|
Class Summary |
| AbsoluteError |
The absolute error: Sum(|label-predicted|)/#examples. |
| AbstractExampleSetEvaluator |
Abstract superclass of operators accepting an ExampleSet and producing
a PerformanceVector. |
| AbstractPerformanceEvaluator |
This performance evaluator operator should be used for regression tasks,
i.e. in cases where the label attribute has a numerical value type. |
| AreaUnderCurve |
This criterion calculates the area under the ROC curve. |
| AreaUnderCurve.Neutral |
|
| AreaUnderCurve.Optimistic |
|
| AreaUnderCurve.Pessimistic |
|
| AttributeCounter |
Returns a performance vector just counting the number of attributes currently
used for the given example set. |
| BinaryClassificationPerformance |
This class encapsulates the well known binary classification criteria
precision and recall. |
| BinominalClassificationPerformanceEvaluator |
This performance evaluator operator should be used for classification tasks,
i.e. in cases where the label attribute has a binominal value type. |
| CorrelationCriterion |
Computes the empirical corelation coefficient 'r' between label and
prediction. |
| CrossEntropy |
Calculates the cross-entropy for the predictions of a classifier. |
| Data2Performance |
This operator can be used to derive a specific value of a given example set
and provide it as a performance value which can be used for optimization
purposes. |
| EstimatedPerformance |
This class is used to store estimated performance values before or
even without the performance test is actually done using a test
set. |
| LenientRelativeError |
The average relative error in a lenient way of calculation:
Sum(|label-predicted|/max(|label|, |predicted|))/#examples. |
| LogisticLoss |
The logistic loss of a classifier, defined as
the average over all ln(1 + exp(-y * f(x))) |
| Margin |
The margin of a classifier, defined as the minimal confidence for the correct label. |
| MDLCriterion |
Measures the length of an example set (i.e. the number of attributes). |
| MeasuredPerformance |
Superclass for performance citeria that are actually measured (not
estimated). |
| MinMaxCriterion |
This criterion should be used as wrapper around other performance criteria
(see MinMaxWrapper). |
| MinMaxWrapper |
Wraps a MinMaxCriterion around each performance criterion of type
MeasuredPerformance. |
| MultiClassificationPerformance |
Measures the accuracy and classification error for both binary classification
problems and multi class problems. |
| NormalizedAbsoluteError |
Normalized absolute error is the total absolute error normalized by the error
simply predicting the average of the actual values. |
| PerformanceCriterion |
Each PerformanceCriterion contains a method to compute this
criterion on a given set of examples, each which has to have a real and a
predicted label. |
| PerformanceEvaluator |
A performance evaluator is an operator that expects a test ExampleSet
as input, whose elements have both true and predicted labels, and delivers as
output a list of performance values according to a list of performance
criteria that it calculates. |
| PerformanceVector |
Handles several performance criteria. |
| PerformanceVector.DefaultComparator |
The default performance comparator compares the main criterion of two
performance vectors. |
| PolynominalClassificationPerformanceEvaluator |
This performance evaluator operator should be used for classification tasks,
i.e. in cases where the label attribute has a (poly-)nominal value type. |
| PredictionAverage |
Returns the average value of the prediction. |
| RankCorrelation |
Computes either the Spearman (rho) or Kendall (tau-b) rank
correlation between the actual label and predicted values of
an example set. |
| RankStatistics |
Provides methods to compute ranks for a single attribute and
rank correlations for two attributes. |
| RegressionPerformanceEvaluator |
This performance evaluator operator should be used for regression tasks,
i.e. in cases where the label attribute has a numerical value type. |
| RelativeError |
The average relative error: Sum(|label-predicted|/label)/#examples. |
| RootMeanSquaredError |
The root-mean-squared error. |
| RootRelativeSquaredError |
Relative squared error is the total squared error made relative to what the
error would have been if the prediction had been the average of the absolute
value. |
| SimpleClassificationError |
This class calculates the classification error without determining the complete
contingency table. |
| SimpleCriterion |
Simple criteria are those which error can be counted for each example and can
be averaged by the number of examples. |
| SimplePerformanceEvaluator |
In contrast to the other performance evaluation methods, this performance
evaluator operator can be used for all types of learning tasks. |
| SoftMarginLoss |
The soft margin loss of a classifier, defined as
the average over all 1 - y * f(x). |
| SquaredCorrelationCriterion |
Computes the square of the empirical corellation coefficient 'r' between
label and prediction. |
| SquaredError |
The squared error. |
| StrictRelativeError |
The average relative error in a strict way of calculation:
Sum(|label-predicted|/min(|label|, |predicted|))/#examples. |
| SupportVectorCounter |
Returns a performance vector just counting the number of support vectors
of a given support vector based model (kernel model). |
| UserBasedPerformanceEvaluator |
This performance evaluator operator should be used for regression tasks,
i.e. in cases where the label attribute has a numerical value type. |
| WeightedMultiClassPerformance |
Measures the weighted mean of all per class recalls or per class precisions based
on the weights defined in the performance evaluator. |
| WeightedPerformanceCreator |
Returns a performance vector containing the weighted fitness value of the
input criteria. |