Name

adams.flow.transformer.WekaBootstrapping

Synopsis

Performs bootstrapping on the incoming evaluation and outputs a spreadsheet where each row represents the results from bootstrapping sub-sample.

Additional information

Flow input/output:
- input: weka.classifiers.Evaluation, adams.flow.container.WekaEvaluationContainer
- output: adams.data.spreadsheet.SpreadSheet

Container information:
- adams.flow.container.WekaEvaluationContainer:
   - Evaluation: evaluation object; weka.classifiers.Evaluation
   - Model: model object; java.lang.Object
   - Prediction output: prediction output text; java.lang.String
   - Original indices: original indices (0-based); array of int
   - Test data: data used for testing; weka.core.Instances

Options

• loggingLevel
  

  The logging level for outputting errors and debugging output.

  command-line	-logging-level <OFF|SEVERE|WARNING|INFO|CONFIG|FINE|FINER|FINEST>
  default	WARNING
  min-user-mode	Expert

  
  
• name
  

  The name of the actor.

  command-line	-name <java.lang.String>
  default	WekaBootstrapping

  
  
• annotations
  

  The annotations to attach to this actor.

  command-line	-annotation <adams.core.base.BaseAnnotation>
  default

  
  
• skip
  

  If set to true, transformation is skipped and the input token is just forwarded as it is.

  command-line	-skip <boolean>
  default	false

  
  
• stopFlowOnError
  

  If set to true, the flow execution at this level gets stopped in case this actor encounters an error; the error gets propagated; useful for critical actors.

  command-line	-stop-flow-on-error <boolean>
  default	false
  min-user-mode	Expert

  
  
• silent
  

  If enabled, then no errors are output in the console; Note: the enclosing actor handler must have this enabled as well.

  command-line	-silent <boolean>
  default	false
  min-user-mode	Expert

  
  
• seed
  

  The seed for generating the random sub-samples.

  command-line	-seed <long>
  default	1

  
  
• numSubSamples
  

  The number of random sub-samples to generate.

  command-line	-num-subsamples <int>
  default	10
  minimum	1

  
  
• percentage
  

  The percentage of the sub-sample size (between 0 and 1).

  command-line	-percentage <double>
  default	0.66
  minimum	1.0E-4
  maximum	1.0

  
  
• statisticValues
  

  The evaluation values to extract and turn into a spreadsheet.

  command-line

  -statistic <Number correct|Number incorrect|Number unclassified|Percent correct|Percent incorrect|Percent unclassified|Kappa statistic|Mean absolute error|Root mean squared error|Relative absolute error|Root relative squared error|Correlation coefficient|SF prior entropy|SF scheme entropy|SF entropy gain|SF mean prior entropy|SF mean scheme entropy|SF mean entropy gain|KB information|KB mean information|KB relative information|True positive rate|Num true positives|False positive rate|Num false positives|True negative rate|Num true negatives|False negative rate|Num false negatives|IR precision|IR recall|F measure|Matthews correlation coefficient|Area under ROC|Area under PRC|Weighted true positive rate|Weighted false positive rate|Weighted true negative rate|Weighted false negative rate|Weighted IR precision|Weighted IR recall|Weighted F measure|Weighted Matthews correlation coefficient|Weighted area under ROC|Weighted area under PRC|Unweighted Macro F measure|Unweighted Micro F measure|Bias|Mean squared logarithmic error|R^2|Standard Deviation of Residuals|Ratio of Performance to Deviation> [-statistic ...]

  default

  
  
• classIndex
  

  The index of class label (eg used for AUC).

  command-line	-class-index <adams.data.weka.WekaLabelIndex>
  default	first
  example	An index is a number starting with 1; apart from label names (case-sensitive), the following placeholders can be used as well: first, second, third, last_2, last_1, last; numeric indices can be enforced by preceding them with '#' (eg '#12'); label names can be surrounded by double quotes.

  
  
• percentiles
  

  The percentiles to calculate for the errors (0-1; 0.95 is 95th percentile).

  command-line	-percentile <adams.core.base.BaseDouble> [-percentile ...]
  default

  
  
• errorCalculation
  

  Determines how to calculate the error.

  command-line	-error-calculation <ACTUAL_MINUS_PREDICTED|PREDICTED_MINUS_ACTUAL|BOTH|ABSOLUTE>
  default	ACTUAL_MINUS_PREDICTED

  
  
• withReplacement
  

  If enabled, predictions are drawn using with replacement (i.e., duplicates are possible).

  command-line	-with-replacement <boolean>
  default	true