PMML43Ext module

class PMML43Ext.AR(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.AR

class PMML43Ext.ARDSquaredExponentialKernel(description=None, gamma='1', noiseVariance='1', Extension=None, Lambda=None)[source]

Bases: PMML43ExtSuper.ARDSquaredExponentialKernel

class PMML43Ext.ARIMA(RMSE=None, transformation='none', constantTerm='0', predictionMethod='conditionalLeastSquares', Extension=None, NonseasonalComponent=None, SeasonalComponent=None, DynamicRegressor=None, MaximumLikelihoodStat=None, OutlierEffect=None)[source]

Bases: PMML43ExtSuper.ARIMA

class PMML43Ext.ARMAPart(constant='0', p=None, q=None, Extension=None, AR=None, MA=None)[source]

Bases: PMML43ExtSuper.ARMAPart

class PMML43Ext.AbsoluteExponentialKernel(description=None, gamma='1', noiseVariance='1', Extension=None, Lambda=None)[source]

Bases: PMML43ExtSuper.AbsoluteExponentialKernel

class PMML43Ext.Adadelta(learningRate=None, rho=None, decayRate=None, epsilon=None, Extension=None)[source]

Bases: PMML43ExtSuper.Adadelta

class PMML43Ext.Adagrad(learningRate=None, decayRate=None, epsilon=None, Extension=None)[source]

Bases: PMML43ExtSuper.Adagrad

class PMML43Ext.Adam(learningRate=None, beta_1=None, beta_2=None, decayRate=None, epsilon=None, Extension=None)[source]

Bases: PMML43ExtSuper.Adam

class PMML43Ext.Adamax(learningRate=None, beta_1=None, beta_2=None, decayRate=None, epsilon=None, Extension=None)[source]

Bases: PMML43ExtSuper.Adamax

class PMML43Ext.Aggregate(field=None, function=None, groupField=None, sqlWhere=None, Extension=None)[source]

Bases: PMML43ExtSuper.Aggregate

summarize or collect groups of values, e.g. compute average

Parameters:field (string) – Column names
class PMML43Ext.Alternate(AnyDistribution=None, GaussianDistribution=None, PoissonDistribution=None, UniformDistribution=None, Extension=None)[source]

Bases: PMML43ExtSuper.Alternate

class PMML43Ext.Annotation(content=None, Extension=None, mixedclass_=None)[source]

Bases: PMML43ExtSuper.Annotation

class PMML43Ext.Anova(target=None, Extension=None, AnovaRow=None)[source]

Bases: PMML43ExtSuper.Anova

class PMML43Ext.AnovaRow(type_=None, sumOfSquares=None, degreesOfFreedom=None, meanOfSquares=None, fValue=None, pValue=None, Extension=None)[source]

Bases: PMML43ExtSuper.AnovaRow

class PMML43Ext.AntecedentSequence(Extension=None, SequenceReference=None, Time=None)[source]

Bases: PMML43ExtSuper.AntecedentSequence

class PMML43Ext.AnyDistribution(mean=None, variance=None, Extension=None)[source]

Bases: PMML43ExtSuper.AnyDistribution

class PMML43Ext.Application(name=None, version=None, Extension=None)[source]

Bases: PMML43ExtSuper.Application

class PMML43Ext.Apply(function=None, mapMissingTo=None, defaultValue=None, invalidValueTreatment='returnInvalid', Extension=None, Apply_member=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.Apply

Apply is an element used in DerivedField

Parameters:
  • function – derive a value by applying a function to one or more parameters
  • mapMissingTo – used to map a missing result to the value specified by the attribute
  • FieldRef – Field references are simply pass-throughs to fields previously defined in the DataDictionary, a DerivedField, or a result field
  • Constant – used in expressions which have multiple arguments. The actual value of a constant is given by the content of the element
  • NormContinuous – used to implement simple normalization functions such as the z-score transformation” (X - m ) / s, where m is the mean value and s is the standard deviation
  • NormDiscrete – refer to a certain input field define a fan-out function which maps a single input field to a set of normalized fields
  • Discretize – Takes the input field as input and maps values less than 0 to negative and other values to positive
  • MapValues – element can be used to create missing value indicators for categorical variables
  • TextIndex – TextIndex expression to extract frequency information from the text input field, for a given term. The TextIndex element fully configures how the text input should be indexed, including case sensitivity, normalization and other settings
  • Aggregate – summarize or collect groups of values, e.g., compute average
  • Lag – defined as the value of the given input field a fixed number of records prior to the current one,If the desired value is not present, for a given record, the lag will be set to missing
class PMML43Ext.ArrayType(content=None, n=None, type_=None, mixedclass_=None)[source]

Bases: PMML43ExtSuper.ArrayType

export(outfile, level, namespace_='', name_='ArrayType', namespacedef_='', pretty_print=True, *args)[source]
export_wrapper(outfile, level, namespace_='', name_='ArrayType', namespacedef_='', pretty_print=True, *args)[source]
class PMML43Ext.AssociationModel(modelName=None, functionName=None, algorithmName=None, numberOfTransactions=None, maxNumberOfItemsPerTA=None, avgNumberOfItemsPerTA=None, minimumSupport=None, minimumConfidence=None, lengthLimit=None, numberOfItems=None, numberOfItemsets=None, numberOfRules=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, LocalTransformations=None, Item=None, Itemset=None, AssociationRule=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.AssociationModel

add_AssociationRule(value, *args)[source]
add_AssociationRule_wrapper(value, *args)[source]
add_Item(value, *args)[source]
add_Item_wrapper(value, *args)[source]
add_Itemset(value, *args)[source]
add_Itemset_wrapper(value, *args)[source]
insert_AssociationRule_at(index, value, *args)[source]
insert_AssociationRule_at_wrapper(index, value, *args)[source]
insert_Item_at(index, value, *args)[source]
insert_Item_at_wrapper(index, value, *args)[source]
insert_Itemset_at(index, value, *args)[source]
insert_Itemset_at_wrapper(index, value, *args)[source]
set_AssociationRule(Rules, *args)[source]
set_AssociationRule_wrapper(Rules, *args)[source]
set_Item(Item, *args)[source]
set_Item_wrapper(Item, *args)[source]
set_Itemset(Itemset, *args)[source]
set_Itemset_wrapper(Itemset, *args)[source]
class PMML43Ext.AssociationRule(antecedent=None, consequent=None, support=None, confidence=None, lift=None, leverage=None, affinity=None, id=None, Extension=None)[source]

Bases: PMML43ExtSuper.AssociationRule

class PMML43Ext.Attribute(reasonCode=None, partialScore=None, Extension=None, SimplePredicate=None, CompoundPredicate=None, SimpleSetPredicate=None, True_=None, False_=None, ComplexPartialScore=None)[source]

Bases: PMML43ExtSuper.Attribute

class PMML43Ext.BaseCumHazardTables(maxTime=None, Extension=None, BaselineStratum=None, BaselineCell=None)[source]

Bases: PMML43ExtSuper.BaseCumHazardTables

class PMML43Ext.Baseline(AnyDistribution=None, GaussianDistribution=None, PoissonDistribution=None, UniformDistribution=None, Extension=None, CountTable=None, NormalizedCountTable=None, FieldRef=None)[source]

Bases: PMML43ExtSuper.Baseline

class PMML43Ext.BaselineCell(time=None, cumHazard=None, Extension=None)[source]

Bases: PMML43ExtSuper.BaselineCell

class PMML43Ext.BaselineModel(modelName=None, functionName=None, algorithmName=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, TestDistributions=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.BaselineModel

class PMML43Ext.BaselineStratum(value=None, label=None, maxTime=None, Extension=None, BaselineCell=None)[source]

Bases: PMML43ExtSuper.BaselineStratum

class PMML43Ext.BayesInput(fieldName=None, Extension=None, TargetValueStats=None, DerivedField=None, PairCounts=None)[source]

Bases: PMML43ExtSuper.BayesInput

BayesInput contains the counts pairing the discrete values of that field with those of the target field

Parameters:
  • fieldName – Name of the input field
  • TargetValueStats – TargetValueStats serves as the envelope for element TargetValueStat
  • DerivedField – which provides a common element for the various mappings
  • PairCounts – PairCounts lists, for a field Ii’s discrete value Iij, the TargetValueCounts that pair the value Iij with each value of the target field
class PMML43Ext.BayesInputs(Extension=None, BayesInput=None)[source]

Bases: PMML43ExtSuper.BayesInputs

BayesInputs element contains several BayesInput elements.

Parameters:BayesInput – contains the counts pairing the discrete values of that field with those of the target field
class PMML43Ext.BayesOutput(fieldName=None, Extension=None, TargetValueCounts=None)[source]

Bases: PMML43ExtSuper.BayesOutput

BayesOutput contains the counts associated with the values of the target field.

Parameters:
  • fieldName – Name of the output field
  • TargetValueCounts – TargetValueCounts lists the counts associated with each value of the target field
class PMML43Ext.BayesianNetworkModel(modelName=None, functionName=None, algorithmName=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, BayesianNetworkNodes=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.BayesianNetworkModel

class PMML43Ext.BayesianNetworkNodes(Extension=None, DiscreteNode=None, ContinuousNode=None)[source]

Bases: PMML43ExtSuper.BayesianNetworkNodes

class PMML43Ext.BlockIndicator(field=None)[source]

Bases: PMML43ExtSuper.BlockIndicator

class PMML43Ext.BoundaryValueMeans(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.BoundaryValueMeans

class PMML43Ext.BoundaryValues(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.BoundaryValues

class PMML43Ext.COUNT_TABLE_TYPE(sample=None, Extension=None, FieldValue=None, FieldValueCount=None)[source]

Bases: PMML43ExtSuper.COUNT_TABLE_TYPE

class PMML43Ext.CategoricalPredictor(name=None, value=None, coefficient=None, Extension=None)[source]

Bases: PMML43ExtSuper.CategoricalPredictor

class PMML43Ext.Categories(Extension=None, Category=None)[source]

Bases: PMML43ExtSuper.Categories

class PMML43Ext.Category(value=None, Extension=None)[source]

Bases: PMML43ExtSuper.Category

class PMML43Ext.Characteristic(name=None, reasonCode=None, baselineScore=None, Extension=None, Attribute=None)[source]

Bases: PMML43ExtSuper.Characteristic

class PMML43Ext.Characteristics(Extension=None, Characteristic=None)[source]

Bases: PMML43ExtSuper.Characteristics

class PMML43Ext.ChildParent(childField=None, parentField=None, parentLevelField=None, isRecursive='no', Extension=None, FieldColumnPair=None, TableLocator=None, InlineTable=None)[source]

Bases: PMML43ExtSuper.ChildParent

class PMML43Ext.ClassLabels(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.ClassLabels

class PMML43Ext.Cluster(id=None, name=None, size=None, Extension=None, KohonenMap=None, Array=None, Partition=None, Covariances=None)[source]

Bases: PMML43ExtSuper.Cluster

defined by a vector of center coordinates. Some distance measure is used to determine the nearest center, that is the nearest cluster for a given input record

Parameters:
  • id – it represents unique identification for cluster
  • name – The name of a cluster is is not required to be unique and is returned as the predictedDisplayValue
  • size – size is descriptive only (not used in predictions) and intended to capture the size of each cluster
  • KohonenMap – The element KohonenMap is appropriate for clustering models that were produced by a Kohonen map algorithm
  • Array – containing the center coordinates for the cluster
  • Partition – A Partition contains statistics for a subset of records, for example it can describe the population in a cluster, each Partition describes the distribution per field
class PMML43Ext.ClusteringField(field=None, isCenterField='true', fieldWeight='1', similarityScale=None, compareFunction=None, Extension=None, Comparisons=None)[source]

Bases: PMML43ExtSuper.ClusteringField

Parameters:
  • field – refers (by name) to a MiningField or to a DerivedField
  • isCenterField – indicates whether the respective field is a center field
  • fieldWeight – used in the comparison functions in order to compute the comparison measure
  • similarityScale – the distance such that similarity becomes 0.5
  • compareFunction – is a function of taking two field values and a similarityScale to define similarity/distance. It can override the general specification of compareFunction in ComparisonMeasure
class PMML43Ext.ClusteringModel(modelName=None, functionName=None, algorithmName=None, modelClass=None, numberOfClusters=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, LocalTransformations=None, ComparisonMeasure=None, ClusteringField=None, MissingValueWeights=None, Cluster=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.ClusteringModel

A cluster model basically consists of a set of clusters

Parameters:
  • modelName – element identifies the model with a unique name in the context of the PMML file
  • functionName – Stores what type of problems it is ex classification or regression
  • algorithmName – Stores algorithm name used in the model
  • modelClass – specifies whether the clusters are defined by center-vectors or whether they are defined by the statistics
  • numberOfClusters – attribute must be equal to the number of Cluster elements in the ClusteringModel
  • isScorable – The isScorable attribute indicates whether the model is valid for scoring
  • MiningSchema – list the fields that have to be provided in order to apply the model
  • Output – describes a set of result values that can be returned from a model
  • ClusteringField – The correspondence between input fields and their coordinates is defined via ClusteringFields
  • MissingValueWeights – used to adjust distance or similarity measures for missing data
  • Cluster – Clusters are identified by an implicit 1-based index, indicating the position in which each cluster appears in the model
  • ModelVerification – ModelVerification schema provides a dataset of model inputs and known results that can be used to verify accurate results are generated, regardless of the environment
add_Cluster(value, *args)[source]
add_Cluster_wrapper(value, *args)[source]
insert_Cluster_at(index, value, *args)[source]
insert_Cluster_at_wrapper(index, value, *args)[source]
set_Cluster(Cluster, *args)[source]
set_Cluster_wrapper(Cluster, *args)[source]
class PMML43Ext.ClusteringModelQuality(dataName=None, SSE=None, SSB=None)[source]

Bases: PMML43ExtSuper.ClusteringModelQuality

class PMML43Ext.Coefficient(value='0', Extension=None)[source]

Bases: PMML43ExtSuper.Coefficient

class PMML43Ext.Coefficients(numberOfCoefficients=None, absoluteValue='0', Extension=None, Coefficient=None)[source]

Bases: PMML43ExtSuper.Coefficients

add_Coefficient(value, *args)[source]
add_Coefficient_wrapper(value, *args)[source]
insert_Coefficient_at(index, value, *args)[source]
insert_Coefficient_at_wrapper(index, value, *args)[source]
set_Coefficient(Coefficient, *args)[source]
set_Coefficient_wrapper(Coefficient, *args)[source]
class PMML43Ext.ComparisonMeasure(kind=None, compareFunction='absDiff', minimum=None, maximum=None, Extension=None, euclidean=None, squaredEuclidean=None, chebychev=None, cityBlock=None, minkowski=None, simpleMatching=None, jaccard=None, tanimoto=None, binarySimilarity=None)[source]

Bases: PMML43ExtSuper.ComparisonMeasure

Per ClusteringModel there is one aggregation function: depending on the attribute kind in ComparisonMeasure the aggregated value is optimal if it is 0 (for distance measure) or greater values indicate optimal fit (for similarity measure)

class PMML43Ext.Comparisons(Extension=None, Matrix=None)[source]

Bases: PMML43ExtSuper.Comparisons

class PMML43Ext.ComplexPartialScore(Extension=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.ComplexPartialScore

class PMML43Ext.CompoundPredicate(booleanOperator=None, Extension=None, SimplePredicate=None, CompoundPredicate_member=None, SimpleSetPredicate=None, True_=None, False_=None)[source]

Bases: PMML43ExtSuper.CompoundPredicate

an encapsulating element for combining two or more elements as defined at the entity PREDICATE

booleanOperator:
The operators and, or and xor are associative binary operators, having their usual semantics. The order of evaluation is irrelevant for all the predicates within one CompoundPredicate
SimplePredicate:
defines a rule in the form of a simple boolean expression. The rule consists of field, operator (booleanOperator) for binary comparison, and value
SimpleSetPredicate:
checks whether a field value is element of a set. The set of values is specified by the array
True_:
a predicate element that identifies the boolean constant TRUE
False_:
a predicate element that identifies the boolean constant False
class PMML43Ext.CompoundRule(Extension=None, SimplePredicate=None, CompoundPredicate=None, SimpleSetPredicate=None, True_=None, False_=None, SimpleRule=None, CompoundRule_member=None)[source]

Bases: PMML43ExtSuper.CompoundRule

class PMML43Ext.Con(from_=None, weight=None, Extension=None)[source]

Bases: PMML43ExtSuper.Con

class PMML43Ext.ConfusionMatrix(Extension=None, ClassLabels=None, Matrix=None)[source]

Bases: PMML43ExtSuper.ConfusionMatrix

class PMML43Ext.ConsequentSequence(Extension=None, SequenceReference=None, Time=None)[source]

Bases: PMML43ExtSuper.ConsequentSequence

class PMML43Ext.Constant(dataType=None, valueOf_=None)[source]

Bases: PMML43ExtSuper.Constant

Used in expressions which have multiple arguments. The actual value of a constant is given by the content of the element

dataType:
Describe the dataType of the ParametersField
valueOf_:
Value of the given Constant
class PMML43Ext.Constraints(minimumNumberOfItems='1', maximumNumberOfItems=None, minimumNumberOfAntecedentItems='1', maximumNumberOfAntecedentItems=None, minimumNumberOfConsequentItems='1', maximumNumberOfConsequentItems=None, minimumSupport='0', minimumConfidence='0', minimumLift='0', minimumTotalSequenceTime='0', maximumTotalSequenceTime=None, minimumItemsetSeparationTime='0', maximumItemsetSeparationTime=None, minimumAntConsSeparationTime='0', maximumAntConsSeparationTime=None, Extension=None)[source]

Bases: PMML43ExtSuper.Constraints

class PMML43Ext.ContStats(totalValuesSum=None, totalSquaresSum=None, Extension=None, Interval=None, NUM_ARRAY=None)[source]

Bases: PMML43ExtSuper.ContStats

class PMML43Ext.ContinuousConditionalProbability(count=None, Extension=None, ParentValue=None, ContinuousDistribution=None)[source]

Bases: PMML43ExtSuper.ContinuousConditionalProbability

class PMML43Ext.ContinuousDistribution(Extension=None, TriangularDistributionForBN=None, NormalDistributionForBN=None, LognormalDistributionForBN=None, UniformDistributionForBN=None)[source]

Bases: PMML43ExtSuper.ContinuousDistribution

class PMML43Ext.ContinuousNode(name=None, count=None, Extension=None, DerivedField=None, ContinuousConditionalProbability=None, ContinuousDistribution=None)[source]

Bases: PMML43ExtSuper.ContinuousNode

class PMML43Ext.CorrelationFields(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.CorrelationFields

class PMML43Ext.CorrelationMethods(Extension=None, Matrix=None)[source]

Bases: PMML43ExtSuper.CorrelationMethods

class PMML43Ext.CorrelationValues(Extension=None, Matrix=None)[source]

Bases: PMML43ExtSuper.CorrelationValues

class PMML43Ext.Correlations(Extension=None, CorrelationFields=None, CorrelationValues=None, CorrelationMethods=None)[source]

Bases: PMML43ExtSuper.Correlations

class PMML43Ext.Counts(totalFreq=None, missingFreq=None, invalidFreq=None, cardinality=None, Extension=None)[source]

Bases: PMML43ExtSuper.Counts

class PMML43Ext.Covariances(Extension=None, Matrix=None)[source]

Bases: PMML43ExtSuper.Covariances

Parameters:Matrix – stores coordinate-by-coordinate variances (diagonal cells) and covariances (non-diagonal cells)
class PMML43Ext.CovariateList(Extension=None, Predictor=None)[source]

Bases: PMML43ExtSuper.CovariateList

class PMML43Ext.DataDictionary(numberOfFields=None, Extension=None, DataField=None, Taxonomy=None)[source]

Bases: PMML43ExtSuper.DataDictionary

The DataDictionary contains definitions for fields as used in mining models. It specifies the types and value ranges. These definitions are assumed to be independent of specific data sets as used for training or scoring a specific model

Parameters:
  • numberOfFields – The value numberOfFields is the number of fields which are defined in the content of DataDictionary, this number can be added for consistency checks
  • DataField – The name of a DataField must be unique from other names in the DataDictionary and, with few exceptions, unique from the names of other fields in the PMML document
  • Taxonomy – The optional attribute taxonomy refers to a taxonomy of values. The value is a name of a taxonomy. It describes a hierarchy of values. The attribute is only applicable to categorical fields
add_DataField(value, *args)[source]
add_DataField_wrapper(value, *args)[source]
insert_DataField_at(index, value, *args)[source]
insert_DataField_at_wrapper(index, value, *args)[source]
set_DataField(DataField, *args)[source]
set_DataField_wrapper(DataField, *args)[source]
class PMML43Ext.DataField(name=None, displayName=None, optype=None, dataType=None, mimeType=None, taxonomy=None, isCyclic='0', Extension=None, Interval=None, Value=None)[source]

Bases: PMML43ExtSuper.DataField

DataField contains features name and target name

Parameters:
  • name – The name of a DataField must be unique from other names in the DataDictionary and, with few exceptions, unique from the names of other fields in the PMML document
  • displayName – DisplayName can be used when the application calls the PMML consumer. Once the consumer has received the parameters and matched to the MiningFields, the displayName is not relevant anymore. Only name is significant for internal processing,
  • optype – The fields are separated into different types depending on which operations are defined on the values; this is defined by the attribute optype
  • dataType – This field contain data type of the feature and target name in the DataDictionary
  • taxonomy – The optional attribute taxonomy refers to a taxonomy of values. The value is a name of a taxonomy. It describes a hierarchy of values. The attribute is only applicable to categorical fields
  • Value – Value is used to define the value ranges for fields in the DataDictionary
class PMML43Ext.Decision(value=None, displayValue=None, description=None, Extension=None)[source]

Bases: PMML43ExtSuper.Decision

Derive a decision from the output of a data mining model. For this result feature, OutputField must contain an EXPRESSION, unless it is used to refer to a decision of segment model through the segmentID attribute.

class PMML43Ext.DecisionTree(modelName=None, functionName=None, algorithmName=None, missingValueStrategy='none', missingValuePenalty='1.0', noTrueChildStrategy='returnNullPrediction', splitCharacteristic='multiSplit', Extension=None, Output=None, ModelStats=None, Targets=None, LocalTransformations=None, ResultField=None, Node=None)[source]

Bases: PMML43ExtSuper.DecisionTree

class PMML43Ext.Decisions(businessProblem=None, description=None, Extension=None, Decision=None)[source]

Bases: PMML43ExtSuper.Decisions

class PMML43Ext.DeepNetwork(modelName=None, functionName=None, algorithmName=None, normalizationMethod='none', numberOfLayers=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, TrainingParameters=None, NetworkLayer=None, NeuralOutputs=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.DeepNetwork

add_NetworkLayer(value, *args)[source]
add_NetworkLayer_wrapper(value, *args)[source]
insert_NetworkLayer_at(index, value, *args)[source]
insert_NetworkLayer_at_wrapper(index, value, *args)[source]
set_NetworkLayer(NetworkLayer, *args)[source]
set_NetworkLayer_wrapper(NetworkLayer, *args)[source]
class PMML43Ext.DefineFunction(name=None, optype=None, dataType=None, Extension=None, ParameterField=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.DefineFunction

class PMML43Ext.Delimiter(delimiter=None, gap=None, Extension=None)[source]

Bases: PMML43ExtSuper.Delimiter

class PMML43Ext.Denominator(Extension=None, NonseasonalFactor=None, SeasonalFactor=None)[source]

Bases: PMML43ExtSuper.Denominator

class PMML43Ext.DerivedField(name=None, displayName=None, optype=None, dataType=None, datasetName=None, trainingBackend=None, architectureName=None, Extension=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None, Value=None)[source]

Bases: PMML43ExtSuper.DerivedField

which provides a common element for the various mappings. They can also appear at several places in the definition of specific models such as neural network or Naïve Bayes models

Parameters:
  • name – name of the element
  • optype – The attribute optype is needed in order to eliminate cases where the resulting type is not known
  • dataType – specifies the data type for the output column
  • FieldRef – Field references are simply pass-throughs to fields previously defined in the DataDictionary, a DerivedField, or a result field
  • Constant – used in expressions which have multiple arguments. The actual value of a constant is given by the content of the element
  • NormContinuous – defines how to normalize an input field by piecewise linear interpolation
  • NormDiscrete – refer to a certain input field define a fan-out function which maps a single input field to a set of normalized fields
  • Discretize – Takes the input field as input and maps values less than 0 to negative and other values to positive
  • MapValues – element can be used to create missing value indicators for categorical variables
  • TextIndex – TextIndex expression to extract frequency information from the text input field, for a given term. The TextIndex element fully configures how the text input should be indexed, including case sensitivity, normalization and other settings
  • Aggregate – summarize or collect groups of values, e.g., compute average
  • Lag – defined as the value of the given input field a fixed number of records prior to the current one,If the desired value is not present, for a given record, the lag will be set to missing
class PMML43Ext.DiscrStats(modalValue=None, Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.DiscrStats

class PMML43Ext.DiscreteConditionalProbability(count=None, Extension=None, ParentValue=None, ValueProbability=None)[source]

Bases: PMML43ExtSuper.DiscreteConditionalProbability

class PMML43Ext.DiscreteNode(name=None, count=None, Extension=None, DerivedField=None, DiscreteConditionalProbability=None, ValueProbability=None)[source]

Bases: PMML43ExtSuper.DiscreteNode

class PMML43Ext.Discretize(field=None, mapMissingTo=None, defaultValue=None, dataType=None, Extension=None, DiscretizeBin=None)[source]

Bases: PMML43ExtSuper.Discretize

Discretization of numerical input fields is a mapping from continuous to discrete values using intervals

Parameters:
  • field – defines the name of the input field
  • mapMissingTo – may be used to map a missing result to the value specified by the attribute. If the attribute is not present, the result remains missing
  • DiscretizeBin – define a set of mappings from an intervali to a binValue
class PMML43Ext.DiscretizeBin(binValue=None, Extension=None, Interval=None)[source]

Bases: PMML43ExtSuper.DiscretizeBin

class PMML43Ext.DocumentTermMatrix(Extension=None, Matrix=None)[source]

Bases: PMML43ExtSuper.DocumentTermMatrix

class PMML43Ext.DynamicRegressor(field=None, transformation='none', delay='0', futureValuesMethod='constant', targetField=None, Extension=None, Numerator=None, Denominator=None, RegressorValues=None)[source]

Bases: PMML43ExtSuper.DynamicRegressor

class PMML43Ext.EventValues(Extension=None, Value=None, Interval=None)[source]

Bases: PMML43ExtSuper.EventValues

class PMML43Ext.ExponentialSmoothing(RMSE=None, transformation='none', Level=None, Trend_ExpoSmooth=None, Seasonality_ExpoSmooth=None, TimeValue=None)[source]

Bases: PMML43ExtSuper.ExponentialSmoothing

class PMML43Ext.Extension(extender=None, name=None, value=None, anytypeobjs_=None)[source]

Bases: PMML43ExtSuper.Extension

class PMML43Ext.FactorList(Extension=None, Predictor=None)[source]

Bases: PMML43ExtSuper.FactorList

class PMML43Ext.False_(Extension=None)[source]

Bases: PMML43ExtSuper.False_

class PMML43Ext.FieldColumnPair(field=None, column=None, Extension=None)[source]

Bases: PMML43ExtSuper.FieldColumnPair

class PMML43Ext.FieldRef(field=None, mapMissingTo=None, Extension=None)[source]

Bases: PMML43ExtSuper.FieldRef

Field references are simply pass-throughs to fields previously defined in the DataDictionary, a DerivedField, or a result field

Parameters:
  • field – Name of the field
  • mapMissingTo – may be used to map a missing result to the value specified by the attribute. If the attribute is not present, the result remains missing
class PMML43Ext.FieldValue(field=None, value=None, Extension=None, FieldValue_member=None, FieldValueCount=None)[source]

Bases: PMML43ExtSuper.FieldValue

class PMML43Ext.FieldValueCount(field=None, value=None, count=None, Extension=None)[source]

Bases: PMML43ExtSuper.FieldValueCount

class PMML43Ext.FinalNoise(Array=None)[source]

Bases: PMML43ExtSuper.FinalNoise

class PMML43Ext.FinalNu(Array=None)[source]

Bases: PMML43ExtSuper.FinalNu

class PMML43Ext.FinalOmega(Matrix=None)[source]

Bases: PMML43ExtSuper.FinalOmega

class PMML43Ext.FinalPredictedNoise(Array=None)[source]

Bases: PMML43ExtSuper.FinalPredictedNoise

class PMML43Ext.FinalStateVector(Array=None)[source]

Bases: PMML43ExtSuper.FinalStateVector

class PMML43Ext.FinalTheta(Theta=None)[source]

Bases: PMML43ExtSuper.FinalTheta

class PMML43Ext.GARCH(Extension=None, ARMAPart=None, GARCHPart=None)[source]

Bases: PMML43ExtSuper.GARCH

class PMML43Ext.GARCHPart(constant='0', gp=None, gq=None, Extension=None, ResidualSquareCoefficients=None, VarianceCoefficients=None)[source]

Bases: PMML43ExtSuper.GARCHPart

class PMML43Ext.GaussianDistribution(mean=None, variance=None, Extension=None)[source]

Bases: PMML43ExtSuper.GaussianDistribution

class PMML43Ext.GaussianProcessModel(modelName=None, functionName=None, algorithmName=None, optimizer=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, RadialBasisKernel=None, ARDSquaredExponentialKernel=None, AbsoluteExponentialKernel=None, GeneralizedExponentialKernel=None, TrainingInstances=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.GaussianProcessModel

class PMML43Ext.GeneralRegressionModel(targetVariableName=None, modelType=None, modelName=None, functionName=None, algorithmName=None, targetReferenceCategory=None, cumulativeLink=None, linkFunction=None, linkParameter=None, trialsVariable=None, trialsValue=None, distribution=None, distParameter=None, offsetVariable=None, offsetValue=None, modelDF=None, endTimeVariable=None, startTimeVariable=None, subjectIDVariable=None, statusVariable=None, baselineStrataVariable=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, ParameterList=None, FactorList=None, CovariateList=None, PPMatrix=None, PCovMatrix=None, ParamMatrix=None, EventValues=None, BaseCumHazardTables=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.GeneralRegressionModel

class PMML43Ext.GeneralizedExponentialKernel(description=None, gamma='1', noiseVariance='1', degree='1', Extension=None, Lambda=None)[source]

Bases: PMML43ExtSuper.GeneralizedExponentialKernel

class PMML43Ext.HVector(Array=None)[source]

Bases: PMML43ExtSuper.HVector

class PMML43Ext.Header(copyright=None, description=None, modelVersion=None, Extension=None, Application=None, Annotation=None, Timestamp=None)[source]

Bases: PMML43ExtSuper.Header

The Header contains useful information about the PMML document

Parameters:
  • copyright (string) – User copyright
  • description (string) – Human readable string describing the model
  • modelVersion (string) – User determined version
exportAttributes(outfile, level, already_processed, namespace_='', name_='Header', *args)[source]
exportAttributes_wrapper(outfile, level, already_processed, namespace_='', name_='Header', *args)[source]
class PMML43Ext.INT_SparseArray(n=None, defaultValue='0', Indices=None, INT_Entries=None)[source]

Bases: PMML43ExtSuper.INT_SparseArray

class PMML43Ext.InlineTable(Extension=None, row=None)[source]

Bases: PMML43ExtSuper.InlineTable

class PMML43Ext.InstanceField(field=None, column=None, Extension=None)[source]

Bases: PMML43ExtSuper.InstanceField

Parameters:
  • field – Contains the name of a DataField or a DerivedField
  • column – Defines the name of the tag or column used by element InlineTable. This attribute is required if element InlineTable is used to represent training data
class PMML43Ext.InstanceFields(Extension=None, InstanceField=None)[source]

Bases: PMML43ExtSuper.InstanceFields

The InstanceFields element serves as an envelope for all the fields included in the training instances

class PMML43Ext.Interval(closure=None, leftMargin=None, rightMargin=None, Extension=None)[source]

Bases: PMML43ExtSuper.Interval

class PMML43Ext.Item(id=None, value=None, field=None, category=None, mappedValue=None, weight=None, Extension=None)[source]

Bases: PMML43ExtSuper.Item

class PMML43Ext.ItemRef(itemRef=None, Extension=None)[source]

Bases: PMML43ExtSuper.ItemRef

class PMML43Ext.Itemset(id=None, support=None, numberOfItems=None, Extension=None, ItemRef=None)[source]

Bases: PMML43ExtSuper.Itemset

add_ItemRef(value, *args)[source]
add_ItemRef_wrapper(value, *args)[source]
insert_ItemRef_at(index, value, *args)[source]
insert_ItemRef_at_wrapper(index, value, *args)[source]
set_ItemRef(ItemRef, *args)[source]
set_ItemRef_wrapper(ItemRef, *args)[source]
class PMML43Ext.KNNInput(field=None, fieldWeight='1', compareFunction=None, Extension=None)[source]

Bases: PMML43ExtSuper.KNNInput

elements which define the fields used to query the k-NN model

Parameters:
  • field – Contains the name of a DataField or a DerivedField
  • fieldWeight – Defines the importance factor for the field. It is used in the comparison functions to compute the comparison measure. The value must be a number greater than 0. The default value is 1.0
  • compareFunction – attribute, this is either defined as default in element ComparisonMeasure or it can be defined per KNNInput
class PMML43Ext.KNNInputs(Extension=None, KNNInput=None)[source]

Bases: PMML43ExtSuper.KNNInputs

This element serves as an envelope for defining all of the k-NN inputs

Parameters:KNNInput – elements which define the fields used to query the k-NN model
class PMML43Ext.KalmanState(FinalOmega=None, FinalStateVector=None, HVector=None)[source]

Bases: PMML43ExtSuper.KalmanState

class PMML43Ext.KohonenMap(coord1=None, coord2=None, coord3=None, Extension=None)[source]

Bases: PMML43ExtSuper.KohonenMap

The element KohonenMap is appropriate for clustering models that were produced by a Kohonen map algorithm

Parameters:
  • coord1 – Describe the position of the current cluster(cluster 1) in a map with up to three dimensions
  • coord2 – Describe the position of the current cluster(cluster2) in a map with up to three dimensions
  • coord3 – Describe the position of the current cluster(cluster3) in a map with up to three dimensions
class PMML43Ext.Lag(field=None, n=1, Extension=None, BlockIndicator=None)[source]

Bases: PMML43ExtSuper.Lag

defined as the value of the given input field a fixed number of records prior to the current one,If the desired value is not present, for a given record, the lag will be set to missing

class PMML43Ext.Lambda(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.Lambda

class PMML43Ext.LayerBias(biasShape=None, biasFlattenAxis=None, content=None, floatType='float32', floatsPerLine=12, src=None, Extension=None, mixedclass_=None)[source]

Bases: PMML43ExtSuper.LayerBias

export(outfile, level, namespace_='', name_='LayerBias', namespacedef_='', pretty_print=True, *args)[source]
export_wrapper(outfile, level, namespace_='', name_='LayerBias', namespacedef_='', pretty_print=True, *args)[source]
weights(*args)[source]
weights_wrapper(*args)[source]
class PMML43Ext.LayerParameters(activationFunction=None, inputDimension=None, outputDimension=None, featureMaps=None, kernel=None, paddingType=None, stride=None, dilationRate=None, poolSize=None, depthMultiplier=None, paddingDims=None, croppingDims=None, upsamplingSize=None, return_sequences=None, return_state=None, stateful=None, inputLength=None, recurrentUnits=None, recurrentActivation=None, recurrentDropout=None, go_backwards=None, batchNormalizationEpsilon=None, flattenAxis=None, batchNormalizationAxis=None, batchNormalizationMomentum=None, batchNormalizationCenter=None, batchNormalizationScale=None, gaussianNoiseStdev=None, gaussianDropoutRate=None, alphaDropoutRate=None, alphaDropoutSeed=None, betaInitializer=None, gammaInitializer=None, movingMeanInitializer=None, movingVarianceInitializer=None, recurrentInitializer=None, betaRegularizer=None, gammaRegularizer=None, betaConstraint=None, gammaConstraint=None, kernelInitializer=None, biasInitializer=None, kernelRegularizer=None, biasRegularizer=None, kernelConstraint=None, biasConstraint=None, depthwiseConstraint=None, pointwiseConstraint=None, recurrentConstraint=None, batchSize=None, dropoutRate=None, dropoutNoiseShape=None, dropoutSeed=None, generalLUAlpha=None, reshapeTarget=None, permuteDims=None, repeatVectorTimes=None, activityRegularizerL1=None, activityRegularizerL2=None, maskValue=None, mergeLayerOp=None, mergeLayerDotOperationAxis=None, mergeLayerDotNormalize=None, mergeLayerConcatOperationAxes=None, slicingAxis=None, Extension=None)[source]

Bases: PMML43ExtSuper.LayerParameters

class PMML43Ext.LayerWeights(weightsShape=None, weightsFlattenAxis=None, content=None, floatType='float32', floatsPerLine=12, src=None, Extension=None, mixedclass_=None)[source]

Bases: PMML43ExtSuper.LayerWeights

export(outfile, level, namespace_='', name_='LayerWeights', namespacedef_='', pretty_print=True, *args)[source]
export_wrapper(outfile, level, namespace_='', name_='LayerWeights', namespacedef_='', pretty_print=True, *args)[source]
weights(*args)[source]
weights_wrapper(*args)[source]
class PMML43Ext.Level(alpha=None, initialLevelValue=None, smoothedValue=None)[source]

Bases: PMML43ExtSuper.Level

class PMML43Ext.LiftData(targetFieldValue=None, targetFieldDisplayValue=None, rankingQuality=None, Extension=None, ModelLiftGraph=None, OptimumLiftGraph=None, RandomLiftGraph=None)[source]

Bases: PMML43ExtSuper.LiftData

class PMML43Ext.LiftGraph(Extension=None, XCoordinates=None, YCoordinates=None, BoundaryValues=None, BoundaryValueMeans=None)[source]

Bases: PMML43ExtSuper.LiftGraph

class PMML43Ext.LinearKernelType(description=None, Extension=None)[source]

Bases: PMML43ExtSuper.LinearKernelType

class PMML43Ext.LinearNorm(orig=None, norm=None, Extension=None)[source]

Bases: PMML43ExtSuper.LinearNorm

class PMML43Ext.LocalTransformations(Extension=None, DerivedField=None)[source]

Bases: PMML43ExtSuper.LocalTransformations

class PMML43Ext.LognormalDistributionForBN(Extension=None, Mean=None, Variance=None)[source]

Bases: PMML43ExtSuper.LognormalDistributionForBN

class PMML43Ext.Losses(loss=None, Extension=None)[source]

Bases: PMML43ExtSuper.Losses

class PMML43Ext.Lower(Extension=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.Lower

class PMML43Ext.MA(Extension=None, Coefficients=None, Residuals=None)[source]

Bases: PMML43ExtSuper.MA

class PMML43Ext.MapValues(mapMissingTo=None, defaultValue=None, outputColumn=None, dataType=None, Extension=None, FieldColumnPair=None, TableLocator=None, InlineTable=None)[source]

Bases: PMML43ExtSuper.MapValues

element can be used to create missing value indicators for categorical variables

Parameters:
  • mapMissingTo (string) – may be used to map a missing result to the value specified by the attribute. If the attribute is not present, the result remains missing
  • InlineTable – used in mapping values
class PMML43Ext.MatCell(row=None, col=None, valueOf_=None)[source]

Bases: PMML43ExtSuper.MatCell

class PMML43Ext.Matrix(kind='any', nbRows=None, nbCols=None, diagDefault=None, offDiagDefault=None, Array=None, MatCell=None)[source]

Bases: PMML43ExtSuper.Matrix

class PMML43Ext.MaximumLikelihoodStat(method=None, periodDeficit='0', KalmanState=None, ThetaRecursionState=None)[source]

Bases: PMML43ExtSuper.MaximumLikelihoodStat

class PMML43Ext.Mean(Extension=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.Mean

class PMML43Ext.MeasurementMatrix(Extension=None, Matrix=None)[source]

Bases: PMML43ExtSuper.MeasurementMatrix

class PMML43Ext.Metrics(top_k_categories_for_accuracy=None, metric=None, Extension=None)[source]

Bases: PMML43ExtSuper.Metrics

class PMML43Ext.MiningBuildTask(Extension=None)[source]

Bases: PMML43ExtSuper.MiningBuildTask

class PMML43Ext.MiningField(name=None, usageType='active', optype=None, importance=None, outliers='asIs', lowValue=None, highValue=None, missingValueReplacement=None, missingValueTreatment=None, invalidValueTreatment='returnInvalid', Extension=None)[source]

Bases: PMML43ExtSuper.MiningField

identify which of the DataFields defined in the DataDictionary are used in the model

Parameters:
  • name (string) – symbolic name of field, must refer to a field in the DataDictionary
  • usageType – field used as input (independent field)
  • optype – The attribute value overrides the corresponding value in the DataField
  • importance – states the relative importance of the field
  • outliers – it shows the records which are out of trend
  • lowValue – Extreme low point to define outliers
  • highValue – Extreme high point to define outliers
  • missingValueReplacement – If this attribute is specified then a missing input value is automatically replaced by the given value
  • missingValueTreatment – this attribute just indicates how the missingValueReplacement was derived
  • invalidValueTreatment – This field specifies how invalid input values are handled
exportAttributes(outfile, level, already_processed, namespace_='', name_='MiningField', *args)[source]
exportAttributes_wrapper(outfile, level, already_processed, namespace_='', name_='MiningField', *args)[source]
class PMML43Ext.MiningModel(modelName=None, functionName=None, algorithmName=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, Regression=None, DecisionTree=None, Segmentation=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.MiningModel

MiningModel contains Segmentation element with a number of Segment elements as well as the attribute multipleModelMethod specifying how all the models applicable to a record should be combined.

Parameters:
  • modelName (string) – element identifies the model with a unique name in the context of the PMML file
  • functionName (string) – Stores what type of problems it is ex classification or regression
  • algorithmName (string) – Stores algorithm name used in the model
  • isScorable – indicates whether the model is valid for scoring
  • MiningSchema – list the fields that have to be provided in order to apply the model
  • Output – describes a set of result values that can be returned from a model
  • Targets – The target values are derived from a variety of elements in the models
  • LocalTransformations – Any pre-processing information goes here
  • Regression – regression equation can be used to define an input transformation in another model which happens to be a TreeModel
  • DecisionTree – DecisionTree contains the essential elements of a TreeModel
  • Segmentation – Segmentation allows representation of different models for different data segments and also can be used for model ensembles and model sequences
  • ModelVerification – ModelVerification schema provides a dataset of model inputs and known results that can be used to verify accurate results are generated, regardless of the environment
class PMML43Ext.MiningSchema(Extension=None, MiningField=None)[source]

Bases: PMML43ExtSuper.MiningSchema

list the fields which a user has to provide in order to apply the model.

Parameters:MiningField – identify which of the DataFields defined in the DataDictionary are used in the model
class PMML43Ext.MissingValueWeights(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.MissingValueWeights

class PMML43Ext.ModelExplanation(Extension=None, PredictiveModelQuality=None, ClusteringModelQuality=None, Correlations=None)[source]

Bases: PMML43ExtSuper.ModelExplanation

class PMML43Ext.ModelLiftGraph(Extension=None, LiftGraph=None)[source]

Bases: PMML43ExtSuper.ModelLiftGraph

class PMML43Ext.ModelStats(Extension=None, UnivariateStats=None, MultivariateStats=None)[source]

Bases: PMML43ExtSuper.ModelStats

class PMML43Ext.ModelVerification(recordCount=None, fieldCount=None, Extension=None, VerificationFields=None, InlineTable=None)[source]

Bases: PMML43ExtSuper.ModelVerification

class PMML43Ext.MultivariateStat(name=None, category=None, exponent='1', isIntercept=False, importance=None, stdError=None, tValue=None, chiSquareValue=None, fStatistic=None, dF=None, pValueAlpha=None, pValueInitial=None, pValueFinal=None, confidenceLevel='0.95', confidenceLowerBound=None, confidenceUpperBound=None, Extension=None)[source]

Bases: PMML43ExtSuper.MultivariateStat

class PMML43Ext.MultivariateStats(targetCategory=None, Extension=None, MultivariateStat=None)[source]

Bases: PMML43ExtSuper.MultivariateStats

class PMML43Ext.Nadam(learningRate=None, beta_1=None, beta_2=None, schedule_decay=None, epsilon=None, Extension=None)[source]

Bases: PMML43ExtSuper.Nadam

class PMML43Ext.NaiveBayesModel(modelName=None, threshold=None, functionName=None, algorithmName=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, BayesInputs=None, BayesOutput=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.NaiveBayesModel

Naïve Bayes uses Bayes’ Theorem, combined with a (“naive”) presumption of conditional independence, to predict the value of a target (output), from evidence given by one or more predictor (input) fields

Parameters:
  • modelName – element identifies the model with a unique name in the context of the PMML file
  • functionName – Stores what type of problems it is ex classification or regression
  • algorithmName – Stores algorithm name used in the model
  • isScorable – The isScorable attribute indicates whether the model is valid for scoring
  • MiningSchema – list the fields that have to be provided in order to apply the model
  • Output – describes a set of result values that can be returned from a model
  • BayesInputs – element contains several BayesInput elements
  • BayesOutput – BayesOutput contains the counts associated with the values of the target field
class PMML43Ext.NearestNeighborModel(modelName=None, functionName=None, algorithmName=None, numberOfNeighbors=None, continuousScoringMethod='average', categoricalScoringMethod='majorityVote', instanceIdVariable=None, threshold='0.001', isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, TrainingInstances=None, ComparisonMeasure=None, KNNInputs=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.NearestNeighborModel

The root element of an XML k-NN model. Each instance of a k-NN model must start with this element

Parameters:
  • modelName – element identifies the model with a unique name in the context of the PMML file
  • functionName – Stores what type of problems it is ex classification or regression
  • algorithmName – Stores algorithm name used in the model
  • numberOfNeighbors – Specifies K, the number of desired neighbors
  • continuousScoringMethod – Specify the scoring (or combining) method based on the continuous or categorical target values of K neighbors
  • categoricalScoringMethod – Specify the scoring (or combining) method based on the continuous or categorical target values of K neighbors
  • instanceIdVariable – Contains the instance ID variable name and so refers to the name of a field in InstanceFields
  • threshold – Defines a very small positive number to be used for “weighted” scoring methods to avoid numerical problems when distance or similarity measure is zero
  • isScorable – The isScorable attribute indicates whether the model is valid for scoring
  • MiningSchema – list the fields that have to be provided in order to apply the model
  • Output – describes a set of result values that can be returned from a model
  • Targets – The target values are derived from a variety of elements in the models
  • TrainingInstances – element serves as an envelope for defining all of the training instances. It contains the definition of the fields included in the training instances as well as a table for representing the training data itself
  • ComparisonMeasure – Defines the distance or similarity measure used to find the k-nearest neighbors
  • KNNInputs – This element serves as an envelope for defining all of the k-NN inputs
class PMML43Ext.NetworkLayer(normalizationMethod='none', layerType=None, layerId=None, connectionLayerId=None, inputFieldName=None, Extension=None, LayerParameters=None, LayerWeights=None, LayerBias=None)[source]

Bases: PMML43ExtSuper.NetworkLayer

class PMML43Ext.NeuralInput(id=None, Extension=None, DerivedField=None)[source]

Bases: PMML43ExtSuper.NeuralInput

defines how input fields are normalized so that the values can be processed in the neural network

Parameters:
  • id – an identifier id which must be unique in all layers
  • DerivedField – which provides a common element for the various mappings. They can also appear at several places in the definition of specific models such as neural network or Naïve Bayes models
class PMML43Ext.NeuralInputs(numberOfInputs=None, Extension=None, NeuralInput=None)[source]

Bases: PMML43ExtSuper.NeuralInputs

NeuralInputs Defines all the NeuralInput in InputLayer

Parameters:
  • numberOfInputs – Describe the number of input features used in Model building
  • NeuralInput – Defines how input fields are normalized so that the values can be processed in the neural network
add_NeuralInput(value, *args)[source]
add_NeuralInput_wrapper(value, *args)[source]
insert_NeuralInput_at(index, value, *args)[source]
insert_NeuralInput_at_wrapper(index, value, *args)[source]
set_NeuralInput(NeuralInput, *args)[source]
set_NeuralInput_wrapper(NeuralInput, *args)[source]
class PMML43Ext.NeuralLayer(numberOfNeurons=None, activationFunction=None, threshold=None, width=None, altitude=None, normalizationMethod=None, Extension=None, Neuron=None)[source]

Bases: PMML43ExtSuper.NeuralLayer

Neural Layer describe the Structure of the neural network i.e. number of layer and number of neuron in each layer etc.

Parameters:
  • numberOfNeurons – Indicate the number of neurons in each Layer
  • activationFunction – Used to determine the output of neural network like yes or no. It maps the resulting values in between 0 to 1 or -1 to 1 etc.
  • threshold – In threshold function, any values above or equal to a given threshold are converted to 1, while anything falling below it is converted to a 0 during activation.
  • width – a positive number describing the width for the radial basis function unit stored either in Neuron element or in NeuralLayer or even in NeuralNetwork.
  • altitude – a positive number stored in Neuron or NeuralLayer or NeuralNetwork. The default is altitude=”1.0”, for that value the activation function reduces to the simple exp(-Z)
  • Neuron – All incoming connections for a certain neuron are contained in the corresponding Neuron element
add_Neuron(value, *args)[source]
add_Neuron_wrapper(value, *args)[source]
insert_Neuron_at(index, value, *args)[source]
insert_Neuron_at_wrapper(index, value, *args)[source]
set_Neuron(Neuron, *args)[source]
set_Neuron_wrapper(Neuron, *args)[source]
class PMML43Ext.NeuralNetwork(modelName=None, functionName=None, algorithmName=None, activationFunction=None, normalizationMethod='none', threshold='0', width=None, altitude='1.0', numberOfLayers=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, NeuralInputs=None, NeuralLayer=None, NeuralOutputs=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.NeuralNetwork

this Defines the structure of Neural Networks

Parameters:
  • modelName – this stores Name of the Model
  • functionName – Stores what type of problems it is ex classification or regression
  • algorithmName – Stores algorithm name used in the model
  • activationFunction – Stores the activation function used in building model
  • isScorable – indicates whether the model is valid for scoring
  • MiningSchema – Stores the MiningField in building PMML
  • Output – Stores the output field
  • NeuralInputs – defines how input fields are normalized so that the values can be processed in the neural network
  • NeuralLayer – defines how neurons are stroed in different NeuralLayer
  • NeuralOutputs – defines how the output of the neural network must be interpreted
add_NeuralLayer(value, *args)[source]
add_NeuralLayer_wrapper(value, *args)[source]
insert_NeuralLayer_at(index, value, *args)[source]
insert_NeuralLayer_at_wrapper(index, value, *args)[source]
set_NeuralLayer(NeuralLayer, *args)[source]
set_NeuralLayer_wrapper(NeuralLayer, *args)[source]
class PMML43Ext.NeuralOutput(outputNeuron=None, Extension=None, DerivedField=None)[source]

Bases: PMML43ExtSuper.NeuralOutput

defines how the output of the neural network must be interpreted

Parameters:
  • outputNeuron – It represent the id’s of output neuron in output layer
  • DerivedField – which provides a common element for the various mappings. They can also appear at several places in the definition of specific models such as neural network or Naïve Bayes models
class PMML43Ext.NeuralOutputs(numberOfOutputs=None, Extension=None, NeuralOutput=None)[source]

Bases: PMML43ExtSuper.NeuralOutputs

It stores the whole part of output layer

Parameters:
  • numberOfOutputs – Represents the number of output neurons in Output layer
  • NeuralOutput – defines how the output of the neural network must be interpreted
add_NeuralOutput(value, *args)[source]
add_NeuralOutput_wrapper(value, *args)[source]
insert_NeuralOutput_at(index, value, *args)[source]
insert_NeuralOutput_at_wrapper(index, value, *args)[source]
set_NeuralOutput(NeuralOutput, *args)[source]
set_NeuralOutput_wrapper(NeuralOutput, *args)[source]
class PMML43Ext.Neuron(id=None, bias=None, width=None, altitude=None, Extension=None, Con=None)[source]

Bases: PMML43ExtSuper.Neuron

All incoming connections for a certain neuron are contained in the corresponding Neuron element

Parameters:
  • id – an identifier which must be unique in all layers
  • bias – The attribute bias implicitly defines a connection to a bias unit where the unit’s value is 1.0 and the weight is the value of bias
  • width – a positive number describing the width for the radial basis function unit stored either in Neuron element or in NeuralLayer or even in NeuralNetwork
  • altitude – a positive number stored in Neuron or NeuralLayer or NeuralNetwork
  • Con – Each connection Con of the element Neuron stores the ID of a node it comes from and the weight
class PMML43Ext.Node(id=None, score=None, recordCount=None, defaultChild=None, SimplePredicate=None, CompoundPredicate=None, SimpleSetPredicate=None, True_=None, False_=None, Partition=None, ScoreDistribution=None, Node_member=None, Extension=None, Regression=None, DecisionTree=None)[source]

Bases: PMML43ExtSuper.Node

Node is an encapsulation for either defining a split or a leaf in a tree model. Every Node contains a predicate that identifies a rule for choosing itself or any of its siblings

Parameters:
  • id – The value of id serves as a unique identifier for any given Node within the tree model
  • score – the predicted value for a record that chooses the Node
  • recordCount – allow to determine the relative size of given values in a ScoreDistribution as well as the relative size of a Node when compared to the parent Node
  • defaultChild – Only applicable when missingValueStrategy is set to defaultChild in the TreeModel element
  • SimplePredicate – defines a rule in the form of a simple boolean expression. The rule consists of field, operator (booleanOperator) for binary comparison, and value
  • CompoundPredicate – an encapsulating element for combining two or more elements as defined at the entity PREDICATE
  • SimpleSetPredicate – checks whether a field value is element of a set. The set of values is specified by the array
  • True – a predicate element that identifies the boolean constant TRUE
  • False – a predicate element that identifies the boolean constant False
  • Partition – Optional element to provide distribution information for all records that belong to the respective Node
  • ScoreDistribution – an element of Node to represent segments of the score that a Node predicts in a classification model. If the Node holds an enumeration, each entry of the enumeration is stored in one ScoreDistribution element
class PMML43Ext.NonseasonalComponent(p=None, d=None, q=None, Extension=None, AR=None, MA=None)[source]

Bases: PMML43ExtSuper.NonseasonalComponent

class PMML43Ext.NonseasonalFactor(difference='0', maximumOrder=None, Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.NonseasonalFactor

class PMML43Ext.NormContinuous(mapMissingTo=None, field=None, outliers='asIs', Extension=None, LinearNorm=None)[source]

Bases: PMML43ExtSuper.NormContinuous

Used to implement simple normalization functions such as the z-score transformation (X - m ) / s, where m is the mean value and s is the standard deviation.

Parameters:
  • mapMissingTo – may be used to map a missing result to the value specified by the attribute. If the attribute is not present, the result remains missing
  • outliers – it shows the records which are out of trend
  • LinearNorm – defines a sequence of points for a stepwise linear interpolation function,LinearNorm must be strictly sorted by ascending value of orig
class PMML43Ext.NormDiscrete(field=None, value=None, mapMissingTo=None, Extension=None)[source]

Bases: PMML43ExtSuper.NormDiscrete

Refer to a certain input field define a fan-out function which maps a single input field to a set of normalized fields.

Parameters:mapMissingTo – may be used to map a missing result to the value specified by the attribute. If the attribute is not present, the result remains missing
class PMML43Ext.NormalDistributionForBN(Extension=None, Mean=None, Variance=None)[source]

Bases: PMML43ExtSuper.NormalDistributionForBN

class PMML43Ext.Numerator(Extension=None, NonseasonalFactor=None, SeasonalFactor=None)[source]

Bases: PMML43ExtSuper.Numerator

class PMML43Ext.NumericInfo(minimum=None, maximum=None, mean=None, standardDeviation=None, median=None, interQuartileRange=None, Extension=None, Quantile=None)[source]

Bases: PMML43ExtSuper.NumericInfo

class PMML43Ext.NumericPredictor(name=None, exponent='1', coefficient=None, Extension=None)[source]

Bases: PMML43ExtSuper.NumericPredictor

Defines a numeric independent variable. The list of valid attributes comprises the name of the variable, the exponent to be used, and the coefficient by which the values of this variable must be multiplied. Note that the exponent defaults to 1, hence it is not always necessary to specify. Also, if the input value is missing, the result evaluates to a missing value.

class PMML43Ext.Optimizers(clipnorm=None, clipvalue=None, Extension=None, SGD=None, RMSprop=None, Adagrad=None, Adadelta=None, Adam=None, Adamax=None, Nadam=None)[source]

Bases: PMML43ExtSuper.Optimizers

class PMML43Ext.OptimumLiftGraph(Extension=None, LiftGraph=None)[source]

Bases: PMML43ExtSuper.OptimumLiftGraph

class PMML43Ext.OutlierEffect(type_=None, startTime=None, magnitude=None, dampingCoefficient=None, Extension=None)[source]

Bases: PMML43ExtSuper.OutlierEffect

class PMML43Ext.Output(Extension=None, OutputField=None)[source]

Bases: PMML43ExtSuper.Output

describes a set of result values that can be returned from a model.

Parameters:OutputField – OutputField elements specify names, types and rules for calculating specific result features
class PMML43Ext.OutputField(name=None, displayName=None, optype=None, dataType=None, targetField=None, feature='predictedValue', value=None, numTopCategories=None, ruleFeature='consequent', algorithm='exclusiveRecommendation', rank='1', rankBasis='confidence', rankOrder='descending', isMultiValued='0', segmentId=None, isFinalResult=True, Extension=None, Decisions=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.OutputField

elements specify names, types and rules for calculating specific result features.

Parameters:
  • name – specifies the name of a the OutputField
  • optype – indicate admissible operations on the values. A clusterId field, for example, can have integer as its dataType, but categorical as its opType. For details, see the description of DataDictionary
  • dataType – specifies the data type for the output column
  • targetField – must refer either to a MiningField of usage type target or a field described in Targets element
  • feature – specifies the value the output field takes from the computed mining result
  • value – used in conjunction with result features referring to specific values
  • ruleFeature – specifies which feature of an association rule to return
  • algorithm – specifies which scoring algorithm to use when computing the output value
  • rank – specify the rank of the feature value from the mining result that should be selected
  • rankBasis – specify which criterion is used to sort the output result
  • rankOrder – determines the sorting order when ranking the results. The default behavior (rankOrder=”descending”) indicates that the result with the highest rank will appear first on the sorted list
  • isMultiValued – indicates that the output can represent multiple output values
  • segmentId – applicable to MiningModels which utilize Segmentation
  • isFinalResult – indicate whether the result should be returned to the user or is only used as input to another OutputField that descrbes a transformed value
  • Decisions – Derive a decision from the output of a data mining model
  • FieldRef – Field references are simply pass-throughs to fields previously defined in the DataDictionary, a DerivedField, or a result field
  • Constant – used in expressions which have multiple arguments. The actual value of a constant is given by the content of the element
  • NormContinuous – defines how to normalize an input field by piecewise linear interpolation
  • NormDiscrete – refer to a certain input field define a fan-out function which maps a single input field to a set of normalized fields
  • Discretize – Discretization of numerical input fields is a mapping from continuous to discrete values using intervals
  • MapValues – element can be used to create missing value indicators for categorical variables
  • TextIndex – TextIndex expression to extract frequency information from the text input field, for a given term. The TextIndex element fully configures how the text input should be indexed, including case sensitivity, normalization and other settings
  • Aggregate – summarize or collect groups of values, e.g., compute average
  • Lag – defined as the value of the given input field a fixed number of records prior to the current one,If the desired value is not present, for a given record, the lag will be set to missing
exportAttributes(outfile, level, already_processed, namespace_='', name_='OutputFields', *args)[source]
exportAttributes_wrapper(outfile, level, already_processed, namespace_='', name_='OutputFields', *args)[source]
class PMML43Ext.PCell(targetCategory=None, parameterName=None, beta=None, df=None, Extension=None)[source]

Bases: PMML43ExtSuper.PCell

class PMML43Ext.PCovCell(pRow=None, pCol=None, tRow=None, tCol=None, value=None, targetCategory=None, Extension=None)[source]

Bases: PMML43ExtSuper.PCovCell

class PMML43Ext.PCovMatrix(type_=None, Extension=None, PCovCell=None)[source]

Bases: PMML43ExtSuper.PCovMatrix

class PMML43Ext.PMML(version='4.3', Header=None, script=None, MiningBuildTask=None, DataDictionary=None, TransformationDictionary=None, AssociationModel=None, BayesianNetworkModel=None, BaselineModel=None, ClusteringModel=None, DeepNetwork=None, GaussianProcessModel=None, GeneralRegressionModel=None, MiningModel=None, NaiveBayesModel=None, NearestNeighborModel=None, NeuralNetwork=None, RegressionModel=None, RuleSetModel=None, SequenceModel=None, Scorecard=None, SupportVectorMachineModel=None, TextModel=None, TimeSeriesModel=None, TreeModel=None, Extension=None)[source]

Bases: PMML43ExtSuper.PMML

this is the root of the pmml document

export(outfile, level, namespace_='', name_='PMML', namespacedef_='', pretty_print=True, *args)[source]
export_wrapper(outfile, level, namespace_='', name_='PMML', namespacedef_='', pretty_print=True, *args)[source]
class PMML43Ext.PPCell(value=None, predictorName=None, parameterName=None, targetCategory=None, Extension=None)[source]

Bases: PMML43ExtSuper.PPCell

class PMML43Ext.PPMatrix(Extension=None, PPCell=None)[source]

Bases: PMML43ExtSuper.PPMatrix

class PMML43Ext.PairCounts(value=None, Extension=None, TargetValueCounts=None)[source]

Bases: PMML43ExtSuper.PairCounts

class PMML43Ext.ParamMatrix(Extension=None, PCell=None)[source]

Bases: PMML43ExtSuper.ParamMatrix

class PMML43Ext.Parameter(name=None, label=None, referencePoint='0', Extension=None)[source]

Bases: PMML43ExtSuper.Parameter

class PMML43Ext.ParameterField(name=None, optype=None, dataType=None)[source]

Bases: PMML43ExtSuper.ParameterField

class PMML43Ext.ParameterList(Extension=None, Parameter=None)[source]

Bases: PMML43ExtSuper.ParameterList

class PMML43Ext.ParentValue(parent=None, value=None, Extension=None)[source]

Bases: PMML43ExtSuper.ParentValue

class PMML43Ext.Partition(name=None, size=None, Extension=None, PartitionFieldStats=None)[source]

Bases: PMML43ExtSuper.Partition

class PMML43Ext.PartitionFieldStats(field=None, weighted='0', Extension=None, Counts=None, NumericInfo=None, Array=None)[source]

Bases: PMML43ExtSuper.PartitionFieldStats

class PMML43Ext.PastVariances(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.PastVariances

class PMML43Ext.PoissonDistribution(mean=None, Extension=None)[source]

Bases: PMML43ExtSuper.PoissonDistribution

class PMML43Ext.PolynomialKernelType(description=None, gamma='1', coef0='1', degree='1', Extension=None)[source]

Bases: PMML43ExtSuper.PolynomialKernelType

class PMML43Ext.PredictiveModelQuality(targetField=None, dataName=None, dataUsage='training', meanError=None, meanAbsoluteError=None, meanSquaredError=None, rootMeanSquaredError=None, r_squared=None, adj_r_squared=None, sumSquaredError=None, sumSquaredRegression=None, numOfRecords=None, numOfRecordsWeighted=None, numOfPredictors=None, degreesOfFreedom=None, fStatistic=None, AIC=None, BIC=None, AICc=None, Extension=None, ConfusionMatrix=None, LiftData=None, ROC=None)[source]

Bases: PMML43ExtSuper.PredictiveModelQuality

class PMML43Ext.Predictor(name=None, contrastMatrixType=None, Extension=None, Categories=None, Matrix=None)[source]

Bases: PMML43ExtSuper.Predictor

class PMML43Ext.PredictorTerm(name=None, coefficient=None, Extension=None, FieldRef=None)[source]

Bases: PMML43ExtSuper.PredictorTerm

Contains one or more fields that are combined by multiplication. That is, this element supports interaction terms. The type of all fields referenced within PredictorTerm must be continuous. Note that if the input value is missing, the result evaluates to a missing value. The name attribute allows this term to be referenced by elements of Statistics and should be unique from any other field names with the scope of this RegressionModel. The content of PredictorTerm might be extended to a sequence of any expression. This feature is not yet needed.

class PMML43Ext.PsiVector(targetField=None, variance=None, Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.PsiVector

class PMML43Ext.Quantile(quantileLimit=None, quantileValue=None, Extension=None)[source]

Bases: PMML43ExtSuper.Quantile

class PMML43Ext.REAL_SparseArray(n=None, defaultValue='0', Indices=None, REAL_Entries=None)[source]

Bases: PMML43ExtSuper.REAL_SparseArray

class PMML43Ext.RMSprop(learningRate=None, rho=None, decayRate=None, epsilon=None, Extension=None)[source]

Bases: PMML43ExtSuper.RMSprop

class PMML43Ext.ROC(positiveTargetFieldValue=None, positiveTargetFieldDisplayValue=None, negativeTargetFieldValue=None, negativeTargetFieldDisplayValue=None, Extension=None, ROCGraph=None)[source]

Bases: PMML43ExtSuper.ROC

class PMML43Ext.ROCGraph(Extension=None, XCoordinates=None, YCoordinates=None, BoundaryValues=None)[source]

Bases: PMML43ExtSuper.ROCGraph

class PMML43Ext.RadialBasisKernel(description=None, gamma='1', noiseVariance='1', lambda_='1', Extension=None)[source]

Bases: PMML43ExtSuper.RadialBasisKernel

class PMML43Ext.RadialBasisKernelType(description=None, gamma='1', Extension=None)[source]

Bases: PMML43ExtSuper.RadialBasisKernelType

class PMML43Ext.RandomLiftGraph(Extension=None, LiftGraph=None)[source]

Bases: PMML43ExtSuper.RandomLiftGraph

class PMML43Ext.Regression(modelName=None, functionName=None, algorithmName=None, normalizationMethod='none', Extension=None, Output=None, ModelStats=None, Targets=None, LocalTransformations=None, ResultField=None, RegressionTable=None)[source]

Bases: PMML43ExtSuper.Regression

class PMML43Ext.RegressionModel(modelName=None, functionName=None, algorithmName=None, modelType=None, targetFieldName=None, normalizationMethod='none', isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, RegressionTable=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.RegressionModel

The root element of an XML regression model. Each instance of a regression model must start with this element

Parameters:
  • modelName – This is a unique identifier specifying the name of the regression model
  • functionName – Can be regression or classification
  • algorithmName – Can be any string describing the algorithm that was used while creating the model
  • modelType – Specifies the type of a regression model. The attribute modelType is for information only. It has been changed to optional and the usage is deprecated. Use functionName and normalizationMethod in order to define the computation. Use algorithmName in order to give further optional information
  • targetFieldName – The name of the target field (also called dependent variable). The attribute targetFieldName is for information only. It has been changed to optional and the usage is deprecated. Use usageType=”target” in MiningField instead
  • isScorable – This attribute indicates if the model is valid for scoring. If this attribute is true or if it is missing, then the model should be processed normally. However, if the attribute is false, then the model producer has indicated that this model is intended for information purposes only and should not be used to generate results. In order to be valid PMML, all required elements and attributes must be present, even for non-scoring models
  • RegressionTable – A table that lists the values of all predictors or independent variables. If the model is used to predict a numerical field, then there is only one RegressionTable and the attribute targetCategory may be missing. If the model is used to predict a categorical field, then there are two or more RegressionTables and each one must have the attribute targetCategory defined with a unique value
class PMML43Ext.RegressionTable(intercept=None, targetCategory=None, Extension=None, NumericPredictor=None, CategoricalPredictor=None, PredictorTerm=None)[source]

Bases: PMML43ExtSuper.RegressionTable

A table that lists the values of all predictors or independent variables. If the model is used to predict a numerical field, then there is only one RegressionTable and the attribute targetCategory may be missing. If the model is used to predict a categorical field, then there are two or more RegressionTables and each one must have the attribute targetCategory defined with a unique value

Parameters:
  • NumericPredictor – Defines a numeric independent variable. The list of valid attributes comprises the name of the variable, the exponent to be used, and the coefficient by which the values of this variable must be multiplied. Note that the exponent defaults to 1, hence it is not always necessary to specify. Also, if the input value is missing, the result evaluates to a missing value
  • CategoricalPredictor – Defines a categorical independent variable. The list of attributes comprises the name of the variable, the value attribute, and the coefficient by which the values of this variable must be multiplied. To do a regression analysis with categorical values, some means must be applied to enable calculations. If the specified value of an independent value occurs, the term variable_name(value) is replaced with 1. Thus the coefficient is multiplied by 1. If the value does not occur, the term variable_name(value) is replaced with 0 so that the product coefficient × variable_name(value) yields 0. Consequently, the product is ignored in the ongoing analysis. If the input value is missing then variable_name(v) yields 0 for any v
  • PredictorTerm – Contains one or more fields that are combined by multiplication. That is, this element supports interaction terms. The type of all fields referenced within PredictorTerm must be continuous. Note that if the input value is missing, the result evaluates to a missing value. The name attribute allows this term to be referenced by elements of Statistics and should be unique from any other field names with the scope of this RegressionModel The content of PredictorTerm might be extended to a sequence of any expression. This feature is not yet needed
class PMML43Ext.RegressorValues(Extension=None, TimeSeries=None, TrendCoefficients=None, TransferFunctionValues=None)[source]

Bases: PMML43ExtSuper.RegressorValues

class PMML43Ext.ResidualSquareCoefficients(Extension=None, Residuals=None, Coefficients=None)[source]

Bases: PMML43ExtSuper.ResidualSquareCoefficients

class PMML43Ext.Residuals(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.Residuals

class PMML43Ext.ResultField(name=None, displayName=None, optype=None, dataType=None, feature=None, value=None, Extension=None)[source]

Bases: PMML43ExtSuper.ResultField

class PMML43Ext.RuleSelectionMethod(criterion=None, Extension=None)[source]

Bases: PMML43ExtSuper.RuleSelectionMethod

class PMML43Ext.RuleSet(recordCount=None, nbCorrect=None, defaultScore=None, defaultConfidence=None, Extension=None, RuleSelectionMethod=None, ScoreDistribution=None, SimpleRule=None, CompoundRule=None)[source]

Bases: PMML43ExtSuper.RuleSet

class PMML43Ext.RuleSetModel(modelName=None, functionName=None, algorithmName=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, RuleSet=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.RuleSetModel

class PMML43Ext.SGD(learningRate=None, momentum=None, decayRate=None, nesterov=None, Extension=None)[source]

Bases: PMML43ExtSuper.SGD

class PMML43Ext.ScoreDistribution(value=None, recordCount=None, confidence=None, probability=None, Extension=None)[source]

Bases: PMML43ExtSuper.ScoreDistribution

an element of Node to represent segments of the score that a Node predicts in a classification model. If the Node holds an enumeration, each entry of the enumeration is stored in one ScoreDistribution element

Parameters:
  • value – This attribute of ScoreDistribution is the label in a classification model
  • recordCount – This attribute of ScoreDistribution is the size (in number of records) associated with the value attribute
  • confidence – This optional attribute of ScoreDistribution assigns a confidence to a given prediction class for this tree node , Confidences are similar to probabilities but more relaxed The confidences may not necessarily sum to 1 across the different classes, like probabilities would. Confidences should normally lie in the range 0.0 to 1.0 though
  • probability – This optional attribute assigns a predicted probability for the given value within the node
class PMML43Ext.Scorecard(modelName=None, functionName=None, algorithmName=None, initialScore='0', useReasonCodes=True, reasonCodeAlgorithm='pointsBelow', baselineScore=None, baselineMethod='other', isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, Characteristics=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.Scorecard

class PMML43Ext.SeasonalComponent(P=None, D=None, Q=None, period=None, Extension=None, AR=None, MA=None)[source]

Bases: PMML43ExtSuper.SeasonalComponent

class PMML43Ext.SeasonalFactor(difference='0', maximumOrder=None, Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.SeasonalFactor

class PMML43Ext.SeasonalTrendDecomposition[source]

Bases: PMML43ExtSuper.SeasonalTrendDecomposition

class PMML43Ext.Seasonality_ExpoSmooth(type_=None, period=None, initialSeasonalTrendValue=None, unit=None, phase=None, delta=None, Array=None)[source]

Bases: PMML43ExtSuper.Seasonality_ExpoSmooth

class PMML43Ext.Segment(id=None, weight='1', Extension=None, SimplePredicate=None, CompoundPredicate=None, SimpleSetPredicate=None, True_=None, False_=None, AssociationModel=None, BayesianNetworkModel=None, BaselineModel=None, ClusteringModel=None, DeepNetwork=None, GaussianProcessModel=None, GeneralRegressionModel=None, MiningModel=None, NaiveBayesModel=None, NearestNeighborModel=None, NeuralNetwork=None, RegressionModel=None, RuleSetModel=None, SequenceModel=None, Scorecard=None, SupportVectorMachineModel=None, TextModel=None, TimeSeriesModel=None, TreeModel=None)[source]

Bases: PMML43ExtSuper.Segment

Segment includes a PREDICATE element specifying the conditions under which that segment is to be used

Parameters:
  • id – The value of id serves as a unique identifier for any given Node within the tree model
  • SimplePredicate – defines a rule in the form of a simple boolean expression. The rule consists of field, operator (booleanOperator) for binary comparison, and value
  • CompoundPredicate – an encapsulating element for combining two or more elements as defined at the entity PREDICATE
  • SimpleSetPredicate – checks whether a field value is element of a set. The set of values is specified by the array
  • True – a predicate element that identifies the boolean constant TRUE
  • False – a predicate element that identifies the boolean constant False
  • TreeModel – TreeModel in PMML allows for defining either a classification or prediction structure
class PMML43Ext.Segmentation(multipleModelMethod=None, Extension=None, Segment=None)[source]

Bases: PMML43ExtSuper.Segmentation

Segmentation allows representation of different models for different data segments and also can be used for model ensembles and model sequences

Parameters:
  • multipleModelMethod – specifying how all the models applicable to a record should be combined
  • Segment – Segment includes a PREDICATE element specifying the conditions under which that segment is to be used
class PMML43Ext.Sequence(id=None, numberOfSets=None, occurrence=None, support=None, Extension=None, Delimiter=None, SetReference=None, Time=None)[source]

Bases: PMML43ExtSuper.Sequence

class PMML43Ext.SequenceModel(modelName=None, functionName=None, algorithmName=None, numberOfTransactions=None, maxNumberOfItemsPerTransaction=None, avgNumberOfItemsPerTransaction=None, numberOfTransactionGroups=None, maxNumberOfTAsPerTAGroup=None, avgNumberOfTAsPerTAGroup=None, isScorable=True, MiningSchema=None, ModelStats=None, LocalTransformations=None, Constraints=None, Item=None, Itemset=None, SetPredicate=None, Sequence=None, SequenceRule=None, Extension=None)[source]

Bases: PMML43ExtSuper.SequenceModel

class PMML43Ext.SequenceReference(seqId=None, Extension=None)[source]

Bases: PMML43ExtSuper.SequenceReference

class PMML43Ext.SequenceRule(id=None, numberOfSets=None, occurrence=None, support=None, confidence=None, lift=None, Extension=None, AntecedentSequence=None, Delimiter=None, ConsequentSequence=None, Time=None)[source]

Bases: PMML43ExtSuper.SequenceRule

class PMML43Ext.SetPredicate(id=None, field=None, operator=None, Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.SetPredicate

class PMML43Ext.SetReference(setId=None, Extension=None)[source]

Bases: PMML43ExtSuper.SetReference

class PMML43Ext.SigmoidKernelType(description=None, gamma='1', coef0='1', Extension=None)[source]

Bases: PMML43ExtSuper.SigmoidKernelType

class PMML43Ext.SimplePredicate(field=None, operator=None, value=None, Extension=None)[source]

Bases: PMML43ExtSuper.SimplePredicate

defines a rule in the form of a simple boolean expression. The rule consists of field, operator (booleanOperator) for binary comparison, and value

Parameters:
  • field – This attribute of the SimplePredicate element is the name attribute of a MiningField or a DerivedField from TransformationDictionary or LocalTransformations
  • operator
    this attribute of SimplePredicate is one of the six pre-defined comparison operators
    Operator Math Symbol equal = notEqual ≠ lessThan <

    lessOrEqual ≤ greaterThan >

  • (greaterOrEqual) –
  • value – This attribute of SimplePredicate element is the information to evaluate / compare against
class PMML43Ext.SimpleRule(id=None, score=None, recordCount=None, nbCorrect=None, confidence='1', weight='1', Extension=None, SimplePredicate=None, CompoundPredicate=None, SimpleSetPredicate=None, True_=None, False_=None, ScoreDistribution=None)[source]

Bases: PMML43ExtSuper.SimpleRule

class PMML43Ext.SimpleSetPredicate(field=None, booleanOperator=None, Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.SimpleSetPredicate

checks whether a field value is element of a set. The set of values is specified by the array

Parameters:
  • booleanOperator – can take one of following boolean operators: isIn, and isNotIn
  • Array – The set of values is specified by the array in the content
class PMML43Ext.SpectralAnalysis[source]

Bases: PMML43ExtSuper.SpectralAnalysis

class PMML43Ext.StateSpaceModel(variance=None, period='none', intercept='0', Extension=None, StateVector=None, TransitionMatrix=None, MeasurementMatrix=None, PsiVector=None, DynamicRegressor=None)[source]

Bases: PMML43ExtSuper.StateSpaceModel

class PMML43Ext.StateVector(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.StateVector

class PMML43Ext.SupportVector(vectorId=None, Extension=None)[source]

Bases: PMML43ExtSuper.SupportVector

class PMML43Ext.SupportVectorMachine(targetCategory=None, alternateTargetCategory=None, threshold=None, Extension=None, SupportVectors=None, Coefficients=None)[source]

Bases: PMML43ExtSuper.SupportVectorMachine

The description of Support Vector Machine (SVM) models assumes some familiarity with the SVM theory. In this specification, Support Vector Machine models for classification and regression are considered. A Support Vector Machine is a function f which is defined in the space spanned by the kernel basis functions K(x,xi) of the support vectors xi

Parameters:
  • targetCategory – The attribute targetCategory is required for classification models and gives the corresponding class label. This attribute is to be used for classification models implementing the one-against-all method. In this method, for n classes, there are exactly n SupportVectorMachine elements. Depending on the model attribute maxWins, the SVM with the largest or the smallest value determines the predicted class label
  • alternateTargetCategory – The attribute alternateTargetCategory is required in case of binary classification models with only one SupportVectorMachine element. It is also required in case of multi-class classification models implementing the one-against-one method
  • threshold – The attribute threshold defines a discrimination boundary to be used in case of binary classification or whenever attribute classificationMethod is defined as OneAgainstOne for multi-class classification tasks
  • SupportVectors – The term Support Vector (SV) has also a geometrical interpretation because these vectors really support the discrimination function f(x) = 0 in the mechanical interpretation
  • Coefficients – Each coefficient αi is described by the element Coefficient and the number of coefficients corresponds to that of the support vectors. Hence the attribute numberOfCoefficients is equal to the number of support vectors. The attribute absoluteValue contains the value of the absolute coefficient b
class PMML43Ext.SupportVectorMachineModel(modelName=None, functionName=None, algorithmName=None, threshold='0', svmRepresentation='SupportVectors', classificationMethod='OneAgainstAll', maxWins=False, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, LinearKernelType=None, PolynomialKernelType=None, RadialBasisKernelType=None, SigmoidKernelType=None, VectorDictionary=None, SupportVectorMachine=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.SupportVectorMachineModel

class PMML43Ext.SupportVectors(numberOfSupportVectors=None, numberOfAttributes=None, Extension=None, SupportVector=None)[source]

Bases: PMML43ExtSuper.SupportVectors

add_SupportVector(value, *args)[source]
add_SupportVector_wrapper(value, *args)[source]
insert_SupportVector_at(index, value, *args)[source]
insert_SupportVector_at_wrapper(index, value, *args)[source]
set_SupportVector(SupportVector, *args)[source]
set_SupportVector_wrapper(SupportVector, *args)[source]
class PMML43Ext.TableLocator(Extension=None)[source]

Bases: PMML43ExtSuper.TableLocator

class PMML43Ext.Target(field=None, optype=None, castInteger=None, min=None, max=None, rescaleConstant=0, rescaleFactor=1, Extension=None, TargetValue=None)[source]

Bases: PMML43ExtSuper.Target

class PMML43Ext.TargetValue(value=None, displayValue=None, priorProbability=None, defaultValue=None, Extension=None, Partition=None)[source]

Bases: PMML43ExtSuper.TargetValue

class PMML43Ext.TargetValueCount(value=None, count=None, Extension=None)[source]

Bases: PMML43ExtSuper.TargetValueCount

class PMML43Ext.TargetValueCounts(Extension=None, TargetValueCount=None)[source]

Bases: PMML43ExtSuper.TargetValueCounts

TargetValueCounts lists the counts associated with each value of the target field.

class PMML43Ext.TargetValueStat(value=None, AnyDistribution=None, GaussianDistribution=None, PoissonDistribution=None, UniformDistribution=None, Extension=None)[source]

Bases: PMML43ExtSuper.TargetValueStat

class PMML43Ext.TargetValueStats(Extension=None, TargetValueStat=None)[source]

Bases: PMML43ExtSuper.TargetValueStats

TargetValueStats serves as the envelope for element TargetValueStat. It is used for a continuous input field Ii to define statistical measures associated with each value of the target field.

class PMML43Ext.Targets(Extension=None, Target=None)[source]

Bases: PMML43ExtSuper.Targets

class PMML43Ext.Taxonomy(name=None, Extension=None, ChildParent=None)[source]

Bases: PMML43ExtSuper.Taxonomy

class PMML43Ext.TestDistributions(field=None, testStatistic=None, resetValue='0.0', windowSize='0', weightField=None, normalizationScheme=None, Baseline=None, Alternate=None, Extension=None)[source]

Bases: PMML43ExtSuper.TestDistributions

class PMML43Ext.TextCorpus(Extension=None, TextDocument=None)[source]

Bases: PMML43ExtSuper.TextCorpus

class PMML43Ext.TextDictionary(Extension=None, Taxonomy=None, Array=None)[source]

Bases: PMML43ExtSuper.TextDictionary

class PMML43Ext.TextDocument(id=None, name=None, length=None, file=None, Extension=None)[source]

Bases: PMML43ExtSuper.TextDocument

class PMML43Ext.TextIndex(textField=None, localTermWeights='termFrequency', isCaseSensitive=False, maxLevenshteinDistance=0, countHits='allHits', wordSeparatorCharacterRE='\s', tokenize=True, Extension=None, TextIndexNormalization=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex_member=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.TextIndex

TextIndex expression to extract frequency information from the text input field, for a given term. The TextIndex element fully configures how the text input should be indexed, including case sensitivity, normalization and other settings,

Parameters:
  • FieldRef – Field references are simply pass-throughs to fields previously defined in the DataDictionary, a DerivedField, or a result field
  • Constant – used in expressions which have multiple arguments. The actual value of a constant is given by the content of the element
  • NormContinuous – defines how to normalize an input field by piecewise linear interpolation
  • NormDiscrete – refer to a certain input field define a fan-out function which maps a single input field to a set of normalized fields
  • Discretize – Discretization of numerical input fields is a mapping from continuous to discrete values using intervals
  • MapValues – element can be used to create missing value indicators for categorical variables
  • Aggregate – summarize or collect groups of values, e.g., compute average
  • Lag – defined as the value of the given input field a fixed number of records prior to the current one,If the desired value is not present, for a given record, the lag will be set to missing
class PMML43Ext.TextIndexNormalization(inField='string', outField='stem', regexField='regex', recursive=False, isCaseSensitive=None, maxLevenshteinDistance=None, wordSeparatorCharacterRE=None, tokenize=None, Extension=None, TableLocator=None, InlineTable=None)[source]

Bases: PMML43ExtSuper.TextIndexNormalization

class PMML43Ext.TextModel(modelName=None, functionName=None, algorithmName=None, numberOfTerms=None, numberOfDocuments=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, TextDictionary=None, TextCorpus=None, DocumentTermMatrix=None, TextModelNormalization=None, TextModelSimiliarity=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.TextModel

class PMML43Ext.TextModelNormalization(localTermWeights='termFrequency', globalTermWeights='inverseDocumentFrequency', documentNormalization='none', Extension=None)[source]

Bases: PMML43ExtSuper.TextModelNormalization

class PMML43Ext.TextModelSimiliarity(similarityType=None, Extension=None)[source]

Bases: PMML43ExtSuper.TextModelSimiliarity

class PMML43Ext.Theta(i=None, j=None, theta=None)[source]

Bases: PMML43ExtSuper.Theta

class PMML43Ext.ThetaRecursionState(FinalNoise=None, FinalPredictedNoise=None, FinalTheta=None, FinalNu=None)[source]

Bases: PMML43ExtSuper.ThetaRecursionState

class PMML43Ext.Time(min=None, max=None, mean=None, standardDeviation=None, Extension=None)[source]

Bases: PMML43ExtSuper.Time

class PMML43Ext.TimeAnchor(type_=None, offset=None, stepsize=None, displayName=None, TimeCycle=None, TimeException=None)[source]

Bases: PMML43ExtSuper.TimeAnchor

class PMML43Ext.TimeCycle(length=None, type_=None, displayName=None, Array=None)[source]

Bases: PMML43ExtSuper.TimeCycle

class PMML43Ext.TimeException(type_=None, count=None, Array=None)[source]

Bases: PMML43ExtSuper.TimeException

class PMML43Ext.TimeSeries(usage='original', startTime=None, endTime=None, interpolationMethod='none', TimeAnchor=None, TimeValue=None)[source]

Bases: PMML43ExtSuper.TimeSeries

class PMML43Ext.TimeSeriesModel(modelName=None, functionName=None, algorithmName=None, bestFit=None, isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, LocalTransformations=None, TimeSeries=None, SpectralAnalysis=None, ARIMA=None, ExponentialSmoothing=None, SeasonalTrendDecomposition=None, StateSpaceModel=None, GARCH=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.TimeSeriesModel

class PMML43Ext.TimeValue(index=None, time=None, value=None, standardError=None, Timestamp=None)[source]

Bases: PMML43ExtSuper.TimeValue

class PMML43Ext.Timestamp(content=None, Extension=None, mixedclass_=None)[source]

Bases: PMML43ExtSuper.Timestamp

export(outfile, level, namespace_='', name_='Timestamp', namespacedef_='', pretty_print=True, *args)[source]
export_wrapper(outfile, level, namespace_='', name_='Timestamp', namespacedef_='', pretty_print=True, *args)[source]
class PMML43Ext.TrainingInstances(isTransformed=False, recordCount=None, fieldCount=None, Extension=None, InstanceFields=None, TableLocator=None, InlineTable=None)[source]

Bases: PMML43ExtSuper.TrainingInstances

This element serves as an envelope for defining all of the training instances. It contains the definition of the fields included in the training instances as well as a table for representing the training data itself

Parameters:
  • isTransformed – Used as a flag to determine whether or not the training instances have already been transformed
  • recordCount – Defines the number of training instances or records. This number needs to match the number of instances defined in the element InlineTable or in the external data if TableLocator is used
  • fieldCount – Defines the number of fields (features + targets). This number needs to match the number of InstanceField elements defined under InstanceFields
  • TableLocator – Allows for the training data to be stored in an external table. Such a table can then be referenced by the TableLocator element which implements a kind of URL for tables
  • InlineTable – Allows for the training instances to be part of the PMML document itself. When used in k-NN models, a row in an InlineTable should contain a sequence of elements representing the input fields
class PMML43Ext.TrainingParameters(architectureName=None, dataset=None, framework=None, Extension=None, Losses=None, Metrics=None, Optimizers=None)[source]

Bases: PMML43ExtSuper.TrainingParameters

class PMML43Ext.TransferFunctionValues(Array=None)[source]

Bases: PMML43ExtSuper.TransferFunctionValues

class PMML43Ext.TransformationDictionary(Extension=None, DefineFunction=None, DerivedField=None)[source]

Bases: PMML43ExtSuper.TransformationDictionary

class PMML43Ext.TransitionMatrix(Extension=None, Matrix=None)[source]

Bases: PMML43ExtSuper.TransitionMatrix

class PMML43Ext.TreeModel(modelName=None, functionName=None, algorithmName=None, missingValueStrategy='none', missingValuePenalty='1.0', noTrueChildStrategy='returnNullPrediction', splitCharacteristic='multiSplit', isScorable=True, MiningSchema=None, Output=None, ModelStats=None, ModelExplanation=None, Targets=None, LocalTransformations=None, Node=None, ModelVerification=None, Extension=None)[source]

Bases: PMML43ExtSuper.TreeModel

TreeModel in PMML allows for defining either a classification or prediction structure. Each Node holds a logical predicate expression that defines the rule for choosing the Node or any of the branching Nodes.

Parameters:
  • modelName – element identifies the model with a unique name in the context of the PMML file
  • functionName – Stores what type of problems it is ex classification or regression
  • algorithmName – Stores algorithm name used in the model
  • missingValueStrategy – defines a strategy for dealing with missing values
  • missingValuePenalty – defines a penalty applied to confidence calculation when missing value handling is performed
  • noTrueChildStrategy – defines what to do in situations where scoring cannot reach a leaf node
  • splitCharacteristic – indicates whether non-leaf Nodes in the tree model have exactly two children, or an unrestricted number of children
  • isScorable – indicates whether the model is valid for scoring
  • MiningSchema – Stores the MiningField in building PMML
  • Node – this element is an encapsulation for either defining a split or a leaf in a tree model. Every Node contains a predicate that identifies a rule for choosing itself or any of its siblings
class PMML43Ext.TrendCoefficients(Extension=None, REAL_SparseArray=None)[source]

Bases: PMML43ExtSuper.TrendCoefficients

class PMML43Ext.Trend_ExpoSmooth(trend='additive', gamma=None, initialTrendValue=None, phi='1', smoothedValue=None, Array=None)[source]

Bases: PMML43ExtSuper.Trend_ExpoSmooth

class PMML43Ext.TriangularDistributionForBN(Extension=None, Mean=None, Lower=None, Upper=None)[source]

Bases: PMML43ExtSuper.TriangularDistributionForBN

class PMML43Ext.True_(Extension=None)[source]

Bases: PMML43ExtSuper.True_

class PMML43Ext.UniformDistribution(lower=None, upper=None, Extension=None)[source]

Bases: PMML43ExtSuper.UniformDistribution

class PMML43Ext.UniformDistributionForBN(Extension=None, Lower=None, Upper=None)[source]

Bases: PMML43ExtSuper.UniformDistributionForBN

class PMML43Ext.UnivariateStats(field=None, weighted='0', Extension=None, Counts=None, NumericInfo=None, DiscrStats=None, ContStats=None, Anova=None)[source]

Bases: PMML43ExtSuper.UnivariateStats

class PMML43Ext.Upper(Extension=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.Upper

class PMML43Ext.Value(value=None, displayValue=None, property='valid', Extension=None)[source]

Bases: PMML43ExtSuper.Value

class PMML43Ext.ValueProbability(value=None, probability=None, Extension=None)[source]

Bases: PMML43ExtSuper.ValueProbability

class PMML43Ext.Variance(Extension=None, Apply=None, FieldRef=None, Constant=None, NormContinuous=None, NormDiscrete=None, Discretize=None, MapValues=None, TextIndex=None, Aggregate=None, Lag=None)[source]

Bases: PMML43ExtSuper.Variance

class PMML43Ext.VarianceCoefficients(Extension=None, PastVariances=None, Coefficients=None)[source]

Bases: PMML43ExtSuper.VarianceCoefficients

class PMML43Ext.VectorDictionary(numberOfVectors=None, Extension=None, VectorFields=None, VectorInstance=None)[source]

Bases: PMML43ExtSuper.VectorDictionary

The VectorDictionary element holds all support vectors from all support vector machines

Parameters:
  • numberOfVectors – The attribute numberOfVectors must be equal to the number of vectors contained in the dictionary
  • VectorFields – VectorFields defines which entries in the vectors correspond to which fields. Note that categorical predictors are usually transformed into groups of dummy continuous variables, each having value 1 if a specific category appears in the case and 0 otherwise. Thus, one categorical field often corresponds to a group of entries in the vector
  • VectorInstance – The elements VectorInstance represent support vectors and are referenced by the id attribute. They do not contain the value of the target mining field
add_VectorInstance(value, *args)[source]
add_VectorInstance_wrapper(value, *args)[source]
insert_VectorInstance_at(index, value, *args)[source]
insert_VectorInstance_at_wrapper(index, value, *args)[source]
set_VectorInstance(VectorInstance, *args)[source]
set_VectorInstance_wrapper(VectorInstance, *args)[source]
class PMML43Ext.VectorFields(numberOfFields=None, Extension=None, FieldRef=None, CategoricalPredictor=None)[source]

Bases: PMML43ExtSuper.VectorFields

VectorFields defines which entries in the vectors correspond to which fields. Note that categorical predictors are usually transformed into groups of dummy continuous variables, each having value 1 if a specific category appears in the case and 0 otherwise. Thus, one categorical field often corresponds to a group of entries in the vector

class PMML43Ext.VectorInstance(id=None, Extension=None, REAL_SparseArray=None, Array=None)[source]

Bases: PMML43ExtSuper.VectorInstance

The elements VectorInstance represent support vectors and are referenced by the id attribute. They do not contain the value of the target mining field. The VectorInstance is a data vector given in dense or sparse array format. The order of the values corresponds to that of the VectorFields. The sizes of the sparse arrays must match the number of fields included in the VectorFields element

class PMML43Ext.VerificationField(field=None, column=None, precision=1e-06, zeroThreshold=1e-16, Extension=None)[source]

Bases: PMML43ExtSuper.VerificationField

class PMML43Ext.VerificationFields(Extension=None, VerificationField=None)[source]

Bases: PMML43ExtSuper.VerificationFields

class PMML43Ext.XCoordinates(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.XCoordinates

class PMML43Ext.YCoordinates(Extension=None, Array=None)[source]

Bases: PMML43ExtSuper.YCoordinates

class PMML43Ext.binarySimilarity(c00_parameter=None, c01_parameter=None, c10_parameter=None, c11_parameter=None, d00_parameter=None, d01_parameter=None, d10_parameter=None, d11_parameter=None, Extension=None)[source]

Bases: PMML43ExtSuper.binarySimilarity

class PMML43Ext.chebychev(Extension=None)[source]

Bases: PMML43ExtSuper.chebychev

class PMML43Ext.cityBlock(Extension=None)[source]

Bases: PMML43ExtSuper.cityBlock

class PMML43Ext.euclidean(Extension=None)[source]

Bases: PMML43ExtSuper.euclidean

PMML43Ext.get_root_tag(node)[source]
class PMML43Ext.jaccard(Extension=None)[source]

Bases: PMML43ExtSuper.jaccard

PMML43Ext.main()[source]
class PMML43Ext.minkowski(p_parameter=None, Extension=None)[source]

Bases: PMML43ExtSuper.minkowski

PMML43Ext.new_init()[source]
PMML43Ext.orig_init()[source]
PMML43Ext.parse(inFileName, silence=False)[source]
PMML43Ext.parseEtree(inFilename, silence=False)[source]
PMML43Ext.parseLiteral(inFilename, silence=False)[source]
PMML43Ext.parseString(inString, silence=False)[source]
PMML43Ext.parseSub(inFilename, silence=False)[source]
PMML43Ext.parsexml_(infile, parser=None, **kwargs)[source]
class PMML43Ext.row(anytypeobjs_=None)[source]

Bases: PMML43ExtSuper.row

class PMML43Ext.script(content=None, for_=None, class_=None, Extension=None)[source]

Bases: PMML43ExtSuper.script

export(outfile, level, namespace_='', name_='script', namespacedef_='', pretty_print=True, *args)[source]
export_wrapper(outfile, level, namespace_='', name_='script', namespacedef_='', pretty_print=True, *args)[source]
PMML43Ext.showIndent(outfile, level, pretty_print=True)[source]
class PMML43Ext.simpleMatching(Extension=None)[source]

Bases: PMML43ExtSuper.simpleMatching

class PMML43Ext.squaredEuclidean(Extension=None)[source]

Bases: PMML43ExtSuper.squaredEuclidean

class PMML43Ext.tanimoto(Extension=None)[source]

Bases: PMML43ExtSuper.tanimoto

PMML43Ext.usage()[source]