfLookAheadSteps

Number of Look Ahead Steps

m_nNrOfGoodOperations

int m_nNrOfGoodOperations

Number of Good Operations per Step

Class weka.classifiers.bayes.net.search.local.LocalScoreSearchAlgorithm extends SearchAlgorithm implements Serializable

serialVersionUID: 3325995552474190374L

Serialized Fields

m_BayesNet

BayesNet m_BayesNet

points to Bayes network for which a structure is searched for

m_fAlpha

double m_fAlpha

Holds prior on count

m_nScoreType

int m_nScoreType

Holds the score type used to measure quality of network

Class weka.classifiers.bayes.net.search.local.RepeatedHillClimber extends HillClimber implements Serializable

serialVersionUID: -6574084564213041174L

Serialized Fields

m_nRuns

int m_nRuns

number of runs

m_nSeed

int m_nSeed

random number seed

m_random

java.util.Random m_random

random number generator

Class weka.classifiers.bayes.net.search.local.SimulatedAnnealing extends LocalScoreSearchAlgorithm implements Serializable

serialVersionUID: 6951955606060513191L

Serialized Fields

m_fTStart

double m_fTStart

start temperature

m_fDelta

double m_fDelta

change in temperature at every run

m_nRuns

int m_nRuns

number of runs

m_bUseArcReversal

boolean m_bUseArcReversal

use the arc reversal operator

m_nSeed

int m_nSeed

random number seed

m_random

java.util.Random m_random

random number generator

Class weka.classifiers.bayes.net.search.local.TabuSearch extends HillClimber implements Serializable

serialVersionUID: 1457344073228786447L

Serialized Fields

m_nRuns

int m_nRuns

number of runs

m_nTabuList

int m_nTabuList

size of tabu list

m_oTabuList

weka.classifiers.bayes.net.search.local.HillClimber.Operation[] m_oTabuList

the actual tabu list

Class weka.classifiers.bayes.net.search.local.TAN extends LocalScoreSearchAlgorithm implements Serializable

serialVersionUID: 965182127977228690L

Package weka.classifiers.evaluation

Class weka.classifiers.evaluation.ConfusionMatrix extends Matrix implements Serializable

serialVersionUID: -181789981401504090L

Serialized Fields

m_ClassNames

java.lang.String[] m_ClassNames

Stores the names of the classes

Class weka.classifiers.evaluation.NominalPrediction extends java.lang.Object implements Serializable

serialVersionUID: -8871333992740492788L

Serialized Fields

m_Distribution

double[] m_Distribution

The predicted probabilities

m_Actual

double m_Actual

The actual class value

m_Predicted

double m_Predicted

The predicted class value

m_Weight

double m_Weight

The weight assigned to this prediction

Class weka.classifiers.evaluation.NumericPrediction extends java.lang.Object implements Serializable

serialVersionUID: -4880216423674233887L

Serialized Fields

m_Actual

double m_Actual

The actual class value

m_Predicted

double m_Predicted

The predicted class value

m_Weight

double m_Weight

The weight assigned to this prediction

Package weka.classifiers.evaluation.output.prediction

Class weka.classifiers.evaluation.output.prediction.AbstractOutput extends java.lang.Object implements Serializable

serialVersionUID: 752696986017306241L

Serialized Fields

m_Header

Instances m_Header

the header of the dataset.

m_Buffer

java.lang.StringBuffer m_Buffer

the buffer to write to.

m_FileBuffer

java.lang.StringBuffer m_FileBuffer

the file buffer to write to.

m_OutputDistribution

boolean m_OutputDistribution

whether to output the class distribution.

m_Attributes

Range m_Attributes

the range of attributes to output.

m_NumDecimals

int m_NumDecimals

the number of decimals after the decimal point.

m_OutputFile

java.io.File m_OutputFile

the file to store the output in.

m_SuppressOutput

boolean m_SuppressOutput

whether to suppress the regular output and only store in file.

Class weka.classifiers.evaluation.output.prediction.CSV extends AbstractOutput implements Serializable

serialVersionUID: 3401604538169573720L

Serialized Fields

m_Delimiter

java.lang.String m_Delimiter

the delimiter.

Class weka.classifiers.evaluation.output.prediction.HTML extends AbstractOutput implements Serializable

serialVersionUID: 7241252244954353300L

Class weka.classifiers.evaluation.output.prediction.Null extends AbstractOutput implements Serializable

serialVersionUID: 4988413155999044966L

Class weka.classifiers.evaluation.output.prediction.PlainText extends AbstractOutput implements Serializable

serialVersionUID: 2033389864898242735L

Class weka.classifiers.evaluation.output.prediction.XML extends AbstractOutput implements Serializable

serialVersionUID: -3165514277316824801L

Package weka.classifiers.functions

Class weka.classifiers.functions.GaussianProcesses extends Classifier implements Serializable

serialVersionUID: -8620066949967678545L

Serialized Fields

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_filterType

int m_filterType

Whether to normalize/standardize/neither

m_Missing

ReplaceMissingValues m_Missing

The filter used to get rid of missing values.

m_checksTurnedOff

boolean m_checksTurnedOff

Turn off all checks and conversions? Turning them off assumes that data is purely numeric, doesn't contain any missing values, and has a numeric class.

m_delta

double m_delta

Gaussian Noise Value.

m_classIndex

int m_classIndex

The class index from the training data

m_Alin

double m_Alin

The parameters of the linear transforamtion realized by the filter on the class attribute

m_Blin

double m_Blin

m_kernel

Kernel m_kernel

Kernel to use

m_NumTrain

int m_NumTrain

The number of training instances

m_avg_target

double m_avg_target

The training data.

m_C

Matrix m_C

The covariance matrix.

m_t

Matrix m_t

The vector of target values.

m_KernelIsLinear

boolean m_KernelIsLinear

whether the kernel is a linear one

Class weka.classifiers.functions.IsotonicRegression extends Classifier implements Serializable

serialVersionUID: 1679336022835454137L

Serialized Fields

m_attribute

Attribute m_attribute

The chosen attribute

m_cuts

double[] m_cuts

The array of cut points

m_values

double[] m_values

The predicted value in each interval.

m_minMsq

double m_minMsq

The minimum mean squared error that has been achieved.

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

Class weka.classifiers.functions.LeastMedSq extends Classifier implements Serializable

serialVersionUID: 4288954049987652970L

Serialized Fields

m_Residuals

double[] m_Residuals

m_weight

double[] m_weight

m_SSR

double m_SSR

m_scalefactor

double m_scalefactor

m_bestMedian

double m_bestMedian

m_currentRegression

LinearRegression m_currentRegression

m_bestRegression

LinearRegression m_bestRegression

m_ls

LinearRegression m_ls

m_Data

Instances m_Data

m_RLSData

Instances m_RLSData

m_SubSample

Instances m_SubSample

m_MissingFilter

ReplaceMissingValues m_MissingFilter

m_TransformFilter

NominalToBinary m_TransformFilter

m_SplitFilter

RemoveRange m_SplitFilter

m_samplesize

int m_samplesize

m_samples

int m_samples

m_israndom

boolean m_israndom

m_debug

boolean m_debug

m_random

java.util.Random m_random

m_randomseed

long m_randomseed

Class weka.classifiers.functions.LibLINEAR extends Classifier implements Serializable

serialVersionUID: 230504711L

Serialized Fields

m_Model

java.lang.Object m_Model

LibLINEAR Model

m_Filter

Filter m_Filter

for normalizing the data

m_Normalize

boolean m_Normalize

normalize input data

m_SVMType

int m_SVMType

the SVM solver type

m_eps

double m_eps

stopping criteria

m_Cost

double m_Cost

cost Parameter C

m_Bias

double m_Bias

bias term value

m_WeightLabel

int[] m_WeightLabel

m_Weight

double[] m_Weight

m_ProbabilityEstimates

boolean m_ProbabilityEstimates

whether to generate probability estimates instead of +1/-1 in case of classification problems

m_ReplaceMissingValues

ReplaceMissingValues m_ReplaceMissingValues

The filter used to get rid of missing values.

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_nominalToBinary

boolean m_nominalToBinary

If true, the nominal to binary filter is applied

m_noReplaceMissingValues

boolean m_noReplaceMissingValues

If true, the replace missing values filter is not applied

Class weka.classifiers.functions.LibSVM extends Classifier implements Serializable

serialVersionUID: 14172L

Serialized Fields

m_Model

java.lang.Object m_Model

LibSVM Model.

m_Filter

Filter m_Filter

for normalizing the data.

m_ReplaceMissingValues

ReplaceMissingValues m_ReplaceMissingValues

The filter used to get rid of missing values.

m_Normalize

boolean m_Normalize

normalize input data.

m_noReplaceMissingValues

boolean m_noReplaceMissingValues

If true, the replace missing values filter is not applied.

m_SVMType

int m_SVMType

the SVM type.

m_KernelType

int m_KernelType

the kernel type.

m_Degree

int m_Degree

for poly - in older versions of libsvm declared as a double. At least since 2.82 it is an int.

m_Gamma

double m_Gamma

for poly/rbf/sigmoid.

m_GammaActual

double m_GammaActual

for poly/rbf/sigmoid (the actual gamma).

m_Coef0

double m_Coef0

for poly/sigmoid.

m_CacheSize

double m_CacheSize

in MB.

m_eps

double m_eps

stopping criteria.

m_Cost

double m_Cost

cost, for C_SVC, EPSILON_SVR and NU_SVR.

m_WeightLabel

int[] m_WeightLabel

for C_SVC.

m_Weight

double[] m_Weight

for C_SVC.

m_nu

double m_nu

for NU_SVC, ONE_CLASS, and NU_SVR.

m_Loss

double m_Loss

loss, for EPSILON_SVR.

m_Shrinking

boolean m_Shrinking

use the shrinking heuristics.

m_ProbabilityEstimates

boolean m_ProbabilityEstimates

whether to generate probability estimates instead of +1/-1 in case of classification problems.

m_ModelFile

java.io.File m_ModelFile

the file to save the libsvm-internal model to.

Class weka.classifiers.functions.LinearRegression extends Classifier implements Serializable

serialVersionUID: -3364580862046573747L

Serialized Fields

m_Coefficients

double[] m_Coefficients

Array for storing coefficients of linear regression.

m_SelectedAttributes

boolean[] m_SelectedAttributes

Which attributes are relevant?

m_TransformedData

Instances m_TransformedData

Variable for storing transformed training data.

m_MissingFilter

ReplaceMissingValues m_MissingFilter

The filter for removing missing values.

m_TransformFilter

NominalToBinary m_TransformFilter

The filter storing the transformation from nominal to binary attributes.

m_ClassStdDev

double m_ClassStdDev

The standard deviations of the class attribute

m_ClassMean

double m_ClassMean

The mean of the class attribute

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Means

double[] m_Means

The attributes means

m_StdDevs

double[] m_StdDevs

The attribute standard deviations

b_Debug

boolean b_Debug

True if debug output will be printed

m_AttributeSelection

int m_AttributeSelection

The current attribute selection method

m_EliminateColinearAttributes

boolean m_EliminateColinearAttributes

Try to eliminate correlated attributes?

m_checksTurnedOff

boolean m_checksTurnedOff

Turn off all checks and conversions?

m_Ridge

double m_Ridge

The ridge parameter

Class weka.classifiers.functions.Logistic extends Classifier implements Serializable

serialVersionUID: 3932117032546553727L

Serialized Fields

m_Par

double[][] m_Par

The coefficients (optimized parameters) of the model

m_Data

double[][] m_Data

The data saved as a matrix

m_NumPredictors

int m_NumPredictors

The number of attributes in the model

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_NumClasses

int m_NumClasses

The number of the class labels

m_Ridge

double m_Ridge

The ridge parameter.

m_AttFilter

RemoveUseless m_AttFilter

An attribute filter

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_ReplaceMissingValues

ReplaceMissingValues m_ReplaceMissingValues

The filter used to get rid of missing values.

m_Debug

boolean m_Debug

Debugging output

m_LL

double m_LL

Log-likelihood of the searched model

m_MaxIts

int m_MaxIts

The maximum number of iterations.

m_structure

Instances m_structure

Class weka.classifiers.functions.MultilayerPerceptron extends Classifier implements Serializable

serialVersionUID: 572250905027665169L

Serialized Fields

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

m_instances

Instances m_instances

The training instances.

m_currentInstance

Instance m_currentInstance

The current instance running through the network.

m_numeric

boolean m_numeric

A flag to say that it's a numeric class.

m_attributeRanges

double[] m_attributeRanges

The ranges for all the attributes.

m_attributeBases

double[] m_attributeBases

The base values for all the attributes.

m_outputs

weka.classifiers.functions.MultilayerPerceptron.NeuralEnd[] m_outputs

The output units.(only feeds the errors, does no calcs)

m_inputs

weka.classifiers.functions.MultilayerPerceptron.NeuralEnd[] m_inputs

The input units.(only feeds the inputs does no calcs)

m_neuralNodes

NeuralConnection[] m_neuralNodes

All the nodes that actually comprise the logical neural net.

m_numClasses

int m_numClasses

The number of classes.

m_numAttributes

int m_numAttributes

The number of attributes.

m_nodePanel

weka.classifiers.functions.MultilayerPerceptron.NodePanel m_nodePanel

The panel the nodes are displayed on.

m_controlPanel

weka.classifiers.functions.MultilayerPerceptron.ControlPanel m_controlPanel

The control panel.

m_nextId

int m_nextId

The next id number available for default naming.

m_selected

FastVector m_selected

A Vector list of the units currently selected.

m_graphers

FastVector m_graphers

A Vector list of the graphers.

m_numEpochs

int m_numEpochs

The number of epochs to train through.

m_stopIt

boolean m_stopIt

a flag to state if the network should be running, or stopped.

m_stopped

boolean m_stopped

a flag to state that the network has in fact stopped.

m_accepted

boolean m_accepted

a flag to state that the network should be accepted the way it is.

m_win

javax.swing.JFrame m_win

The window for the network.

m_autoBuild

boolean m_autoBuild

A flag to tell the build classifier to automatically build a neural net.

m_gui

boolean m_gui

A flag to state that the gui for the network should be brought up. To allow interaction while training.

m_valSize

int m_valSize

An int to say how big the validation set should be.

m_driftThreshold

int m_driftThreshold

The number to to use to quit on validation testing.

m_randomSeed

int m_randomSeed

The number used to seed the random number generator.

m_random

java.util.Random m_random

The actual random number generator.

m_useNomToBin

boolean m_useNomToBin

A flag to state that a nominal to binary filter should be used.

m_nominalToBinaryFilter

NominalToBinary m_nominalToBinaryFilter

The actual filter.

m_hiddenLayers

java.lang.String m_hiddenLayers

The string that defines the hidden layers

m_normalizeAttributes

boolean m_normalizeAttributes

This flag states that the user wants the input values normalized.

m_decay

boolean m_decay

This flag states that the user wants the learning rate to decay.

m_learningRate

double m_learningRate

This is the learning rate for the network.

m_momentum

double m_momentum

This is the momentum for the network.

m_epoch

int m_epoch

Shows the number of the epoch that the network just finished.

m_error

double m_error

Shows the error of the epoch that the network just finished.

m_reset

boolean m_reset

This flag states that the user wants the network to restart if it is found to be generating infinity or NaN for the error value. This would restart the network with the current options except that the learning rate would be smaller than before, (perhaps half of its current value). This option will not be available if the gui is chosen (if the gui is open the user can fix the network themselves, it is an architectural minefield for the network to be reset with the gui open).

m_normalizeClass

boolean m_normalizeClass

This flag states that the user wants the class to be normalized while processing in the network is done. (the final answer will be in the original range regardless). This option will only be used when the class is numeric.

m_sigmoidUnit

SigmoidUnit m_sigmoidUnit

this is a sigmoid unit.

m_linearUnit

LinearUnit m_linearUnit

This is a linear unit.

Class weka.classifiers.functions.MultilayerPerceptron.NeuralEnd extends NeuralConnection implements Serializable

serialVersionUID: 7305185603191183338L

Serialized Fields

m_link

int m_link

the value that represents the instance value this node represents. For an input it is the attribute number, for an output, if nominal it is the class value.

m_input

boolean m_input

True if node is an input, False if it's an output.

Class weka.classifiers.functions.PaceRegression extends Classifier implements Serializable

serialVersionUID: 7230266976059115435L

Serialized Fields

m_Model

Instances m_Model

The model used

m_Coefficients

double[] m_Coefficients

Array for storing coefficients of linear regression.

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Debug

boolean m_Debug

True if debug output will be printed

paceEstimator

int paceEstimator

the estimator

olscThreshold

double olscThreshold

Class weka.classifiers.functions.PLSClassifier extends Classifier implements Serializable

serialVersionUID: 4819775160590973256L

Serialized Fields

m_Filter

PLSFilter m_Filter

the PLS filter

m_ActualFilter

PLSFilter m_ActualFilter

the actual filter to use

Class weka.classifiers.functions.RBFNetwork extends Classifier implements Serializable

serialVersionUID: -3669814959712675720L

Serialized Fields

m_logistic

Logistic m_logistic

The logistic regression for classification problems

m_linear

LinearRegression m_linear

The linear regression for numeric problems

m_basisFilter

ClusterMembership m_basisFilter

The filter for producing the meta data

m_standardize

Standardize m_standardize

Filter used for normalizing the data

m_numClusters

int m_numClusters

The number of clusters (basis functions to generate)

m_ridge

double m_ridge

The ridge parameter for the logistic regression.

m_maxIts

int m_maxIts

The maximum number of iterations for logistic regression.

m_clusteringSeed

int m_clusteringSeed

The seed to pass on to K-means

m_minStdDev

double m_minStdDev

The minimum standard deviation

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

Class weka.classifiers.functions.SimpleLinearRegression extends Classifier implements Serializable

serialVersionUID: 1679336022895414137L

Serialized Fields

m_attribute

Attribute m_attribute

The chosen attribute

m_attributeIndex

int m_attributeIndex

The index of the chosen attribute

m_slope

double m_slope

The slope

m_intercept

double m_intercept

The intercept

m_suppressErrorMessage

boolean m_suppressErrorMessage

If true, suppress error message if no useful attribute was found

Class weka.classifiers.functions.SimpleLogistic extends Classifier implements Serializable

serialVersionUID: 7397710626304705059L

Serialized Fields

m_boostedModel

LogisticBase m_boostedModel

The actual logistic regression model

m_NominalToBinary

NominalToBinary m_NominalToBinary

Filter for converting nominal attributes to binary ones

m_ReplaceMissingValues

ReplaceMissingValues m_ReplaceMissingValues

Filter for replacing missing values

m_numBoostingIterations

int m_numBoostingIterations

If non-negative, use this as fixed number of LogitBoost iterations

m_maxBoostingIterations

int m_maxBoostingIterations

Maximum number of iterations for LogitBoost

m_heuristicStop

int m_heuristicStop

Parameter for the heuristic for early stopping of LogitBoost

m_useCrossValidation

boolean m_useCrossValidation

If true, cross-validate number of LogitBoost iterations

m_errorOnProbabilities

boolean m_errorOnProbabilities

If true, use minimize error on probabilities instead of misclassification error

m_weightTrimBeta

double m_weightTrimBeta

Threshold for trimming weights. Instances with a weight lower than this (as a percentage of total weights) are not included in the regression fit.

m_useAIC

boolean m_useAIC

If true, the AIC is used to choose the best iteration

Class weka.classifiers.functions.SMO extends Classifier implements Serializable

serialVersionUID: -6585883636378691736L

Serialized Fields

m_classifiers

SMO.BinarySMO[][] m_classifiers

The binary classifier(s)

m_C

double m_C

The complexity parameter.

m_eps

double m_eps

Epsilon for rounding.

m_tol

double m_tol

Tolerance for accuracy of result.

m_filterType

int m_filterType

Whether to normalize/standardize/neither

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_Missing

ReplaceMissingValues m_Missing

The filter used to get rid of missing values.

m_classIndex

int m_classIndex

The class index from the training data

m_classAttribute

Attribute m_classAttribute

The class attribute

m_KernelIsLinear

boolean m_KernelIsLinear

whether the kernel is a linear one

m_checksTurnedOff

boolean m_checksTurnedOff

Turn off all checks and conversions? Turning them off assumes that data is purely numeric, doesn't contain any missing values, and has a nominal class. Turning them off also means that no header information will be stored if the machine is linear. Finally, it also assumes that no instance has a weight equal to 0.

m_fitLogisticModels

boolean m_fitLogisticModels

Whether logistic models are to be fit

m_numFolds

int m_numFolds

The number of folds for the internal cross-validation

m_randomSeed

int m_randomSeed

The random number seed

m_kernel

Kernel m_kernel

the kernel to use

Class weka.classifiers.functions.SMO.BinarySMO extends java.lang.Object implements Serializable

serialVersionUID: -8246163625699362456L

Serialized Fields

m_alpha

double[] m_alpha

The Lagrange multipliers.

m_b

double m_b

The thresholds.

m_bLow

double m_bLow

The thresholds.

m_bUp

double m_bUp

The thresholds.

m_iLow

int m_iLow

The indices for m_bLow and m_bUp

m_iUp

int m_iUp

The indices for m_bLow and m_bUp

m_data

Instances m_data

The training data.

m_weights

double[] m_weights

Weight vector for linear machine.

m_sparseWeights

double[] m_sparseWeights

Variables to hold weight vector in sparse form. (To reduce storage requirements.)

m_sparseIndices

int[] m_sparseIndices

m_kernel

Kernel m_kernel

Kernel to use

m_class

double[] m_class

The transformed class values.

m_errors

double[] m_errors

The current set of errors for all non-bound examples.

m_I0

SMOset m_I0

{i: 0 < m_alpha[i] < C}

m_I1

SMOset m_I1

{i: m_class[i] = 1, m_alpha[i] = 0}

m_I2

SMOset m_I2

{i: m_class[i] = -1, m_alpha[i] =C}

m_I3

SMOset m_I3

{i: m_class[i] = 1, m_alpha[i] = C}

m_I4

SMOset m_I4

{i: m_class[i] = -1, m_alpha[i] = 0}

m_supportVectors

SMOset m_supportVectors

The set of support vectors

m_logistic

Logistic m_logistic

Stores logistic regression model for probability estimate

m_sumOfWeights

double m_sumOfWeights

Stores the weight of the training instances

Class weka.classifiers.functions.SMOreg extends Classifier implements Serializable

serialVersionUID: -7149606251113102827L

Serialized Fields

m_filterType

int m_filterType

Whether to normalize/standardize/neither

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_Missing

ReplaceMissingValues m_Missing

The filter used to get rid of missing values.

m_onlyNumeric

boolean m_onlyNumeric

Only numeric attributes in the dataset? If so, less need to filter

m_C

double m_C

capacity parameter

m_x1

double m_x1

coefficients used by normalization filter for doing its linear transformation so that result = svmoutput * m_x1 + m_x0

m_x0

double m_x0

m_optimizer

RegOptimizer m_optimizer

contains the algorithm used for learning

m_kernel

Kernel m_kernel

the configured kernel

Class weka.classifiers.functions.VotedPerceptron extends Classifier implements Serializable

serialVersionUID: -1072429260104568698L

Serialized Fields

m_MaxK

int m_MaxK

The maximum number of alterations to the perceptron

m_NumIterations

int m_NumIterations

The number of iterations

m_Exponent

double m_Exponent

The exponent

m_K

int m_K

The actual number of alterations

m_Additions

int[] m_Additions

The training instances added to the perceptron

m_IsAddition

boolean[] m_IsAddition

Addition or subtraction?

m_Weights

int[] m_Weights

The weights for each perceptron

m_Train

Instances m_Train

The training instances

m_Seed

int m_Seed

Seed used for shuffling the dataset

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_ReplaceMissingValues

ReplaceMissingValues m_ReplaceMissingValues

The filter used to get rid of missing values.

Class weka.classifiers.functions.Winnow extends Classifier implements Serializable

serialVersionUID: 3543770107994321324L

Serialized Fields

m_Balanced

boolean m_Balanced

Use the balanced variant?

m_numIterations

int m_numIterations

The number of iterations

m_Alpha

double m_Alpha

The promotion coefficient

m_Beta

double m_Beta

The demotion coefficient

m_Threshold

double m_Threshold

Prediction threshold, <0 == numAttributes

m_Seed

int m_Seed

Random seed used for shuffling the dataset, -1 == disable

m_Mistakes

int m_Mistakes

Accumulated mistake count (for statistics)

m_defaultWeight

double m_defaultWeight

Starting weights for the prediction vector(s)

m_predPosVector

double[] m_predPosVector

The weight vector for prediction (pos)

m_predNegVector

double[] m_predNegVector

The weight vector for prediction (neg)

m_actualThreshold

double m_actualThreshold

The true threshold used for prediction

m_Train

Instances m_Train

The training instances

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_ReplaceMissingValues

ReplaceMissingValues m_ReplaceMissingValues

The filter used to get rid of missing values.

Package weka.classifiers.functions.neural

Class weka.classifiers.functions.neural.LinearUnit extends java.lang.Object implements Serializable

serialVersionUID: 8572152807755673630L

Class weka.classifiers.functions.neural.NeuralConnection extends java.lang.Object implements Serializable

serialVersionUID: -286208828571059163L

Serialized Fields

m_inputList

NeuralConnection[] m_inputList

The list of inputs to this unit.

m_outputList

NeuralConnection[] m_outputList

The list of outputs from this unit.

m_inputNums

int[] m_inputNums

The numbering for the connections at the other end of the input lines.

m_outputNums

int[] m_outputNums

The numbering for the connections at the other end of the out lines.

m_numInputs

int m_numInputs

The number of inputs.

m_numOutputs

int m_numOutputs

The number of outputs.

m_unitValue

double m_unitValue

The output value for this unit, NaN if not calculated.

m_unitError

double m_unitError

The error value for this unit, NaN if not calculated.

m_weightsUpdated

boolean m_weightsUpdated

True if the weights have already been updated.

m_id

java.lang.String m_id

The string that uniquely (provided naming is done properly) identifies this unit.

m_type

int m_type

The type of unit this is.

m_x

double m_x

The x coord of this unit purely for displaying purposes.

m_y

double m_y

The y coord of this unit purely for displaying purposes.

Class weka.classifiers.functions.neural.NeuralNode extends NeuralConnection implements Serializable

serialVersionUID: -1085750607680839163L

Serialized Fields

m_weights

double[] m_weights

The weights for each of the input connections, and the threshold.

m_bestWeights

double[] m_bestWeights

The best (lowest error) weights. Only used when validation set is used

m_changeInWeights

double[] m_changeInWeights

The change in the weights.

m_random

java.util.Random m_random

m_methods

NeuralMethod m_methods

Performs the operations for this node. Currently this defines that the node is either a sigmoid or a linear unit.

Class weka.classifiers.functions.neural.SigmoidUnit extends java.lang.Object implements Serializable

serialVersionUID: -5162958458177475652L

Package weka.classifiers.functions.pace

Class weka.classifiers.functions.pace.PaceMatrix extends Matrix implements Serializable

serialVersionUID: 2699925616857843973L

Package weka.classifiers.functions.supportVector

Class weka.classifiers.functions.supportVector.CachedKernel extends Kernel implements Serializable

serialVersionUID: 702810182699015136L

Serialized Fields

m_kernelEvals

int m_kernelEvals

Counts the number of kernel evaluations.

m_cacheHits

int m_cacheHits

Counts the number of kernel cache hits.

m_cacheSize

int m_cacheSize

The size of the cache (a prime number)

m_storage

double[] m_storage

Kernel cache

m_keys

long[] m_keys

m_kernelMatrix

double[][] m_kernelMatrix

The kernel matrix if full cache is used (i.e. size is set to 0)

m_numInsts

int m_numInsts

The number of instance in the dataset

m_cacheSlots

int m_cacheSlots

number of cache slots in an entry

Class weka.classifiers.functions.supportVector.Kernel extends java.lang.Object implements Serializable

serialVersionUID: -6102771099905817064L

Serialized Fields

m_data

Instances m_data

The dataset

m_Debug

boolean m_Debug

enables debugging output

m_ChecksTurnedOff

boolean m_ChecksTurnedOff

Turns off all checks

Class weka.classifiers.functions.supportVector.NormalizedPolyKernel extends PolyKernel implements Serializable

serialVersionUID: 1248574185532130851L

Class weka.classifiers.functions.supportVector.PolyKernel extends CachedKernel implements Serializable

serialVersionUID: -321831645846363201L

Serialized Fields

m_lowerOrder

boolean m_lowerOrder

Use lower-order terms?

m_exponent

double m_exponent

The exponent for the polynomial kernel.

Class weka.classifiers.functions.supportVector.PrecomputedKernelMatrixKernel extends Kernel implements Serializable

serialVersionUID: -321831645846363333L

Serialized Fields

m_KernelMatrixFile

java.io.File m_KernelMatrixFile

The file holding the kernel matrix.

m_KernelMatrix

Matrix m_KernelMatrix

The kernel matrix.

m_Counter

int m_Counter

A classifier counter.

Class weka.classifiers.functions.supportVector.Puk extends CachedKernel implements Serializable

serialVersionUID: 1682161522559978851L

Serialized Fields

m_kernelPrecalc

double[] m_kernelPrecalc

The precalculated dotproducts of <inst_i,inst_i>

m_omega

double m_omega

Omega for the Puk kernel.

m_sigma

double m_sigma

Sigma for the Puk kernel.

m_factor

double m_factor

Cached factor for the Puk kernel.

Class weka.classifiers.functions.supportVector.RBFKernel extends CachedKernel implements Serializable

serialVersionUID: 5247117544316387852L

Serialized Fields

m_kernelPrecalc

double[] m_kernelPrecalc

The precalculated dotproducts of <inst_i,inst_i>

m_gamma

double m_gamma

Gamma for the RBF kernel.

Class weka.classifiers.functions.supportVector.RegOptimizer extends java.lang.Object implements Serializable

serialVersionUID: -2198266997254461814L

Serialized Fields

m_alpha

double[] m_alpha

alpha and alpha* arrays containing weights for solving dual problem

m_alphaStar

double[] m_alphaStar

m_b

double m_b

offset

m_epsilon

double m_epsilon

epsilon of epsilon-insensitive cost function

m_C

double m_C

capacity parameter, copied from SMOreg

m_target

double[] m_target

class values/desired output vector

m_data

Instances m_data

points to data set

m_kernel

Kernel m_kernel

the kernel

m_classIndex

int m_classIndex

index of class variable in data set

m_nInstances

int m_nInstances

number of instances in data set

m_random

java.util.Random m_random

random number generator

m_nSeed

int m_nSeed

seed for initializing random number generator

m_supportVectors

SMOset m_supportVectors

set of support vectors, that is, vectors with alpha(*)!=0

m_nEvals

int m_nEvals

number of kernel evaluations, used for printing statistics only

m_nCacheHits

int m_nCacheHits

number of kernel cache hits, used for printing statistics only

m_weights

double[] m_weights

weights for linear kernel

m_sparseWeights

double[] m_sparseWeights

Variables to hold weight vector in sparse form. (To reduce storage requirements.)

m_sparseIndices

int[] m_sparseIndices

m_bModelBuilt

boolean m_bModelBuilt

flag to indicate whether the model is built yet

m_SVM

SMOreg m_SVM

parent SMOreg class

Class weka.classifiers.functions.supportVector.RegSMO extends RegOptimizer implements Serializable

serialVersionUID: -7504070793279598638L

Serialized Fields

m_eps

double m_eps

tolerance parameter, smaller changes on alpha in inner loop will be ignored

m_error

double[] m_error

error cache containing m_error[i] = SVMOutput(i) - m_target[i] - m_b
note, we don't need m_b in the cache, since if we do, we need to maintain it when m_b is updated

m_alpha1

double m_alpha1

alpha value for first candidate

m_alpha1Star

double m_alpha1Star

alpha* value for first candidate

m_alpha2

double m_alpha2

alpha value for second candidate

m_alpha2Star

double m_alpha2Star

alpha* value for second candidate

Class weka.classifiers.functions.supportVector.RegSMOImproved extends RegSMO implements Serializable

serialVersionUID: 471692841446029784L

Serialized Fields

m_I0

SMOset m_I0

The different sets used by the algorithm.

m_iSet

int[] m_iSet

Index set {i: 0 < m_alpha[i] < C || 0 < m_alphaStar[i] < C}}

m_bUp

double m_bUp

b.up and b.low boundaries used to determine stopping criterion

m_bLow

double m_bLow

b.up and b.low boundaries used to determine stopping criterion

m_iUp

int m_iUp

index of the instance that gave us b.up and b.low

m_iLow

int m_iLow

index of the instance that gave us b.up and b.low

m_fTolerance

double m_fTolerance

tolerance parameter used for checking stopping criterion b.up < b.low + 2 tol

m_bUseVariant1

boolean m_bUseVariant1

set true to use variant 1 of the paper, otherwise use variant 2

Class weka.classifiers.functions.supportVector.SMOset extends java.lang.Object implements Serializable

serialVersionUID: -8364829283188675777L

Serialized Fields

m_number

int m_number

The current number of elements in the set

m_first

int m_first

The first element in the set

m_indicators

boolean[] m_indicators

Indicators

m_next

int[] m_next

The next element for each element

m_previous

int[] m_previous

The previous element for each element

Class weka.classifiers.functions.supportVector.StringKernel extends Kernel implements Serializable

serialVersionUID: -4902954211202690123L

Serialized Fields

m_cacheSize

int m_cacheSize

The size of the cache (a prime number)

m_internalCacheSize

int m_internalCacheSize

The size of the internal cache for intermediate results (a prime number)

m_strAttr

int m_strAttr

The attribute number of the string attribute

m_storage

double[] m_storage

Kernel cache (i.e., cache for kernel evaluations)

m_keys

long[] m_keys

m_kernelEvals

int m_kernelEvals

Counts the number of kernel evaluations.

m_numInsts

int m_numInsts

The number of instance in the dataset

m_PruningMethod

int m_PruningMethod

the pruning method

m_lambda

double m_lambda

the decay factor that penalizes non-continuous substring matches. See [1] for details.

m_subsequenceLength

int m_subsequenceLength

The substring length

m_maxSubsequenceLength

int m_maxSubsequenceLength

The maximum substring length for lambda pruning

m_powersOflambda

double[] m_powersOflambda

the precalculated powers of lambda

m_normalize

boolean m_normalize

flag for switching normalization on or off. This defaults to false and can be turned on by the switch for feature space normalization in SMO

maxCache

int maxCache

private cache for intermediate results

cachekh

double[] cachekh

cachekhK

int[] cachekhK

cachekh2

double[] cachekh2

cachekh2K

int[] cachekh2K

m_multX

int m_multX

cached indexes for private cache

m_multY

int m_multY

m_multZ

int m_multZ

m_multZZ

int m_multZZ

m_useRecursionCache

boolean m_useRecursionCache

Package weka.classifiers.lazy

Class weka.classifiers.lazy.IB1 extends Classifier implements Serializable

serialVersionUID: -6152184127304895851L

Serialized Fields

m_Train

Instances m_Train

The training instances used for classification.

m_MinArray

double[] m_MinArray

The minimum values for numeric attributes.

m_MaxArray

double[] m_MaxArray

The maximum values for numeric attributes.

Class weka.classifiers.lazy.IBk extends Classifier implements Serializable

serialVersionUID: -3080186098777067172L

Serialized Fields

m_Train

Instances m_Train

The training instances used for classification.

m_NumClasses

int m_NumClasses

The number of class values (or 1 if predicting numeric).

m_ClassType

int m_ClassType

The class attribute type.

m_kNN

int m_kNN

The number of neighbours to use for classification (currently).

m_kNNUpper

int m_kNNUpper

The value of kNN provided by the user. This may differ from m_kNN if cross-validation is being used.

m_kNNValid

boolean m_kNNValid

Whether the value of k selected by cross validation has been invalidated by a change in the training instances.

m_WindowSize

int m_WindowSize

The maximum number of training instances allowed. When this limit is reached, old training instances are removed, so the training data is "windowed". Set to 0 for unlimited numbers of instances.

m_DistanceWeighting

int m_DistanceWeighting

Whether the neighbours should be distance-weighted.

m_CrossValidate

boolean m_CrossValidate

Whether to select k by cross validation.

m_MeanSquared

boolean m_MeanSquared

Whether to minimise mean squared error rather than mean absolute error when cross-validating on numeric prediction tasks.

m_NNSearch

NearestNeighbourSearch m_NNSearch

for nearest-neighbor search.

m_NumAttributesUsed

double m_NumAttributesUsed

The number of attributes the contribute to a prediction.

Class weka.classifiers.lazy.KStar extends Classifier implements Serializable

serialVersionUID: 332458330800479083L

Serialized Fields

m_Train

Instances m_Train

The training instances used for classification.

m_NumInstances

int m_NumInstances

The number of instances in the dataset

m_NumClasses

int m_NumClasses

The number of class values

m_NumAttributes

int m_NumAttributes

The number of attributes

m_ClassType

int m_ClassType

The class attribute type

m_RandClassCols

int[][] m_RandClassCols

Table of random class value colomns

m_ComputeRandomCols

int m_ComputeRandomCols

Flag turning on and off the computation of random class colomns

m_InitFlag

int m_InitFlag

Flag turning on and off the initialisation of config variables

m_Cache

KStarCache[] m_Cache

A custom data structure for caching distinct attribute values and their scale factor or stop parameter.

m_MissingMode

int m_MissingMode

missing value treatment

m_BlendMethod

int m_BlendMethod

0 = use specified blend, 1 = entropic blend setting

m_GlobalBlend

int m_GlobalBlend

default sphere of influence blend setting

Class weka.classifiers.lazy.LBR extends Classifier implements Serializable

serialVersionUID: 5648559277738985156L

Serialized Fields

m_Counts

int[][][] m_Counts

All the counts for nominal attributes.

m_tCounts

int[][][] m_tCounts

All the counts for nominal attributes.

m_Priors

int[] m_Priors

The prior probabilities of the classes.

m_tPriors

int[] m_tPriors

The prior probabilities of the classes.

m_numAtts

int m_numAtts

number of attributes for the dataset

m_numClasses

int m_numClasses

number of classes for dataset

m_numInsts

int m_numInsts

number of instances in dataset

m_Instances

Instances m_Instances

The set of instances used for current training.

m_Errors

int m_Errors

leave-one-out errors on the training dataset.

m_ErrorFlags

boolean[] m_ErrorFlags

leave-one-out error flags on the training dataaet.

leftHand

java.util.ArrayList<E> leftHand

best attribute's index list. maybe as output result

m_subOldErrorFlags

boolean[] m_subOldErrorFlags

following is defined by wangzh, the number of instances to be classified incorrectly on the subset.

m_RemainderErrors

int m_RemainderErrors

the number of instances to be classified incorrectly besides the subset.

m_Number

int m_Number

the number of instance to be processed

m_NumberOfInstances

int m_NumberOfInstances

the Number of Instances to be used in building a classifiers

m_NCV

boolean m_NCV

for printing in n-fold cross validation

m_subInstances

LBR.Indexes m_subInstances

index of instances and attributes for the given dataset

tempSubInstances

LBR.Indexes tempSubInstances

index of instances and attributes for the given dataset

posteriorsArray

double[] posteriorsArray

probability values array

bestCnt

int bestCnt

tempCnt

int tempCnt

forCnt

int forCnt

whileCnt

int whileCnt

Class weka.classifiers.lazy.LBR.Indexes extends java.lang.Object implements Serializable

serialVersionUID: -2771490019751421307L

Serialized Fields

m_InstIndexes

boolean[] m_InstIndexes

the array instance indexes

m_AttIndexes

boolean[] m_AttIndexes

the array attribute indexes

m_NumInstances

int m_NumInstances

the number of instances indexed

m_NumAtts

int m_NumAtts

the number of attributes indexed

m_SequentialInstIndexes

int[] m_SequentialInstIndexes

the array of instance indexes that are set to a either true or false

m_SequentialAttIndexes

int[] m_SequentialAttIndexes

an array of attribute indexes that are set to either true or false

m_SequentialInstanceIndex_valid

boolean m_SequentialInstanceIndex_valid

flag to check if sequential array must be rebuilt due to changes to the instance index

m_SequentialAttIndex_valid

boolean m_SequentialAttIndex_valid

flag to check if sequential array must be rebuilt due to changes to the attribute index

m_NumInstsSet

int m_NumInstsSet

the number of instances "in use" or set to a the original value (true or false)

m_NumAttsSet

int m_NumAttsSet

the number of attributes "in use" or set to a the original value (true or false)

m_NumSeqInstsSet

int m_NumSeqInstsSet

the number of sequential instances "in use" or set to a the original value (true or false)

m_NumSeqAttsSet

int m_NumSeqAttsSet

the number of sequential attributes "in use" or set to a the original value (true or false)

m_ClassIndex

int m_ClassIndex

the Class Index for the data set

Class weka.classifiers.lazy.LWL extends SingleClassifierEnhancer implements Serializable

serialVersionUID: 1979797405383665815L

Serialized Fields

m_Train

Instances m_Train

The training instances used for classification.

m_kNN

int m_kNN

The number of neighbours used to select the kernel bandwidth.

m_WeightKernel

int m_WeightKernel

The weighting kernel method currently selected.

m_UseAllK

boolean m_UseAllK

True if m_kNN should be set to all instances.

m_NNSearch

NearestNeighbourSearch m_NNSearch

The nearest neighbour search algorithm to use. (Default: weka.core.neighboursearch.LinearNNSearch)

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data.

Package weka.classifiers.lazy.kstar

Class weka.classifiers.lazy.kstar.KStarCache extends java.lang.Object implements Serializable

serialVersionUID: -7693632394267140678L

Serialized Fields

m_Cache

KStarCache.CacheTable m_Cache

cache table

Class weka.classifiers.lazy.kstar.KStarCache.CacheTable extends java.lang.Object implements Serializable

serialVersionUID: -8086106452588253423L

Serialized Fields

m_Table

KStarCache.TableEntry[] m_Table

The hash table data.

m_Count

int m_Count

The total number of entries in the hash table.

m_Threshold

int m_Threshold

Rehashes the table when count exceeds this threshold.

m_LoadFactor

float m_LoadFactor

The load factor for the hashtable.

DEFAULT_TABLE_SIZE

int DEFAULT_TABLE_SIZE

The default size of the hashtable

DEFAULT_LOAD_FACTOR

float DEFAULT_LOAD_FACTOR

The default load factor for the hashtable

EPSILON

double EPSILON

Accuracy value for equality

Class weka.classifiers.lazy.kstar.KStarCache.TableEntry extends java.lang.Object implements Serializable

serialVersionUID: 4057602386766259138L

Serialized Fields

hash

int hash

attribute value hash code

key

double key

attribute value

value

double value

scale factor or stop parameter

pmiss

double pmiss

transformation probability to missing value

next

KStarCache.TableEntry next

next table entry (separate chaining)

Package weka.classifiers.meta

Class weka.classifiers.meta.AdaBoostM1 extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -7378107808933117974L

Serialized Fields

m_Betas

double[] m_Betas

Array for storing the weights for the votes.

m_NumIterationsPerformed

int m_NumIterationsPerformed

The number of successfully generated base classifiers.

m_WeightThreshold

int m_WeightThreshold

Weight Threshold. The percentage of weight mass used in training

m_UseResampling

boolean m_UseResampling

Use boosting with reweighting?

m_NumClasses

int m_NumClasses

The number of classes

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

Class weka.classifiers.meta.AdditiveRegression extends IteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -2368937577670527151L

Serialized Fields

m_shrinkage

double m_shrinkage

Shrinkage (Learning rate). Default = no shrinkage.

m_NumIterationsPerformed

int m_NumIterationsPerformed

The number of successfully generated base classifiers.

m_zeroR

ZeroR m_zeroR

The model for the mean

m_SuitableData

boolean m_SuitableData

whether we have suitable data or nor (if not, ZeroR model is used)

Class weka.classifiers.meta.AttributeSelectedClassifier extends SingleClassifierEnhancer implements Serializable

serialVersionUID: -5951805453487947577L

Serialized Fields

m_AttributeSelection

AttributeSelection m_AttributeSelection

The attribute selection object

m_Evaluator

ASEvaluation m_Evaluator

The attribute evaluator to use

m_Search

ASSearch m_Search

The search method to use

m_ReducedHeader

Instances m_ReducedHeader

The header of the dimensionally reduced data

m_numClasses

int m_numClasses

The number of class vals in the training data (1 if class is numeric)

m_numAttributesSelected

double m_numAttributesSelected

The number of attributes selected by the attribute selection phase

m_selectionTime

double m_selectionTime

The time taken to select attributes in milliseconds

m_totalTime

double m_totalTime

The time taken to select attributes AND build the classifier

Class weka.classifiers.meta.Bagging extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -505879962237199703L

Serialized Fields

m_BagSizePercent

int m_BagSizePercent

The size of each bag sample, as a percentage of the training size

m_CalcOutOfBag

boolean m_CalcOutOfBag

Whether to calculate the out of bag error

m_OutOfBagError

double m_OutOfBagError

The out of bag error that has been calculated

Class weka.classifiers.meta.ClassificationViaClustering extends Classifier implements Serializable

serialVersionUID: -5687069451420259135L

Serialized Fields

m_Clusterer

Clusterer m_Clusterer

the cluster algorithm used (template)

m_ActualClusterer

Clusterer m_ActualClusterer

the actual cluster algorithm being used

m_OriginalHeader

Instances m_OriginalHeader

the original training data header

m_ClusteringHeader

Instances m_ClusteringHeader

the modified training data header

m_ClustersToClasses

double[] m_ClustersToClasses

the mapping between clusters and classes

m_ZeroR

Classifier m_ZeroR

the default model

Class weka.classifiers.meta.ClassificationViaRegression extends SingleClassifierEnhancer implements Serializable

serialVersionUID: 4500023123618669859L

Serialized Fields

m_Classifiers

Classifier[] m_Classifiers

The classifiers. (One for each class.)

m_ClassFilters

MakeIndicator[] m_ClassFilters

The filters used to transform the class.

Class weka.classifiers.meta.CostSensitiveClassifier extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: -720658209263002404L

Serialized Fields

m_MatrixSource

int m_MatrixSource

Indicates the current cost matrix source

m_OnDemandDirectory

java.io.File m_OnDemandDirectory

The directory used when loading cost files on demand, null indicates current directory

m_CostFile

java.lang.String m_CostFile

The name of the cost file, for command line options

m_CostMatrix

CostMatrix m_CostMatrix

The cost matrix

m_MinimizeExpectedCost

boolean m_MinimizeExpectedCost

True if the costs should be used by selecting the minimum expected cost (false means weight training data by the costs)

Class weka.classifiers.meta.CVParameterSelection extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: -6529603380876641265L

Serialized Fields

m_ClassifierOptions

java.lang.String[] m_ClassifierOptions

The base classifier options (not including those being set by cross-validation)

m_BestClassifierOptions

java.lang.String[] m_BestClassifierOptions

The set of all classifier options as determined by cross-validation

m_InitOptions

java.lang.String[] m_InitOptions

The set of all options at initialization time. So that getOptions can return this.

m_BestPerformance

double m_BestPerformance

The cross-validated performance of the best options

m_CVParams

FastVector m_CVParams

The set of parameters to cross-validate over

m_NumAttributes

int m_NumAttributes

The number of attributes in the data

m_TrainFoldSize

int m_TrainFoldSize

The number of instances in a training fold

m_NumFolds

int m_NumFolds

The number of folds used in cross-validation

Class weka.classifiers.meta.CVParameterSelection.CVParameter extends java.lang.Object implements Serializable

serialVersionUID: -4668812017709421953L

Serialized Fields

m_ParamChar

char m_ParamChar

Char used to identify the option of interest

m_Lower

double m_Lower

Lower bound for the CV search

m_Upper

double m_Upper

Upper bound for the CV search

m_Steps

double m_Steps

Number of steps during the search

m_ParamValue

double m_ParamValue

The parameter value with the best performance

m_AddAtEnd

boolean m_AddAtEnd

True if the parameter should be added at the end of the argument list

m_RoundParam

boolean m_RoundParam

True if the parameter should be rounded to an integer

Class weka.classifiers.meta.Dagging extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: 4560165876570074309L

Serialized Fields

m_NumFolds

int m_NumFolds

the number of folds to use to split the training data

m_Vote

Vote m_Vote

the classifier used for voting

m_Verbose

boolean m_Verbose

whether to output some progress information during building

Class weka.classifiers.meta.Decorate extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -6020193348750269931L

Serialized Fields

m_Committee

java.util.Vector<E> m_Committee

Vector of classifiers that make up the committee/ensemble.

m_DesiredSize

int m_DesiredSize

The desired ensemble size.

m_ArtSize

double m_ArtSize

Amount of artificial/random instances to use - specified as a fraction of the training data size.

m_Random

java.util.Random m_Random

The random number generator.

m_AttributeStats

java.util.Vector<E> m_AttributeStats

Attribute statistics - used for generating artificial examples.

Class weka.classifiers.meta.END extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -4143242362912214956L

Serialized Fields

m_hashtable

java.util.Hashtable<K,V> m_hashtable

The hashtable containing the classifiers for the END.

Class weka.classifiers.meta.EnsembleSelection extends RandomizableClassifier implements Serializable

serialVersionUID: -1744155148765058511L

Serialized Fields

m_library

EnsembleSelectionLibrary m_library

The Library of models, from which we can select our ensemble. Usually loaded from a model list file (.mlf or .model.xml) using the -L command-line option.

m_chosen_models

EnsembleSelectionLibraryModel[] m_chosen_models

List of models chosen by EnsembleSelection. Populated by buildClassifier.

m_chosen_model_weights

int[] m_chosen_model_weights

An array of weights for the chosen models. Elements are parallel to those in m_chosen_models. That is, m_chosen_model_weights[i] is the weight associated with the model at m_chosen_models[i].

m_total_weight

int m_total_weight

Total weight of all chosen models.

m_modelRatio

double m_modelRatio

ratio of library models that will be randomly chosen to be used for each model bag

m_validationRatio

double m_validationRatio

Indicates the fraction of the given training set that should be used for hillclimbing/validation. This fraction is set aside and not used for training. It is assumed that any loaded models were also not trained on set-aside data. (If the same percentage and random seed were used previously to train the models in the library, this will work as expected - i.e., those models will be valid)

m_modelLibraryFileName

java.lang.String m_modelLibraryFileName

The name of the Model Library File (if one is specified) which lists models from which ensemble selection will choose. This is only used when run from the command-line, as otherwise m_library is responsible for this.

m_numModelBags

int m_numModelBags

The number of "model bags". Using 1 is equivalent to no bagging at all.

m_hillclimbMetric

int m_hillclimbMetric

The metric for which the ensemble will be optimized.

m_algorithm

int m_algorithm

The algorithm used for ensemble selection.

m_hillclimbIterations

int m_hillclimbIterations

number of hillclimbing iterations for the ensemble selection algorithm

m_sortInitializationRatio

double m_sortInitializationRatio

ratio of library models to be used for sort initialization

m_replacement

boolean m_replacement

specifies whether or not the ensemble algorithm is allowed to include a specific model in the library more than once in each ensemble

m_greedySortInitialization

boolean m_greedySortInitialization

specifies whether we use "greedy" sort initialization. If false, we simply add the best m_sortInitializationRatio models of the bag blindly. If true, we add the best models in order up to m_sortInitializationRatio until adding the next model would not help performance.

m_verboseOutput

boolean m_verboseOutput

Specifies whether or not we will output metrics for all models

m_cachedPredictions

java.util.Map<K,V> m_cachedPredictions

Hash map of cached predictions. The key is a stringified Instance. Each entry is a 2d array, first indexed by classifier index (i.e., the one used in m_chosen_model). The second index is the usual "distribution" index across classes.

m_workingDirectory

java.io.File m_workingDirectory

This string will store the working directory where all models , temporary prediction values, and modellist logs are to be built and stored.

m_NumFolds

int m_NumFolds

Indicates the number of folds for cross-validation. A value of 1 indicates there is no cross-validation. Cross validation is done in the "embedded" fashion described by Caruana, Niculescu, and Munson (unpublished work - tech report forthcoming)

Class weka.classifiers.meta.FilteredClassifier extends SingleClassifierEnhancer implements Serializable

serialVersionUID: -4523450618538717400L

Serialized Fields

m_Filter

Filter m_Filter

The filter

m_FilteredInstances

Instances m_FilteredInstances

The instance structure of the filtered instances

Class weka.classifiers.meta.Grading extends Stacking implements Serializable

serialVersionUID: 5207837947890081170L

Serialized Fields

m_MetaClassifiers

Classifier[] m_MetaClassifiers

The meta classifiers, one for each base classifier.

m_InstPerClass

double[] m_InstPerClass

InstPerClass

Class weka.classifiers.meta.GridSearch extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: -3034773968581595348L

Serialized Fields

m_Filter

Filter m_Filter

the Filter

m_BestFilter

Filter m_BestFilter

the Filter with the best setup

m_BestClassifier

Classifier m_BestClassifier

the Classifier with the best setup

m_Values

weka.classifiers.meta.GridSearch.PointDouble m_Values

the best values

m_Evaluation

int m_Evaluation

the type of evaluation

m_Y_Property

java.lang.String m_Y_Property

the Y option to work on (without leading dash, preceding 'classifier.' means to set the option for the classifier 'filter.' for the filter)

m_Y_Min

double m_Y_Min

the minimum of Y

m_Y_Max

double m_Y_Max

the maximum of Y

m_Y_Step

double m_Y_Step

the step size of Y

m_Y_Base

double m_Y_Base

the base for Y

m_Y_Expression

java.lang.String m_Y_Expression

The expression for the Y property. Available parameters for the expression:

• BASE
• FROM (= min)
• TO (= max)
• STEP
• I - the current value (from 'from' to 'to' with stepsize 'step')

See Also:

MathematicalExpression, MathExpression

m_X_Property

java.lang.String m_X_Property

the X option to work on (without leading dash, preceding 'classifier.' means to set the option for the classifier 'filter.' for the filter)

m_X_Min

double m_X_Min

the minimum of X

m_X_Max

double m_X_Max

the maximum of X

m_X_Step

double m_X_Step

the step size of

m_X_Base

double m_X_Base

the base for

m_X_Expression

java.lang.String m_X_Expression

The expression for the X property. Available parameters for the expression:

• BASE
• FROM (= min)
• TO (= max)
• STEP
• I - the current value (from 'from' to 'to' with stepsize 'step')

See Also:

MathematicalExpression, MathExpression

m_GridIsExtendable

boolean m_GridIsExtendable

whether the grid can be extended

m_MaxGridExtensions

int m_MaxGridExtensions

maximum number of grid extensions (-1 means unlimited)

m_GridExtensionsPerformed

int m_GridExtensionsPerformed

the number of extensions performed

m_SampleSize

double m_SampleSize

the sample size to search the initial grid with

m_Traversal

int m_Traversal

the traversal

m_LogFile

java.io.File m_LogFile

the log file to use

m_Grid

weka.classifiers.meta.GridSearch.Grid m_Grid

the value-pairs grid

m_Data

Instances m_Data

the training data

m_Cache

weka.classifiers.meta.GridSearch.PerformanceCache m_Cache

the cache for points in the grid that got calculated

m_UniformPerformance

boolean m_UniformPerformance

whether all performances in the grid are the same

Class weka.classifiers.meta.GridSearch.Grid extends java.lang.Object implements Serializable

serialVersionUID: 7290732613611243139L

Serialized Fields

m_MinX

double m_MinX

the minimum on the X axis

m_MaxX

double m_MaxX

the maximum on the X axis

m_StepX

double m_StepX

the step size for the X axis

m_LabelX

java.lang.String m_LabelX

the label for the X axis

m_MinY

double m_MinY

the minimum on the Y axis

m_MaxY

double m_MaxY

the maximum on the Y axis

m_StepY

double m_StepY

the step size for the Y axis

m_LabelY

java.lang.String m_LabelY

the label for the Y axis

m_Width

int m_Width

the number of points on the X axis

m_Height

int m_Height

the number of points on the Y axis

Class weka.classifiers.meta.GridSearch.Performance extends java.lang.Object implements Serializable

serialVersionUID: -4374706475277588755L

Serialized Fields

m_Values

weka.classifiers.meta.GridSearch.PointDouble m_Values

the value pair the classifier was built with

m_CC

double m_CC

the Correlation coefficient

m_RMSE

double m_RMSE

the Root mean squared error

m_RRSE

double m_RRSE

the Root relative squared error

m_MAE

double m_MAE

the Mean absolute error

m_RAE

double m_RAE

the Relative absolute error

m_ACC

double m_ACC

the Accuracy

m_Kappa

double m_Kappa

the kappa value

Class weka.classifiers.meta.GridSearch.PerformanceCache extends java.lang.Object implements Serializable

serialVersionUID: 5838863230451530252L

Serialized Fields

m_Cache

java.util.Hashtable<K,V> m_Cache

the cache for points in the grid that got calculated

Class weka.classifiers.meta.GridSearch.PerformanceComparator extends java.lang.Object implements Serializable

serialVersionUID: 6507592831825393847L

Serialized Fields

m_Evaluation

int m_Evaluation

the performance measure to use for comparison

See Also:

GridSearch.TAGS_EVALUATION

Class weka.classifiers.meta.GridSearch.PerformanceTable extends java.lang.Object implements Serializable

serialVersionUID: 5486491313460338379L

Serialized Fields

m_Grid

weka.classifiers.meta.GridSearch.Grid m_Grid

the corresponding grid

m_Performances

java.util.Vector<E> m_Performances

the performances

m_Type

int m_Type

the type of performance the table was generated for

m_Table

double[][] m_Table

the table with the values

m_Min

double m_Min

the minimum performance

m_Max

double m_Max

the maximum performance

Class weka.classifiers.meta.GridSearch.PointDouble extends java.awt.geom.Point2D.Double implements Serializable

serialVersionUID: 7151661776161898119L

Class weka.classifiers.meta.GridSearch.PointInt extends java.awt.Point implements Serializable

serialVersionUID: -5900415163698021618L

Class weka.classifiers.meta.LogitBoost extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -3905660358715833753L

Serialized Fields

m_Classifiers

Classifier[][] m_Classifiers

Array for storing the generated base classifiers. Note: we are hiding the variable from IteratedSingleClassifierEnhancer

m_NumClasses

int m_NumClasses

The number of classes

m_NumGenerated

int m_NumGenerated

The number of successfully generated base classifiers.

m_NumFolds

int m_NumFolds

The number of folds for the internal cross-validation.

m_NumRuns

int m_NumRuns

The number of runs for the internal cross-validation.

m_WeightThreshold

int m_WeightThreshold

Weight thresholding. The percentage of weight mass used in training

m_NumericClassData

Instances m_NumericClassData

Dummy dataset with a numeric class

m_ClassAttribute

Attribute m_ClassAttribute

The actual class attribute (for getting class names)

m_UseResampling

boolean m_UseResampling

Use boosting with reweighting?

m_Precision

double m_Precision

The threshold on the improvement of the likelihood

m_Shrinkage

double m_Shrinkage

The value of the shrinkage parameter

m_RandomInstance

java.util.Random m_RandomInstance

The random number generator used

m_Offset

double m_Offset

The value by which the actual target value for the true class is offset.

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

Class weka.classifiers.meta.MetaCost extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: 1205317833344726855L

Serialized Fields

m_MatrixSource

int m_MatrixSource

Indicates the current cost matrix source

m_OnDemandDirectory

java.io.File m_OnDemandDirectory

The directory used when loading cost files on demand, null indicates current directory

m_CostFile

java.lang.String m_CostFile

The name of the cost file, for command line options

m_CostMatrix

CostMatrix m_CostMatrix

The cost matrix

m_NumIterations

int m_NumIterations

The number of iterations.

m_BagSizePercent

int m_BagSizePercent

The size of each bag sample, as a percentage of the training size

Class weka.classifiers.meta.MultiBoostAB extends AdaBoostM1 implements Serializable

serialVersionUID: -6681619178187935148L

Serialized Fields

m_NumSubCmtys

int m_NumSubCmtys

The number of sub-committees to use

m_Random

java.util.Random m_Random

Random number generator

Class weka.classifiers.meta.MultiClassClassifier extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: -3879602011542849141L

Serialized Fields

m_Classifiers

Classifier[] m_Classifiers

The classifiers.

m_pairwiseCoupling

boolean m_pairwiseCoupling

Use pairwise coupling with 1-vs-1

m_SumOfWeights

double[] m_SumOfWeights

Needed for pairwise coupling

m_ClassFilters

Filter[] m_ClassFilters

The filters used to transform the class.

m_ZeroR

ZeroR m_ZeroR

ZeroR classifier for when all base classifier return zero probability.

m_ClassAttribute

Attribute m_ClassAttribute

Internal copy of the class attribute for output purposes

m_TwoClassDataset

Instances m_TwoClassDataset

A transformed dataset header used by the 1-against-1 method

m_RandomWidthFactor

double m_RandomWidthFactor

The multiplier when generating random codes. Will generate numClasses * m_RandomWidthFactor codes

m_Method

int m_Method

The multiclass method to use

Class weka.classifiers.meta.MultiScheme extends RandomizableMultipleClassifiersCombiner implements Serializable

serialVersionUID: 5710744346128957520L

Serialized Fields

m_Classifier

Classifier m_Classifier

The classifier that had the best performance on training data.

m_ClassifierIndex

int m_ClassifierIndex

The index into the vector for the selected scheme

m_NumXValFolds

int m_NumXValFolds

Number of folds to use for cross validation (0 means use training error for selection)

Class weka.classifiers.meta.OrdinalClassClassifier extends SingleClassifierEnhancer implements Serializable

serialVersionUID: -3461971774059603636L

Serialized Fields

m_Classifiers

Classifier[] m_Classifiers

The classifiers. (One for each class.)

m_ClassFilters

MakeIndicator[] m_ClassFilters

The filters used to transform the class.

m_ZeroR

ZeroR m_ZeroR

ZeroR classifier for when all base classifier return zero probability.

m_UseSmoothing

boolean m_UseSmoothing

Whether to use smoothing to prevent negative "probabilities".

Class weka.classifiers.meta.RacedIncrementalLogitBoost extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: 908598343772170052L

Serialized Fields

m_committees

FastVector m_committees

The committees

m_PruningType

int m_PruningType

The pruning type used

m_UseResampling

boolean m_UseResampling

Whether to use resampling

m_NumClasses

int m_NumClasses

The number of classes

m_NumericClassData

Instances m_NumericClassData

Dummy dataset with a numeric class

m_ClassAttribute

Attribute m_ClassAttribute

The actual class attribute (for getting class names)

m_minChunkSize

int m_minChunkSize

The minimum chunk size used for training

m_maxChunkSize

int m_maxChunkSize

The maimum chunk size used for training

m_validationChunkSize

int m_validationChunkSize

The size of the validation set

m_numInstancesConsumed

int m_numInstancesConsumed

The number of instances consumed

m_validationSet

Instances m_validationSet

The instances used for validation

m_currentSet

Instances m_currentSet

The instances currently in memory for training

m_bestCommittee

weka.classifiers.meta.RacedIncrementalLogitBoost.Committee m_bestCommittee

The current best committee

m_zeroR

ZeroR m_zeroR

The default scheme used when committees aren't ready

m_validationSetChanged

boolean m_validationSetChanged

Whether the validation set has recently been changed

m_maxBatchSizeRequired

int m_maxBatchSizeRequired

The maximum number of instances required for processing

m_RandomInstance

java.util.Random m_RandomInstance

The random number generator used

Class weka.classifiers.meta.RacedIncrementalLogitBoost.Committee extends java.lang.Object implements Serializable

serialVersionUID: 5559880306684082199L

Serialized Fields

m_chunkSize

int m_chunkSize

m_instancesConsumed

int m_instancesConsumed

number eaten from m_currentSet

m_models

FastVector m_models

m_lastValidationError

double m_lastValidationError

m_lastLogLikelihood

double m_lastLogLikelihood

m_modelHasChanged

boolean m_modelHasChanged

m_modelHasChangedLL

boolean m_modelHasChangedLL

m_validationFs

double[][] m_validationFs

m_newValidationFs

double[][] m_newValidationFs

Class weka.classifiers.meta.RandomCommittee extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -9204394360557300092L

Class weka.classifiers.meta.RandomSubSpace extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: 1278172513912424947L

Serialized Fields

m_SubSpaceSize

double m_SubSpaceSize

The size of each bag sample, as a percentage of the training size

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

Class weka.classifiers.meta.RegressionByDiscretization extends SingleClassifierEnhancer implements Serializable

serialVersionUID: 5066426153134050378L

Serialized Fields

m_Discretizer

Discretize m_Discretizer

The discretization filter.

m_NumBins

int m_NumBins

The number of discretization intervals.

m_ClassMeans

double[] m_ClassMeans

The mean values for each Discretized class interval.

m_DeleteEmptyBins

boolean m_DeleteEmptyBins

Whether to delete empty intervals.

m_DiscretizedHeader

Instances m_DiscretizedHeader

Header of discretized data.

m_UseEqualFrequency

boolean m_UseEqualFrequency

Use equal-frequency binning

Class weka.classifiers.meta.RotationForest extends RandomizableIteratedSingleClassifierEnhancer implements Serializable

serialVersionUID: -3255631880798499936L

Serialized Fields

m_MinGroup

int m_MinGroup

The minimum size of a group

m_MaxGroup

int m_MaxGroup

The maximum size of a group

m_NumberOfGroups

boolean m_NumberOfGroups

Whether minGroup and maxGroup refer to the number of groups or their size

m_RemovedPercentage

int m_RemovedPercentage

The percentage of instances to be removed

m_Groups

int[][][] m_Groups

The attributes of each group

m_ProjectionFilter

Filter m_ProjectionFilter

The type of projection filter

m_ProjectionFilters

Filter[][] m_ProjectionFilters

The projection filters

m_Headers

Instances[] m_Headers

Headers of the transformed dataset

m_ReducedHeaders

Instances[][] m_ReducedHeaders

Headers of the reduced datasets

m_RemoveUseless

RemoveUseless m_RemoveUseless

Filter that remove useless attributes

m_Normalize

Normalize m_Normalize

Filter that normalized the attributes

Class weka.classifiers.meta.Stacking extends RandomizableMultipleClassifiersCombiner implements Serializable

serialVersionUID: 5134738557155845452L

Serialized Fields

m_MetaClassifier

Classifier m_MetaClassifier

The meta classifier

m_MetaFormat

Instances m_MetaFormat

Format for meta data

m_BaseFormat

Instances m_BaseFormat

Format for base data

m_NumFolds

int m_NumFolds

Set the number of folds for the cross-validation

Class weka.classifiers.meta.StackingC extends Stacking implements Serializable

serialVersionUID: -6717545616603725198L

Serialized Fields

m_MetaClassifiers

Classifier[] m_MetaClassifiers

The meta classifiers (one for each class, like in ClassificationViaRegression)

m_attrFilter

Remove m_attrFilter

Filter to transform metaData - Remove

m_makeIndicatorFilter

MakeIndicator m_makeIndicatorFilter

Filter to transform metaData - MakeIndicator

Class weka.classifiers.meta.ThresholdSelector extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: -1795038053239867444L

Serialized Fields

m_HighThreshold

double m_HighThreshold

The upper threshold used as the basis of correction

m_LowThreshold

double m_LowThreshold

The lower threshold used as the basis of correction

m_BestThreshold

double m_BestThreshold

The threshold that lead to the best performance

m_BestValue

double m_BestValue

The best value that has been observed

m_NumXValFolds

int m_NumXValFolds

The number of folds used in cross-validation

m_DesignatedClass

int m_DesignatedClass

Designated class value, determined during building

m_ClassMode

int m_ClassMode

Method to determine which class to optimize for

m_EvalMode

int m_EvalMode

The evaluation mode

m_RangeMode

int m_RangeMode

The range correction mode

m_nMeasure

int m_nMeasure

evaluation measure used for determining threshold

m_manualThreshold

boolean m_manualThreshold

True if a manually set threshold is being used

m_manualThresholdValue

double m_manualThresholdValue

-1 = not used by default

Class weka.classifiers.meta.Vote extends RandomizableMultipleClassifiersCombiner implements Serializable

serialVersionUID: -637891196294399624L

Serialized Fields

m_CombinationRule

int m_CombinationRule

Combination Rule variable

m_Random

java.util.Random m_Random

the random number generator used for breaking ties in majority voting

See Also:

Vote.distributionForInstanceMajorityVoting(Instance)

Package weka.classifiers.meta.ensembleSelection

Class weka.classifiers.meta.ensembleSelection.EnsembleModelMismatchException extends java.lang.Exception implements Serializable

serialVersionUID: 4660917211181280739L

Class weka.classifiers.meta.ensembleSelection.EnsembleSelectionLibrary extends EnsembleLibrary implements Serializable

serialVersionUID: -6444026512552917835L

Serialized Fields

m_workingDirectory

java.io.File m_workingDirectory

the working ensemble library directory.

m_modelListFile

java.lang.String m_modelListFile

tha name of the model list file storing the list of models currently being used by the model library

m_trainingData

Instances[] m_trainingData

the training data used to build the library. One per fold.

m_hillclimbData

Instances[] m_hillclimbData

the test data used for hillclimbing. One per fold.

m_predictions

double[][][] m_predictions

the predictions of each model. Built by trainAll. First index is for the model. Second is for the instance. third is for the class (we use distributionForInstance).

m_seed

int m_seed

the random seed used to partition the training data into validation and training folds

m_folds

int m_folds

the number of folds

m_validationRatio

double m_validationRatio

the ratio of validation data used to train the model

Class weka.classifiers.meta.ensembleSelection.EnsembleSelectionLibraryModel extends EnsembleLibraryModel implements Serializable

serialVersionUID: -6426075459862947640L

Serialized Fields

m_models

Classifier[] m_models

the models

m_seed

int m_seed

The seed that was used to create this model

m_checksum

java.lang.String m_checksum

The checksum of the instances.arff object that was used to create this model

m_validationRatio

double m_validationRatio

The validation ratio that was used to create this model

m_folds

int m_folds

The number of folds, or number of CV models that was used to create this "model"

m_fileName

java.lang.String m_fileName

The .elm file name that this model should be saved/loaded to/from

m_validationPredictions

double[][] m_validationPredictions

the validation predictions of this model. First index for the instance. third is for the class (we use distributionForInstance).

Package weka.classifiers.meta.nestedDichotomies

Class weka.classifiers.meta.nestedDichotomies.ClassBalancedND extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: 5944063630650811903L

Serialized Fields

m_FilteredClassifier

FilteredClassifier m_FilteredClassifier

The filtered classifier in which the base classifier is wrapped.

m_classifiers

java.util.Hashtable<K,V> m_classifiers

The hashtable for this node.

m_FirstSuccessor

ClassBalancedND m_FirstSuccessor

The first successor

m_SecondSuccessor

ClassBalancedND m_SecondSuccessor

The second successor

m_Range

Range m_Range

The classes that are grouped together at the current node

m_hashtablegiven

boolean m_hashtablegiven

Is Hashtable given from END?

Class weka.classifiers.meta.nestedDichotomies.DataNearBalancedND extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: 5117477294209496368L

Serialized Fields

m_FilteredClassifier

FilteredClassifier m_FilteredClassifier

The filtered classifier in which the base classifier is wrapped.

m_classifiers

java.util.Hashtable<K,V> m_classifiers

The hashtable for this node.

m_FirstSuccessor

DataNearBalancedND m_FirstSuccessor

The first successor

m_SecondSuccessor

DataNearBalancedND m_SecondSuccessor

The second successor

m_Range

Range m_Range

The classes that are grouped together at the current node

m_hashtablegiven

boolean m_hashtablegiven

Is Hashtable given from END?

Class weka.classifiers.meta.nestedDichotomies.ND extends RandomizableSingleClassifierEnhancer implements Serializable

serialVersionUID: -6355893369855683820L

Serialized Fields

m_ndtree

weka.classifiers.meta.nestedDichotomies.ND.NDTree m_ndtree

The tree of classes

m_classifiers

java.util.Hashtable<K,V> m_classifiers

The hashtable containing all the classifiers

m_hashtablegiven

boolean m_hashtablegiven

Is Hashtable given from END?

Class weka.classifiers.meta.nestedDichotomies.ND.NDTree extends java.lang.Object implements Serializable

serialVersionUID: 4284655952754474880L

Serialized Fields

m_indices

FastVector m_indices

The indices associated with this node

m_parent

weka.classifiers.meta.nestedDichotomies.ND.NDTree m_parent

The parent

m_left

weka.classifiers.meta.nestedDichotomies.ND.NDTree m_left

The left successor

m_right

weka.classifiers.meta.nestedDichotomies.ND.NDTree m_right

The right successor

Package weka.classifiers.mi

Class weka.classifiers.mi.CitationKNN extends Classifier implements Serializable

serialVersionUID: -8435377743874094852L

Serialized Fields

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_NumClasses

int m_NumClasses

The number of the class labels

m_IdIndex

int m_IdIndex

m_Debug

boolean m_Debug

Debugging output

m_Classes

int[] m_Classes

Class labels for each bag

m_Attributes

Instances m_Attributes

attribute name structure of the relational attribute

m_NumReferences

int m_NumReferences

Number of references

m_NumCiters

int m_NumCiters

Number of citers

m_TrainBags

Instances m_TrainBags

Training bags

m_CNNDebug

boolean m_CNNDebug

Different debugging output

m_CitersDebug

boolean m_CitersDebug

m_ReferencesDebug

boolean m_ReferencesDebug

m_HDistanceDebug

boolean m_HDistanceDebug

m_NeighborListDebug

boolean m_NeighborListDebug

m_CNN

weka.classifiers.mi.CitationKNN.NeighborList[] m_CNN

C nearest neighbors considering all the bags

m_Citers

int[] m_Citers

C nearest citers

m_References

int[] m_References

R nearest references

m_HDRank

int m_HDRank

Rank associated to the Hausdorff distance

m_Diffs

double[] m_Diffs

Normalization of the euclidean distance

m_Min

double[] m_Min

m_MinNorm

double m_MinNorm

m_Max

double[] m_Max

m_MaxNorm

double m_MaxNorm

Class weka.classifiers.mi.MDD extends Classifier implements Serializable

serialVersionUID: -7273119490545290581L

Serialized Fields

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Par

double[] m_Par

m_NumClasses

int m_NumClasses

The number of the class labels

m_Classes

int[] m_Classes

Class labels for each bag

m_Data

double[][][] m_Data

MI data

m_Attributes

Instances m_Attributes

All attribute names

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_filterType

int m_filterType

Whether to normalize/standardize/neither, default:standardize

m_Missing

ReplaceMissingValues m_Missing

The filter used to get rid of missing values.

Class weka.classifiers.mi.MIBoost extends SingleClassifierEnhancer implements Serializable

serialVersionUID: -3808427225599279539L

Serialized Fields

m_Models

Classifier[] m_Models

the models for the iterations

m_NumClasses

int m_NumClasses

The number of the class labels

m_Classes

int[] m_Classes

Class labels for each bag

m_Attributes

Instances m_Attributes

attributes name for the new dataset used to build the model

m_NumIterations

int m_NumIterations

Number of iterations

m_Beta

double[] m_Beta

Voting weights of models

m_MaxIterations

int m_MaxIterations

the maximum number of boost iterations

m_DiscretizeBin

int m_DiscretizeBin

the number of discretization bins

m_Filter

Discretize m_Filter

filter used for discretization

m_ConvertToSI

MultiInstanceToPropositional m_ConvertToSI

filter used to convert the MI dataset into single-instance dataset

Class weka.classifiers.mi.MIDD extends Classifier implements Serializable

serialVersionUID: 4263507733600536168L

Serialized Fields

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Par

double[] m_Par

m_NumClasses

int m_NumClasses

The number of the class labels

m_Classes

int[] m_Classes

Class labels for each bag

m_Data

double[][][] m_Data

MI data

m_Attributes

Instances m_Attributes

All attribute names

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_filterType

int m_filterType

Whether to normalize/standardize/neither, default:standardize

m_Missing

ReplaceMissingValues m_Missing

The filter used to get rid of missing values.

Class weka.classifiers.mi.MIEMDD extends RandomizableClassifier implements Serializable

serialVersionUID: 3899547154866223734L

Serialized Fields

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Par

double[] m_Par

m_NumClasses

int m_NumClasses

The number of the class labels

m_Classes

int[] m_Classes

Class labels for each bag

m_Data

double[][][] m_Data

MI data

m_Attributes

Instances m_Attributes

All attribute names

m_emData

double[][] m_emData

MI data

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_filterType

int m_filterType

Whether to normalize/standardize/neither, default:standardize

m_Missing

ReplaceMissingValues m_Missing

The filter used to get rid of missing values.

Class weka.classifiers.mi.MILR extends Classifier implements Serializable

serialVersionUID: 1996101190172373826L

Serialized Fields

m_Par

double[] m_Par

m_NumClasses

int m_NumClasses

The number of the class labels

m_Ridge

double m_Ridge

The ridge parameter.

m_Classes

int[] m_Classes

Class labels for each bag

m_Data

double[][][] m_Data

MI data

m_Attributes

Instances m_Attributes

All attribute names

xMean

double[] xMean

xSD

double[] xSD

m_AlgorithmType

int m_AlgorithmType

the type of processing

Class weka.classifiers.mi.MINND extends Classifier implements Serializable

serialVersionUID: -4512599203273864994L

Serialized Fields

m_Neighbour

int m_Neighbour

The number of nearest neighbour for prediction

m_Mean

double[][] m_Mean

The mean for each attribute of each exemplar

m_Variance

double[][] m_Variance

The variance for each attribute of each exemplar

m_Dimension

int m_Dimension

The dimension of each exemplar, i.e. (numAttributes-2)

m_Attributes

Instances m_Attributes

header info of the data

m_Class

double[] m_Class

The class label of each exemplar

m_NumClasses

int m_NumClasses

The number of class labels in the data

m_Weights

double[] m_Weights

The weight of each exemplar

m_Rate

double m_Rate

The learning rate in the gradient descent

m_MinArray

double[] m_MinArray

The minimum values for numeric attributes.

m_MaxArray

double[] m_MaxArray

The maximum values for numeric attributes.

m_STOP

double m_STOP

The stopping criteria of gradient descent

m_Change

double[][] m_Change

The weights that alter the dimnesion of each exemplar

m_NoiseM

double[][] m_NoiseM

The noise data of each exemplar

m_NoiseV

double[][] m_NoiseV

The noise data of each exemplar

m_ValidM

double[][] m_ValidM

The noise data of each exemplar

m_ValidV

double[][] m_ValidV

The noise data of each exemplar

m_Select

int m_Select

The number of nearest neighbour instances in the selection of noises in the training data

m_Choose

int m_Choose

The number of nearest neighbour exemplars in the selection of noises in the test data

m_Decay

double m_Decay

The decay rate of learning rate

Class weka.classifiers.mi.MIOptimalBall extends Classifier implements Serializable

serialVersionUID: -6465750129576777254L

Serialized Fields

m_Center

double[] m_Center

center of the optimal ball

m_Radius

double m_Radius

radius of the optimal ball

m_Distance

double[][][] m_Distance

the distances from each instance in a positive bag to each bag

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_filterType

int m_filterType

Whether to normalize/standardize/neither

m_ConvertToSI

MultiInstanceToPropositional m_ConvertToSI

filter used to convert the MI dataset into single-instance dataset

m_ConvertToMI

PropositionalToMultiInstance m_ConvertToMI

filter used to convert the single-instance dataset into MI dataset

Class weka.classifiers.mi.MISMO extends Classifier implements Serializable

serialVersionUID: -5834036950143719712L

Serialized Fields

m_classifiers

weka.classifiers.mi.MISMO.BinaryMISMO[][] m_classifiers

The binary classifier(s)

m_C

double m_C

The complexity parameter.

m_eps

double m_eps

Epsilon for rounding.

m_tol

double m_tol

Tolerance for accuracy of result.

m_filterType

int m_filterType

Whether to normalize/standardize/neither

m_minimax

boolean m_minimax

Use MIMinimax feature space?

m_NominalToBinary

NominalToBinary m_NominalToBinary

The filter used to make attributes numeric.

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_Missing

ReplaceMissingValues m_Missing

The filter used to get rid of missing values.

m_classIndex

int m_classIndex

The class index from the training data

m_classAttribute

Attribute m_classAttribute

The class attribute

m_kernel

Kernel m_kernel

Kernel to use

m_checksTurnedOff

boolean m_checksTurnedOff

Turn off all checks and conversions? Turning them off assumes that data is purely numeric, doesn't contain any missing values, and has a nominal class. Turning them off also means that no header information will be stored if the machine is linear. Finally, it also assumes that no instance has a weight equal to 0.

m_fitLogisticModels

boolean m_fitLogisticModels

Whether logistic models are to be fit

m_numFolds

int m_numFolds

The number of folds for the internal cross-validation

m_randomSeed

int m_randomSeed

The random number seed

Class weka.classifiers.mi.MISMO.BinaryMISMO extends java.lang.Object implements Serializable

serialVersionUID: -7107082483475433531L

Serialized Fields

m_alpha

double[] m_alpha

The Lagrange multipliers.

m_b

double m_b

The thresholds.

m_bLow

double m_bLow

The thresholds.

m_bUp

double m_bUp

The thresholds.

m_iLow

int m_iLow

The indices for m_bLow and m_bUp

m_iUp

int m_iUp

The indices for m_bLow and m_bUp

m_data

Instances m_data

The training data.

m_weights

double[] m_weights

Weight vector for linear machine.

m_sparseWeights

double[] m_sparseWeights

Variables to hold weight vector in sparse form. (To reduce storage requirements.)

m_sparseIndices

int[] m_sparseIndices

m_kernel

Kernel m_kernel

Kernel to use

m_class

double[] m_class

The transformed class values.

m_errors

double[] m_errors

The current set of errors for all non-bound examples.

m_I0

SMOset m_I0

{i: 0 < m_alpha[i] < C}

m_I1

SMOset m_I1

{i: m_class[i] = 1, m_alpha[i] = 0}

m_I2

SMOset m_I2

{i: m_class[i] = -1, m_alpha[i] = C}

m_I3

SMOset m_I3

{i: m_class[i] = 1, m_alpha[i] = C}

m_I4

SMOset m_I4

{i: m_class[i] = -1, m_alpha[i] = 0}

m_supportVectors

SMOset m_supportVectors

The set of support vectors {i: 0 < m_alpha[i]}

m_logistic

Logistic m_logistic

Stores logistic regression model for probability estimate

m_sumOfWeights

double m_sumOfWeights

Stores the weight of the training instances

Class weka.classifiers.mi.MISVM extends Classifier implements Serializable

serialVersionUID: 7622231064035278145L

Serialized Fields

m_SparseFilter

Filter m_SparseFilter

The filter used to transform the sparse datasets to nonsparse

m_SVM

weka.classifiers.mi.MISVM.SVM m_SVM

The SMO classifier used to compute SVM soluton w,b for the dataset

m_kernel

Kernel m_kernel

the kernel to use

m_C

double m_C

The complexity parameter.

m_Filter

Filter m_Filter

The filter used to standardize/normalize all values.

m_filterType

int m_filterType

Whether to normalize/standardize/neither

m_MaxIterations

int m_MaxIterations

the maximum number of iterations to perform

m_ConvertToProp

MultiInstanceToPropositional m_ConvertToProp

filter used to convert the MI dataset into single-instance dataset

Class weka.classifiers.mi.MIWrapper extends SingleClassifierEnhancer implements Serializable

serialVersionUID: -7707766152904315910L

Serialized Fields

m_NumClasses

int m_NumClasses

The number of the class labels

m_Method

int m_Method

the test method

m_ConvertToProp

MultiInstanceToPropositional m_ConvertToProp

Filter used to convert MI dataset into single-instance dataset

m_WeightMethod

int m_WeightMethod

the single-instance weight setting method

Class weka.classifiers.mi.SimpleMI extends SingleClassifierEnhancer implements Serializable

serialVersionUID: 9137795893666592662L

Serialized Fields

m_TransformMethod

int m_TransformMethod

the method used in transformation

Class weka.classifiers.mi.TLD extends RandomizableClassifier implements Serializable

serialVersionUID: 6657315525171152210L

Serialized Fields

m_MeanP

double[][] m_MeanP

The mean for each attribute of each positive exemplar

m_VarianceP

double[][] m_VarianceP

The variance for each attribute of each positive exemplar

m_MeanN

double[][] m_MeanN

The mean for each attribute of each negative exemplar

m_VarianceN

double[][] m_VarianceN

The variance for each attribute of each negative exemplar

m_SumP

double[][] m_SumP

The effective sum of weights of each positive exemplar in each dimension

m_SumN

double[][] m_SumN

The effective sum of weights of each negative exemplar in each dimension

m_ParamsP

double[] m_ParamsP

The parameters to be estimated for each positive exemplar

m_ParamsN

double[] m_ParamsN

The parameters to be estimated for each negative exemplar

m_Dimension

int m_Dimension

The dimension of each exemplar, i.e. (numAttributes-2)

m_Class

double[] m_Class

The class label of each exemplar

m_NumClasses

int m_NumClasses

The number of class labels in the data

m_Run

int m_Run

The number of runs to perform

m_Cutoff

double m_Cutoff

m_UseEmpiricalCutOff

boolean m_UseEmpiricalCutOff

Class weka.classifiers.mi.TLDSimple extends RandomizableClassifier implements Serializable

serialVersionUID: 9040995947243286591L

Serialized Fields

m_MeanP

double[][] m_MeanP

The mean for each attribute of each positive exemplar

m_MeanN

double[][] m_MeanN

The mean for each attribute of each negative exemplar

m_SumP

double[][] m_SumP

The effective sum of weights of each positive exemplar in each dimension

m_SumN

double[][] m_SumN

The effective sum of weights of each negative exemplar in each dimension

m_SgmSqP

double[] m_SgmSqP

Estimated sigma^2 in positive bags

m_SgmSqN

double[] m_SgmSqN

Estimated sigma^2 in negative bags

m_ParamsP

double[] m_ParamsP

The parameters to be estimated for each positive exemplar

m_ParamsN

double[] m_ParamsN

The parameters to be estimated for each negative exemplar

m_Dimension

int m_Dimension

The dimension of each exemplar, i.e. (numAttributes-2)

m_Class

double[] m_Class

The class label of each exemplar

m_NumClasses

int m_NumClasses

The number of class labels in the data

m_Run

int m_Run

m_Cutoff

double m_Cutoff

m_UseEmpiricalCutOff

boolean m_UseEmpiricalCutOff

m_LkRatio

double[] m_LkRatio

m_Attribute

Instances m_Attribute

Package weka.classifiers.mi.supportVector

Class weka.classifiers.mi.supportVector.MIPolyKernel extends PolyKernel implements Serializable

serialVersionUID: 7926421479341051777L

Class weka.classifiers.mi.supportVector.MIRBFKernel extends RBFKernel implements Serializable

serialVersionUID: -8711882393708956962L

Serialized Fields

m_kernelPrecalc

double[][] m_kernelPrecalc

The precalculated dotproducts of <inst_i,inst_i>

Package weka.classifiers.misc

Class weka.classifiers.misc.FLR extends Classifier implements Serializable

serialVersionUID: 3337906540579569626L

Serialized Fields

learnedCode

java.util.Vector<E> learnedCode

the RuleSet: a vector keeping the learned Fuzzy Lattices

m_Rhoa

double m_Rhoa

a double keeping the vignilance parameter rhoa

bounds

weka.classifiers.misc.FLR.FuzzyLattice bounds

a Fuzzy Lattice keeping the metric space

m_BoundsFile

java.io.File m_BoundsFile

a File pointing to the boundaries file (bounds.txt)

m_showRules

boolean m_showRules

a flag indicating whether the RuleSet will be displayed

index

int[] index

an index of the RuleSet (keeps how many rules are needed for each class)

classNames

java.lang.String[] classNames

an array of the names of the classes

Class weka.classifiers.misc.HyperPipes extends Classifier implements Serializable

serialVersionUID: -7527596632268975274L

Serialized Fields

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_Instances

Instances m_Instances

The structure of the training data

m_HyperPipes

weka.classifiers.misc.HyperPipes.HyperPipe[] m_HyperPipes

Stores the HyperPipe for each class

m_ZeroR

Classifier m_ZeroR

a ZeroR model in case no model can be built from the data

Class weka.classifiers.misc.SerializedClassifier extends Classifier implements Serializable

serialVersionUID: 4599593909947628642L

Serialized Fields

m_ModelFile

java.io.File m_ModelFile

the file where the serialized model is stored

Class weka.classifiers.misc.VFI extends Classifier implements Serializable

serialVersionUID: 8081692166331321866L

Serialized Fields

m_ClassIndex

int m_ClassIndex

The index of the class attribute

m_NumClasses

int m_NumClasses

The number of classes

m_Instances

Instances m_Instances

The training data

m_counts

double[][][] m_counts

The class counts for each interval of each attribute

m_globalCounts

double[] m_globalCounts

The global class counts

m_intervalBounds

double[][] m_intervalBounds

The lower bounds for each attribute

m_maxEntrop

double m_maxEntrop

The maximum entropy for the class

m_weightByConfidence

boolean m_weightByConfidence

Exponentially bias more confident intervals

m_bias

double m_bias

Bias towards more confident intervals

TINY

double TINY

Package weka.classifiers.pmml.consumer

Class