edu.stanford.nlp.tmt

stage

package stage

Visibility
  1. Public
  2. All
Impl.
  1. Concrete
  2. Abstract

Type Members

  1. case class GroupedTopTerms [Grouping] (topic: String, grouping: Grouping, terms: List[String]) extends Product

    Describes the top-k terms in a topic by grouping.

  2. case class GroupedTopicUsage [Grouping] (topic: String, grouping: Grouping, numDocuments: Double, numWords: Double) extends Product

    Describes the usage of a topic within a grouping.

  3. case class TopTerms (topic: String, terms: List[String]) extends Product

    Describes the top-k terms in a topic.

  4. case class TopicUsage (topic: String, numDocuments: Double, numWords: Double) extends Product

    Describes the usage of a topic within a grouping.

Value Members

  1. def EstimateLabeledLDAPerWordTopicDistributions (model: edu.stanford.nlp.tmt.model.llda.LabeledLDA[_, _, _], dataset: Iterable[LabeledLDADocumentParams], perDocTopicDistributions: Iterable[(String, SparseArray[Double])]) : LazyIterable[(String, Array[SparseArray[Double]])]

    Estimates the per-word topic distributions using the given model counts and the per-document topic distributions.

    Estimates the per-word topic distributions using the given model counts and the per-document topic distributions. This is not as exact as inference, but is nearly so, and is much faster.

    definition classes: root

  2. def EstimatePerWordTopicDistributions (model: edu.stanford.nlp.tmt.model.lda.LDA[_, _, _], dataset: Iterable[LDADocumentParams], perDocTopicDistributions: Iterable[(String, Array[Double])]) : LazyIterable[(String, Array[Array[Double]])]

    Estimates the per-word topic distributions using the given model counts and the per-document topic distributions.

    Estimates the per-word topic distributions using the given model counts and the per-document topic distributions. This is not as exact as inference, but is nearly so, and is much faster.

    definition classes: root

  3. object GroupedTopTerms extends AnyRef

  4. object GroupedTopicUsage extends AnyRef

  5. def InferCVB0DocumentTopicDistributions (model: CVB0LDA, dataset: Iterable[LDADocumentParams]) : LazyIterable[(String, Array[Double])]

    Returns an array of per-term, per-topic probabilities.

    Returns an array of per-term, per-topic probabilities. Loops while the largest difference between iterations in probabilities for any given topic is greater than delta.

    definition classes: root

  6. def InferCVB0LabeledLDADocumentTopicDistributions (model: CVB0LabeledLDA, dataset: Iterable[LabeledLDADocumentParams]) : LazyIterable[(String, SparseArray[Double])]

    Returns an array of per-term, per-topic probabilities.

    Returns an array of per-term, per-topic probabilities. Loops while the largest difference between iterations in probabilities for any given topic is greater than delta.

    definition classes: root

  7. def InferGibbsDocumentTopicDistributions (model: GibbsLDA, dataset: Iterable[LDADocumentParams]) : LazyIterable[(String, Array[Double])]

    Returns an array of per-term, per-topic probabilities.

    Returns an array of per-term, per-topic probabilities.

    definition classes: root

  8. def InferGibbsLabeledLDADocumentTopicDistributions (model: GibbsLabeledLDA, dataset: Iterable[LabeledLDADocumentParams]) : LazyIterable[(String, SparseArray[Double])]

    Returns an array of per-term, per-topic probabilities.

    Returns an array of per-term, per-topic probabilities.

    definition classes: root

  9. def LoadCVB0LDA (path: File) : CVB0LDA

    Loads a GibbsLDA model from the given path.

    Loads a GibbsLDA model from the given path.

    definition classes: root

  10. def LoadCVB0LabeledLDA (path: File) : CVB0LabeledLDA

    Loads a GibbsLDA model from the given path.

    Loads a GibbsLDA model from the given path.

    definition classes: root

  11. def LoadGibbsLDA (path: File) : GibbsLDA

    Loads a GibbsLDA model from the given path.

    Loads a GibbsLDA model from the given path.

    definition classes: root

  12. def LoadGibbsLabeledLDA (path: File) : GibbsLabeledLDA

    Loads a GibbsLDA model from the given path.

    Loads a GibbsLDA model from the given path.

    definition classes: root

  13. def LoadLDADocumentTopicDistributions (path: CSVFile) : LazyIterable[(String, Array[Double])]

    Loads the document-topic assignments from the given path.

    Loads the document-topic assignments from the given path.

    definition classes: root

  14. def QueryTopTerms [ID, Dist, Grouping] (model: edu.stanford.nlp.tmt.model.TopicModel[_, _, _, _, _] with ClosedTopicSet, dataset: Iterable[LDADocumentParams], perDocWordTopicDistribution: Iterable[(String, Array[Dist])], numTopTerms: Int, grouping: Iterable[Item[ID, Grouping]])(implicit active: CanGetActiveValues[Dist, Int, Double], ordering: Ordering[Grouping]) : Iterable[GroupedTopTerms[Grouping]]

    Gets the top terms in each topic, but counting terms instances separately for members of each group.

    Gets the top terms in each topic, but counting terms instances separately for members of each group.

    definition classes: root

  15. def QueryTopTerms [ID, Dist] (model: edu.stanford.nlp.tmt.model.TopicModel[_, _, _, _, _] with ClosedTopicSet, dataset: Iterable[LDADocumentParams], perDocWordTopicDistribution: Iterable[(String, Array[Dist])], numTopTerms: Int)(implicit active: CanGetActiveValues[Dist, Int, Double]) : Iterable[TopTerms]

    Gets the top terms in each topic, but counting terms instances separately for members of each group.

    Gets the top terms in each topic, but counting terms instances separately for members of each group.

    definition classes: root

  16. def QueryTopTerms (model: edu.stanford.nlp.tmt.model.TopicModel[_, _, _, _, _] with ClosedTopicSet, numTerms: Int) : LazyIterable[(String, String, Int, Double)]

    Returns the top terms associated with the model.

    Returns the top terms associated with the model.

    definition classes: root

  17. def QueryTopicUsage [ID, Grouping] (model: edu.stanford.nlp.tmt.model.TopicModel[_, _, _, _, _] with ClosedTopicSet, dataset: Iterable[LDADocumentParams], perDocTopicDistribution: Iterable[(String, Array[Double])], grouping: Iterable[Item[ID, Grouping]])(implicit active: CanGetActiveValues[Array[Double], Int, Double], ordering: Ordering[Grouping]) : Iterable[GroupedTopicUsage[Grouping]]

    Gets the usage of each topic by sub-group.

    Gets the usage of each topic by sub-group.

    definition classes: root

  18. def QueryTopicUsage [ID] (model: edu.stanford.nlp.tmt.model.TopicModel[_, _, _, _, _] with ClosedTopicSet, dataset: Iterable[LDADocumentParams], perDocTopicDistribution: Iterable[(String, Array[Double])])(implicit active: CanGetActiveValues[Array[Double], Int, Double]) : Iterable[TopicUsage]

    Gets the usage of each topic overall within a corpus.

    Gets the usage of each topic overall within a corpus.

    definition classes: root

  19. object TopTerms extends TableRowCompanion[TopTerms, (String, List[String])]

  20. object TopicUsage extends TableRowCompanion[TopicUsage, (String, Double, Double)]

  21. def TrainCVB0LDA (modelParams: LDAModelParams, dataset: Iterable[LDADocumentParams], output: File, maxIterations: Int =1000) : CVB0LDA

    Trains a CVB0LDA model using the given model parametesr.

    Trains a CVB0LDA model using the given model parametesr.

    definition classes: root

  22. def TrainCVB0LabeledLDA (modelParams: LabeledLDAModelParams, dataset: Iterable[LabeledLDADocumentParams], output: File, maxIterations: Int =1000) : CVB0LabeledLDA

    Trains a CVB0LDA model using the given model parametesr.

    Trains a CVB0LDA model using the given model parametesr.

    definition classes: root

  23. def TrainGibbsLDA (modelParams: LDAModelParams, dataset: Iterable[LDADocumentParams], output: File, maxIterations: Int =1500) : GibbsLDA

    Trains a GibbsLDA model using the given model parametesr.

    Trains a GibbsLDA model using the given model parametesr.

    definition classes: root

  24. def TrainGibbsLabeledLDA (modelParams: LabeledLDAModelParams, dataset: Iterable[LabeledLDADocumentParams], output: File, maxIterations: Int =1500) : GibbsLabeledLDA

    Trains a GibbsLDA model using the given model parametesr.

    Trains a GibbsLDA model using the given model parametesr.

    definition classes: root