lamp-core/lamp.nn/FreeRunningRNN

FreeRunningRNN

case class FreeRunningRNN[T, M <: StatefulModule[Variable, Variable, T]](module: M & StatefulModule[Variable, Variable, T], timeSteps: Int) extends StatefulModule[Variable, Variable, T]

Wraps a (sequence x batch) long -> (sequence x batch x dim) double stateful module and runs in it greedy (argmax) generation mode over timeSteps steps.

Companion:
object
trait Serializable
trait Product
trait Equals
trait GenericModule[(Variable, T), (Variable, T)]
class Object
trait Matchable
class Any

Value members

Concrete methods

def forward[S : Sc](x: (Variable, T)): (Variable, T)
def loop[S : Sc](lastOutput: Variable, lastState: T, n: Int, buffer: Seq[Variable]): (Seq[Variable], T)
def state: Seq[(Constant, PTag)]

Inherited methods

def apply[S : Sc](a: (Variable, T)): (Variable, T)

Alias of forward

Alias of forward

Inherited from:
GenericModule
final def gradients(loss: Variable, zeroGrad: Boolean): Seq[Option[STen]]

Computes the gradient of loss with respect to the parameters.

Computes the gradient of loss with respect to the parameters.

Inherited from:
GenericModule
final def learnableParameters: Long

Returns the total number of optimizable parameters.

Returns the total number of optimizable parameters.

Inherited from:
GenericModule
final def parameters: Seq[(Constant, PTag)]

Returns the state variables which need gradient computation.

Returns the state variables which need gradient computation.

Inherited from:
GenericModule
def productElementNames: Iterator[String]
Inherited from:
Product
def productIterator: Iterator[Any]
Inherited from:
Product
final def zeroGrad(): Unit
Inherited from:
GenericModule