All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Modules Pages
TanhComponent Class Reference

#include <nnet-component.h>

Inheritance diagram for TanhComponent:
Collaboration diagram for TanhComponent:

Public Member Functions

 TanhComponent (int32 dim)
 
 TanhComponent (const TanhComponent &other)
 
 TanhComponent ()
 
virtual std::string Type () const
 
virtual ComponentCopy () const
 Copy component (deep copy). More...
 
virtual bool BackpropNeedsInput () const
 
virtual bool BackpropNeedsOutput () const
 
virtual void Propagate (const ChunkInfo &in_info, const ChunkInfo &out_info, const CuMatrixBase< BaseFloat > &in, CuMatrixBase< BaseFloat > *out) const
 Perform forward pass propagation Input->Output. More...
 
virtual void Backprop (const ChunkInfo &in_info, const ChunkInfo &out_info, const CuMatrixBase< BaseFloat > &in_value, const CuMatrixBase< BaseFloat > &out_value, const CuMatrixBase< BaseFloat > &out_deriv, Component *to_update, CuMatrix< BaseFloat > *in_deriv) const
 Perform backward pass propagation of the derivative, and also either update the model (if to_update == this) or update another model or compute the model derivative (otherwise). More...
 
- Public Member Functions inherited from NonlinearComponent
void Init (int32 dim)
 
 NonlinearComponent (int32 dim)
 
 NonlinearComponent ()
 
 NonlinearComponent (const NonlinearComponent &other)
 
virtual int32 InputDim () const
 Get size of input vectors. More...
 
virtual int32 OutputDim () const
 Get size of output vectors. More...
 
virtual void InitFromString (std::string args)
 We implement InitFromString at this level. More...
 
virtual void Read (std::istream &is, bool binary)
 We implement Read at this level as it just needs the Type(). More...
 
virtual void Write (std::ostream &os, bool binary) const
 Write component to stream. More...
 
void Scale (BaseFloat scale)
 
void Add (BaseFloat alpha, const NonlinearComponent &other)
 
const CuVector< double > & ValueSum () const
 
const CuVector< double > & DerivSum () const
 
double Count () const
 
void SetDim (int32 dim)
 
- Public Member Functions inherited from Component
 Component ()
 
virtual int32 Index () const
 Returns the index in the sequence of layers in the neural net; intended only to be used in debugging information. More...
 
virtual void SetIndex (int32 index)
 
virtual std::vector< int32 > Context () const
 Return a vector describing the temporal context this component requires for each frame of output, as a sorted list. More...
 
void Propagate (const ChunkInfo &in_info, const ChunkInfo &out_info, const CuMatrixBase< BaseFloat > &in, CuMatrix< BaseFloat > *out) const
 A non-virtual propagate function that first resizes output if necessary. More...
 
virtual std::string Info () const
 
virtual ~Component ()
 

Private Member Functions

TanhComponentoperator= (const TanhComponent &other)
 

Additional Inherited Members

- Static Public Member Functions inherited from Component
static ComponentReadNew (std::istream &is, bool binary)
 Read component from stream. More...
 
static ComponentNewFromString (const std::string &initializer_line)
 Initialize the Component from one line that will contain first the type, e.g. More...
 
static ComponentNewComponentOfType (const std::string &type)
 Return a new Component of the given type e.g. More...
 
- Protected Member Functions inherited from NonlinearComponent
void UpdateStats (const CuMatrixBase< BaseFloat > &out_value, const CuMatrixBase< BaseFloat > *deriv=NULL)
 
const NonlinearComponentoperator= (const NonlinearComponent &other)
 
- Protected Attributes inherited from NonlinearComponent
int32 dim_
 
CuVector< double > value_sum_
 
CuVector< double > deriv_sum_
 
double count_
 
std::mutex mutex_
 

Detailed Description

Definition at line 610 of file nnet-component.h.

Constructor & Destructor Documentation

TanhComponent ( int32  dim)
inlineexplicit

Definition at line 612 of file nnet-component.h.

TanhComponent ( const TanhComponent other)
inlineexplicit

Definition at line 613 of file nnet-component.h.

TanhComponent ( )
inline

Definition at line 614 of file nnet-component.h.

Referenced by TanhComponent::Copy().

614 { }

Member Function Documentation

void Backprop ( const ChunkInfo in_info,
const ChunkInfo out_info,
const CuMatrixBase< BaseFloat > &  in_value,
const CuMatrixBase< BaseFloat > &  out_value,
const CuMatrixBase< BaseFloat > &  out_deriv,
Component to_update,
CuMatrix< BaseFloat > *  in_deriv 
) const
virtual

Perform backward pass propagation of the derivative, and also either update the model (if to_update == this) or update another model or compute the model derivative (otherwise).

Note: in_value and out_value are the values of the input and output of the component, and these may be dummy variables if respectively BackpropNeedsInput() or BackpropNeedsOutput() return false for that component (not all components need these).

num_chunks lets us treat the input matrix as contiguous-in-time chunks of equal size; it only matters if splicing is involved.

Implements Component.

Definition at line 664 of file nnet-component.cc.

References CuMatrixBase< Real >::Add(), CuMatrixBase< Real >::ApplyPow(), CuMatrixBase< Real >::CopyFromMat(), CuMatrixBase< Real >::MulElements(), CuMatrixBase< Real >::NumCols(), CuMatrixBase< Real >::NumRows(), CuMatrix< Real >::Resize(), CuMatrixBase< Real >::Scale(), and NonlinearComponent::UpdateStats().

670  {
671  /*
672  Note on the derivative of the tanh function:
673  tanh'(x) = sech^2(x) = -(tanh(x)+1) (tanh(x)-1) = 1 - tanh^2(x)
674 
675  The element by element equation of what we're doing would be:
676  in_deriv = out_deriv * (1.0 - out_value^2).
677  We can accomplish this via calls to the matrix library. */
678 
679  in_deriv->Resize(out_deriv.NumRows(), out_deriv.NumCols());
680  in_deriv->CopyFromMat(out_value);
681  in_deriv->ApplyPow(2.0);
682  in_deriv->Scale(-1.0);
683  in_deriv->Add(1.0);
684  // now in_deriv = (1.0 - out_value^2), the element-by-element derivative of
685  // the nonlinearity.
686  if (to_update != NULL)
687  dynamic_cast<NonlinearComponent*>(to_update)->UpdateStats(out_value,
688  in_deriv);
689  in_deriv->MulElements(out_deriv);
690 }
void Scale(Real value)
Definition: cu-matrix.cc:608
MatrixIndexT NumCols() const
Definition: cu-matrix.h:196
void CopyFromMat(const MatrixBase< OtherReal > &src, MatrixTransposeType trans=kNoTrans)
Definition: cu-matrix.cc:337
void UpdateStats(const CuMatrixBase< BaseFloat > &out_value, const CuMatrixBase< BaseFloat > *deriv=NULL)
void Resize(MatrixIndexT rows, MatrixIndexT cols, MatrixResizeType resize_type=kSetZero, MatrixStrideType stride_type=kDefaultStride)
Allocate the memory.
Definition: cu-matrix.cc:47
MatrixIndexT NumRows() const
Dimensions.
Definition: cu-matrix.h:195
void ApplyPow(Real power)
Apply power to the absolute value of each element.
Definition: cu-matrix.cc:2271
void MulElements(const CuMatrixBase< Real > &A)
Multiply two matrices elementwise: C = C .* A.
Definition: cu-matrix.cc:652
void Add(Real value)
Definition: cu-matrix.cc:546
virtual bool BackpropNeedsInput ( ) const
inlinevirtual

Reimplemented from Component.

Definition at line 617 of file nnet-component.h.

617 { return false; }
virtual bool BackpropNeedsOutput ( ) const
inlinevirtual

Reimplemented from Component.

Definition at line 618 of file nnet-component.h.

618 { return true; }
virtual Component* Copy ( ) const
inlinevirtual

Copy component (deep copy).

Implements Component.

Definition at line 616 of file nnet-component.h.

References TanhComponent::TanhComponent().

616 { return new TanhComponent(*this); }
TanhComponent& operator= ( const TanhComponent other)
private
void Propagate ( const ChunkInfo in_info,
const ChunkInfo out_info,
const CuMatrixBase< BaseFloat > &  in,
CuMatrixBase< BaseFloat > *  out 
) const
virtual

Perform forward pass propagation Input->Output.

Each row is one frame or training example. Interpreted as "num_chunks" equally sized chunks of frames; this only matters for layers that do things like context splicing. Typically this variable will either be 1 (when we're processing a single contiguous chunk of data) or will be the same as in.NumFrames(), but other values are possible if some layers do splicing.

Implements Component.

Definition at line 650 of file nnet-component.cc.

References ChunkInfo::CheckSize(), KALDI_ASSERT, ChunkInfo::NumChunks(), and CuMatrixBase< Real >::Tanh().

653  {
654  // Apply tanh function to each element of the output...
655  // the tanh function may be written as -1 + ( 2 / (1 + e^{-2 x})),
656  // which is a scaled and shifted sigmoid.
657 
658  in_info.CheckSize(in);
659  out_info.CheckSize(*out);
660  KALDI_ASSERT(in_info.NumChunks() == out_info.NumChunks());
661  out->Tanh(in);
662 }
void Tanh(const CuMatrixBase< Real > &src)
Compute the hyperbolic tangent (tanh) function; element by element, *this = tanh(src).
Definition: cu-matrix.cc:1632
#define KALDI_ASSERT(cond)
Definition: kaldi-error.h:169
virtual std::string Type ( ) const
inlinevirtual

Implements Component.

Definition at line 615 of file nnet-component.h.

615 { return "TanhComponent"; }

The documentation for this class was generated from the following files: