keystoneml.nodes.nlp

StupidBackoffEstimator

case class StupidBackoffEstimator[T](unigramCounts: Map[T, Int], alpha: Double = 0.4)(implicit evidence$3: ClassTag[T]) extends Estimator[(NGram[T], Int), (NGram[T], Double)] with Product with Serializable

Estimates a Stupid Backoff ngram language model, which was introduced in the following paper:

Brants, Thorsten, et al. "Large language models in machine translation." 2007.

The results are scores indicating likeliness of each ngram, but they are not normalized probabilities. The score for an n-gram is defined recursively:

S(w_i | w_{i - n + 1}{i - 1}) := if numerator > 0: freq(w_{i - n + 1}i) / freq(w_{i - n + 1}{i - 1}) otherwise: \alpha * S(w_i | w_{i - n + 2}{i - 1})

S(w_i) := freq(w_i) / N, where N is the total number of tokens in training corpus.

unigramCounts

the pre-computed unigram counts of the training corpus

alpha

hyperparameter that gets multiplied once per backoff

Linear Supertypes
Product, Equals, Estimator[(NGram[T], Int), (NGram[T], Double)], EstimatorOperator, Serializable, Serializable, Operator, AnyRef, Any
Ordering
  1. Alphabetic
  2. By inheritance
Inherited
  1. StupidBackoffEstimator
  2. Product
  3. Equals
  4. Estimator
  5. EstimatorOperator
  6. Serializable
  7. Serializable
  8. Operator
  9. AnyRef
  10. Any
  1. Hide All
  2. Show all
Learn more about member selection
Visibility
  1. Public
  2. All

Instance Constructors

  1. new StupidBackoffEstimator(unigramCounts: Map[T, Int], alpha: Double = 0.4)(implicit arg0: ClassTag[T])

    unigramCounts

    the pre-computed unigram counts of the training corpus

    alpha

    hyperparameter that gets multiplied once per backoff

Value Members

  1. final def !=(arg0: AnyRef): Boolean

    Definition Classes
    AnyRef
  2. final def !=(arg0: Any): Boolean

    Definition Classes
    Any
  3. final def ##(): Int

    Definition Classes
    AnyRef → Any
  4. final def ==(arg0: AnyRef): Boolean

    Definition Classes
    AnyRef
  5. final def ==(arg0: Any): Boolean

    Definition Classes
    Any
  6. val alpha: Double

    hyperparameter that gets multiplied once per backoff

  7. final def asInstanceOf[T0]: T0

    Definition Classes
    Any
  8. def clone(): AnyRef

    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  9. final def eq(arg0: AnyRef): Boolean

    Definition Classes
    AnyRef
  10. def execute(deps: Seq[Expression]): TransformerExpression

    Definition Classes
    EstimatorOperator → Operator
  11. def finalize(): Unit

    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  12. def fit(data: RDD[(NGram[T], Int)]): StupidBackoffModel[T]

    The type-safe method that ML developers need to implement when writing new Estimators.

    The type-safe method that ML developers need to implement when writing new Estimators.

    data

    The estimator's training data.

    returns

    A new transformer

    Definition Classes
    StupidBackoffEstimatorEstimator
  13. final def getClass(): Class[_]

    Definition Classes
    AnyRef → Any
  14. final def isInstanceOf[T0]: Boolean

    Definition Classes
    Any
  15. def label: String

    Definition Classes
    Operator
  16. final def ne(arg0: AnyRef): Boolean

    Definition Classes
    AnyRef
  17. final def notify(): Unit

    Definition Classes
    AnyRef
  18. final def notifyAll(): Unit

    Definition Classes
    AnyRef
  19. final def synchronized[T0](arg0: ⇒ T0): T0

    Definition Classes
    AnyRef
  20. val unigramCounts: Map[T, Int]

    the pre-computed unigram counts of the training corpus

  21. final def wait(): Unit

    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  22. final def wait(arg0: Long, arg1: Int): Unit

    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  23. final def wait(arg0: Long): Unit

    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  24. final def withData(data: PipelineDataset[(NGram[T], Int)]): Pipeline[(NGram[T], Int), (NGram[T], Double)]

    Constructs a pipeline that fits this estimator to training data, then applies the resultant transformer to the Pipeline input.

    Constructs a pipeline that fits this estimator to training data, then applies the resultant transformer to the Pipeline input.

    data

    The training data

    returns

    A pipeline that fits this estimator and applies the result to inputs.

    Definition Classes
    Estimator
  25. final def withData(data: RDD[(NGram[T], Int)]): Pipeline[(NGram[T], Int), (NGram[T], Double)]

    Constructs a pipeline that fits this estimator to training data, then applies the resultant transformer to the Pipeline input.

    Constructs a pipeline that fits this estimator to training data, then applies the resultant transformer to the Pipeline input.

    data

    The training data

    returns

    A pipeline that fits this estimator and applies the result to inputs.

    Definition Classes
    Estimator

Inherited from Product

Inherited from Equals

Inherited from Estimator[(NGram[T], Int), (NGram[T], Double)]

Inherited from EstimatorOperator

Inherited from Serializable

Inherited from Serializable

Inherited from Operator

Inherited from AnyRef

Inherited from Any

Ungrouped