Name

adams.flow.transformer.Evaluator

Synopsis

If input is Instances, build this Evaluator. If Instance, use the built Evaluator.
The name of this evaluator is used for storing the evaluation result.

Additional information

Flow input/output:
- input: weka.core.Instances, weka.core.Instance, adams.data.instance.WekaInstanceContainer, adams.flow.container.EvaluationContainer
- output: adams.flow.container.EvaluationContainer

Container information:
- adams.flow.container.EvaluationContainer:
   - Instance: data row; weka.core.Instance
   - Instances: dataset; weka.core.Instances
   - Evaluations: mapping of evaluations (String/Object); java.util.Map
   - Evaluator: evaluator used; adams.data.evaluator.instance.AbstractEvaluator
   - Classification: classification; java.lang.Double or java.lang.String
   - Abstention classification: abstention classification; java.lang.Double or java.lang.String
   - Distribution: class distribution; double[]
   - Component: component name; java.lang.String
   - Version: version; java.lang.String
   - ID: (optional) ID; java.lang.String
   - Report: report for storing meta-data; adams.data.report.Report

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	Evaluator

  
  
• 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

  
  
• noCopy
  

  If enabled, no copy of the evaluation container is generated before adding the additional evaluations.

  command-line	-no-copy <boolean>
  default	false

  
  
• evaluator
  

  The Evaluator to train on the input data.

  command-line	-evaluator <adams.data.evaluator.instance.AbstractEvaluator>
  default	adams.data.evaluator.instance.NullEvaluator -missing-evaluation NaN

  
  
• modelLoadingType
  

  Determines how to load the model, in case of AUTO, first the model file is checked, then the callable actor and then the storage.

  command-line	-model-loading-type <AUTO|FILE|SOURCE_ACTOR|STORAGE>
  default	AUTO

  
  
• modelFile
  

  The file to load the model from, ignored if pointing to a directory.

  command-line	-model <adams.core.io.PlaceholderFile>
  default	${CWD}

  
  
• modelActor
  

  The callable actor (source) to obtain the model from, ignored if not present.

  command-line	-model-actor <adams.flow.core.CallableActorReference>
  default

  
  
• modelStorage
  

  The storage item to obtain the model from, ignored if not present.

  command-line	-model-storage <adams.flow.control.StorageName>
  default	storage

  
  
• useEvaluatorResetVariable
  

  If enabled, chnages to the specified variable are monitored in order to reset the evaluator, eg when a storage evaluator changed.

  command-line	-use-evaluator-reset-variable <boolean>
  default	false

  
  
• evaluatorResetVariable
  

  The variable to monitor for changes in order to reset the evaluator, eg when a storage evaluator changed.

  command-line	-evaluator-reset-variable <adams.core.VariableName>
  default	variable

  
  
• component
  

  The component identifier.

  command-line	-component <java.lang.String>
  default

  
  
• version
  

  The version.

  command-line	-version <java.lang.String>
  default

  
  
• cleanAfterBuild
  

  If enabled, the internnaly built evaluator gets discarded again.

  command-line	-clean-after-build <boolean>
  default	false

  
  
• preferJobRunner
  

  If enabled, tries to offload the processing onto a adams.flow.standalone.JobRunnerInstance; applies only to training.

  command-line	-prefer-jobrunner <boolean>
  default	false