ai.libs.mlpipeline_evaluation

Class ConsistentMLPipelineEvaluator

java.lang.Object

ai.libs.mlpipeline_evaluation.ConsistentMLPipelineEvaluator

public class ConsistentMLPipelineEvaluator
extends java.lang.Object

For consistent evaluations of MLPipelines.

Method Summary

Modifier and Type	Method and Description
static double	evaluateClassifier(java.lang.String splitTechnique, java.lang.String evaluationTechnique, int seed, weka.core.Instances data, weka.classifiers.Classifier classifier) Get the error rate of the classifier according to the given info about the split and evaluation technique.
static double	evaluateClassifier(java.lang.String testSplitTechnique, java.lang.String testEvaluationTechnique, int testSeed, java.lang.String valSplitTechnique, java.lang.String valEvaluationTechnique, int valSeed, weka.core.Instances data, weka.classifiers.Classifier classifier) Get the error rate of the classifier according to the given info about the split and evaluation technique.
static ai.libs.jaicore.ml.evaluation.evaluators.weka.IClassifierEvaluator	getEvaluatorForSplitTechnique(java.lang.String split_technique, weka.core.Instances data, int seed) Get an evaluator object for the given split configuration for the datasets, which can then be used to evaluate a classifier.
static weka.core.Instances	getTestSplit(java.lang.String split_technique, weka.core.Instances data, int seed) Split the dataset according to the given parameters and return the test portion of the split.
static weka.core.Instances	getTrainSplit(java.lang.String split_technique, weka.core.Instances data, int seed) Split the dataset according to the given parameters and return the train portion of the split.

Methods inherited from class java.lang.Object

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

Method Detail

evaluateClassifier

public static double evaluateClassifier(java.lang.String testSplitTechnique,
                                        java.lang.String testEvaluationTechnique,
                                        int testSeed,
                                        java.lang.String valSplitTechnique,
                                        java.lang.String valEvaluationTechnique,
                                        int valSeed,
                                        weka.core.Instances data,
                                        weka.classifiers.Classifier classifier)
                                 throws java.lang.Exception

Get the error rate of the classifier according to the given info about the split and evaluation technique.

Parameters:

testSplitTechnique -

testEvaluationTechnique -

testSeed -

valSplitTechnique -

valEvaluationTechnique -

valSeed -

data -

classifier -

Returns:

Throws:

java.lang.Exception

evaluateClassifier

public static double evaluateClassifier(java.lang.String splitTechnique,
                                        java.lang.String evaluationTechnique,
                                        int seed,
                                        weka.core.Instances data,
                                        weka.classifiers.Classifier classifier)
                                 throws java.lang.Exception

Get the error rate of the classifier according to the given info about the split and evaluation technique.

Parameters:

splitTechnique -

evaluationTechnique -

seed -

data -

classifier -

Returns:

Throws:

java.lang.Exception

getEvaluatorForSplitTechnique

public static ai.libs.jaicore.ml.evaluation.evaluators.weka.IClassifierEvaluator getEvaluatorForSplitTechnique(java.lang.String split_technique,
                                                                                                               weka.core.Instances data,
                                                                                                               int seed)

Get an evaluator object for the given split configuration for the datasets, which can then be used to evaluate a classifier.

Parameters:

split_technique -

data -

seed -

Returns:

getTrainSplit

public static weka.core.Instances getTrainSplit(java.lang.String split_technique,
                                                weka.core.Instances data,
                                                int seed)

Split the dataset according to the given parameters and return the train portion of the split.

Parameters:

split_technique -

data -

seed -

Returns:

getTestSplit

public static weka.core.Instances getTestSplit(java.lang.String split_technique,
                                               weka.core.Instances data,
                                               int seed)

Split the dataset according to the given parameters and return the test portion of the split.

Parameters:

split_technique -

data -

seed -

Returns: