Packages

c

scalation.analytics

StoppingRule

class StoppingRule extends AnyRef

The StoppingRule class provides a stopping rule to terminating the iterative steps in an optimization early.

Linear Supertypes
AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. StoppingRule
  2. AnyRef
  3. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new StoppingRule(up_limit: Int = 4)

    up_limit

    the stopping rule limit (number of increasing steps allowed)

Value Members

  1. def stopWhen(b: NetParams, sse: Double): (NetParams, Double)

    Stop when too many steps have the cost measure (e.g., sse) increasing.

    Stop when too many steps have the cost measure (e.g., sse) increasing. Signal a stopping condition by returning the best parameter vector, else null.

    b

    the current parameter value (weights and biases)

    sse

    the current value of cost measure (e.g., sum of sqaured errors)

  2. def stopWhen(b: VectoD, sse: Double): (VectoD, Double)

    Stop when too many steps have the cost measure (e.g., sse) increasing.

    Stop when too many steps have the cost measure (e.g., sse) increasing. Signal a stopping condition by returning the best parameter vector, else null.

    b

    the current value of the parameter vector

    sse

    the current value of cost measure (e.g., sum of sqaured errors)