PLearn 0.1
Public Types | Public Member Functions | Static Public Member Functions | Static Public Attributes | Static Protected Member Functions | Protected Attributes | Private Member Functions
PLearn::LimitedGaussianSmoother Class Reference

#include <LimitedGaussianSmoother.h>

Inheritance diagram for PLearn::LimitedGaussianSmoother:
Inheritance graph
[legend]
Collaboration diagram for PLearn::LimitedGaussianSmoother:
Collaboration graph
[legend]

List of all members.

Public Types

typedef Smoother inherited

Public Member Functions

 LimitedGaussianSmoother ()
 LimitedGaussianSmoother (real window_size_wrt_sigma_, real sigma_bin_)
virtual void build ()
 Post-constructor.
virtual void makeDeepCopyFromShallowCopy (CopiesMap &copies)
 Transforms a shallow copy into a deep copy.
virtual string classname () const
virtual OptionListgetOptionList () const
virtual OptionMapgetOptionMap () const
virtual RemoteMethodMapgetRemoteMethodMap () const
virtual LimitedGaussianSmootherdeepCopy (CopiesMap &copies) const
virtual real smooth (const Vec &source_function, Vec &smoothed_function, Vec bin_positions=Vec(), Vec dest_bin_positions=Vec()) const

Static Public Member Functions

static string _classname_ ()
 Declares name and deepCopy methods.
static OptionList_getOptionList_ ()
static RemoteMethodMap_getRemoteMethodMap_ ()
static Object_new_instance_for_typemap_ ()
static bool _isa_ (const Object *o)
static void _static_initialize_ ()
static const PPathdeclaringFile ()

Static Public Attributes

static StaticInitializer _static_initializer_

Static Protected Member Functions

static void declareOptions (OptionList &ol)
 Declares this class' options.

Protected Attributes

real window_size_wrt_sigma
real sigma_bin

Private Member Functions

void build_ ()
 This does the actual building.

Detailed Description

Definition at line 58 of file LimitedGaussianSmoother.h.


Member Typedef Documentation

Reimplemented from PLearn::Smoother.

Definition at line 70 of file LimitedGaussianSmoother.h.


Constructor & Destructor Documentation

PLearn::LimitedGaussianSmoother::LimitedGaussianSmoother ( )

Definition at line 46 of file LimitedGaussianSmoother.cc.

    :Smoother()
/* ### Initialise all fields to their default value */
{
    // ...

    // ### You may or may not want to call build_() to finish building the object
    // build_();
}
PLearn::LimitedGaussianSmoother::LimitedGaussianSmoother ( real  window_size_wrt_sigma_,
real  sigma_bin_ 
)

Definition at line 56 of file LimitedGaussianSmoother.cc.

    :Smoother(), window_size_wrt_sigma(window_size_wrt_sigma_), sigma_bin(sigma_bin_)
{}

Member Function Documentation

string PLearn::LimitedGaussianSmoother::_classname_ ( ) [static]

Declares name and deepCopy methods.

Reimplemented from PLearn::Smoother.

Definition at line 60 of file LimitedGaussianSmoother.cc.

OptionList & PLearn::LimitedGaussianSmoother::_getOptionList_ ( ) [static]

Reimplemented from PLearn::Smoother.

Definition at line 60 of file LimitedGaussianSmoother.cc.

RemoteMethodMap & PLearn::LimitedGaussianSmoother::_getRemoteMethodMap_ ( ) [static]

Reimplemented from PLearn::Smoother.

Definition at line 60 of file LimitedGaussianSmoother.cc.

bool PLearn::LimitedGaussianSmoother::_isa_ ( const Object o) [static]

Reimplemented from PLearn::Smoother.

Definition at line 60 of file LimitedGaussianSmoother.cc.

Object * PLearn::LimitedGaussianSmoother::_new_instance_for_typemap_ ( ) [static]

Reimplemented from PLearn::Object.

Definition at line 60 of file LimitedGaussianSmoother.cc.

StaticInitializer LimitedGaussianSmoother::_static_initializer_ & PLearn::LimitedGaussianSmoother::_static_initialize_ ( ) [static]

Reimplemented from PLearn::Smoother.

Definition at line 60 of file LimitedGaussianSmoother.cc.

void PLearn::LimitedGaussianSmoother::build ( ) [virtual]

Post-constructor.

The normal implementation should call simply inherited::build(), then this class's build_(). This method should be callable again at later times, after modifying some option fields to change the "architecture" of the object.

Reimplemented from PLearn::Smoother.

Definition at line 91 of file LimitedGaussianSmoother.cc.

References PLearn::Smoother::build(), and build_().

Here is the call graph for this function:

void PLearn::LimitedGaussianSmoother::build_ ( ) [private]

This does the actual building.

Reimplemented from PLearn::Smoother.

Definition at line 79 of file LimitedGaussianSmoother.cc.

Referenced by build().

{
    // ### This method should do the real building of the object,
    // ### according to set 'options', in *any* situation. 
    // ### Typical situations include:
    // ###  - Initial building of an object from a few user-specified options
    // ###  - Building of a "reloaded" object: i.e. from the complete set of all serialised options.
    // ###  - Updating or "re-building" of an object after a few "tuning" options have been modified.
    // ### You should assume that the parent class' build_() has already been called.
}

Here is the caller graph for this function:

string PLearn::LimitedGaussianSmoother::classname ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 60 of file LimitedGaussianSmoother.cc.

void PLearn::LimitedGaussianSmoother::declareOptions ( OptionList ol) [static, protected]

Declares this class' options.

Reimplemented from PLearn::Smoother.

Definition at line 62 of file LimitedGaussianSmoother.cc.

References PLearn::Smoother::declareOptions().

{
    // ### Declare all of this object's options here
    // ### For the "flags" of each option, you should typically specify  
    // ### one of OptionBase::buildoption, OptionBase::learntoption or 
    // ### OptionBase::tuningoption. Another possible flag to be combined with
    // ### is OptionBase::nosave

    // ### ex:
    // declareOption(ol, "myoption", &LimitedGaussianSmoother::myoption, OptionBase::buildoption,
    //               "Help text describing this option");
    // ...

    // Now call the parent class' declareOptions
    inherited::declareOptions(ol);
}

Here is the call graph for this function:

static const PPath& PLearn::LimitedGaussianSmoother::declaringFile ( ) [inline, static]

Reimplemented from PLearn::Smoother.

Definition at line 113 of file LimitedGaussianSmoother.h.

:
    virtual real smooth(const Vec& source_function, Vec& smoothed_function, 
LimitedGaussianSmoother * PLearn::LimitedGaussianSmoother::deepCopy ( CopiesMap copies) const [virtual]

Reimplemented from PLearn::Smoother.

Definition at line 60 of file LimitedGaussianSmoother.cc.

OptionList & PLearn::LimitedGaussianSmoother::getOptionList ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 60 of file LimitedGaussianSmoother.cc.

OptionMap & PLearn::LimitedGaussianSmoother::getOptionMap ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 60 of file LimitedGaussianSmoother.cc.

RemoteMethodMap & PLearn::LimitedGaussianSmoother::getRemoteMethodMap ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 60 of file LimitedGaussianSmoother.cc.

void PLearn::LimitedGaussianSmoother::makeDeepCopyFromShallowCopy ( CopiesMap copies) [virtual]

Transforms a shallow copy into a deep copy.

Reimplemented from PLearn::Smoother.

Definition at line 98 of file LimitedGaussianSmoother.cc.

References PLearn::Smoother::makeDeepCopyFromShallowCopy(), and PLERROR.

{
    inherited::makeDeepCopyFromShallowCopy(copies);

    // ### Call deepCopyField on all "pointer-like" fields 
    // ### that you wish to be deepCopied rather than 
    // ### shallow-copied.
    // ### ex:
    // deepCopyField(trainvec, copies);

    // ### Remove this line when you have fully implemented this method.
    PLERROR("LimitedGaussianSmoother::makeDeepCopyFromShallowCopy not fully (correctly) implemented yet!");
}

Here is the call graph for this function:

real PLearn::LimitedGaussianSmoother::smooth ( const Vec source_function,
Vec smoothed_function,
Vec  bin_positions = Vec(),
Vec  dest_bin_positions = Vec() 
) const [virtual]

Implements PLearn::Smoother.

Definition at line 113 of file LimitedGaussianSmoother.cc.

References PLearn::TVec< T >::fill(), PLearn::gauss_cum(), i, j, PLearn::TVec< T >::length(), PLearn::TVec< T >::resize(), sigma_bin, and window_size_wrt_sigma.

{
    //parzen regressor?? kernel smoothing??
// smoothed_function[k] = sum_{j=max(0,k-window_size)}^{min(l-1,k+window_size)} w_{k,j} source_function[j]
//                        / sum_{j=max(0,k-window_size)}^{min(l-1,k+window_size)} w_{k,j} 
// with w_{k,j} = phi(bin_positions[j+1];mu_k,sigma_k)-phi(bin_positions[j];mu_k,sigma_k)
// where mu_k = 0.5*(bin_positions[k+1]+bin_positions[k]),
//       sigma_k = bin_positions[k+window_size]-bin_positions[k]
// where phi(x;mu,sigma) = cdf of normal(mu,sigma) at x,
// window_size = window_size_wrt_sigma * sigma_bin

// for dest_bin_positions != bin_positions: 2 methods:
// 1- trouver sigma_bin en fonction du voisinage
//    d'une position dest.  Somme ponderee avec 
//    gaussienne centree sur le pt. dest.
// 2- Un sigma_bin pour chaque bin_position (src).
//    Une gaussienne centree sur sur ch. pos src.
//    



    smoothed_function.resize(source_function.length());
    smoothed_function.fill(0.0);
    real window_size= window_size_wrt_sigma * sigma_bin;
    for(int i= 0; i < smoothed_function.length()-1; ++i)
    {
        int min_j= i-static_cast<int>(window_size), max_j= i+static_cast<int>(window_size);
        if(min_j < 0) min_j= 0;
        if(max_j > smoothed_function.length()) max_j= smoothed_function.length();
        real sum_weights= 0.0;
        real mu= 0.5*(bin_positions[i+1]+bin_positions[i]),
            sigma= bin_positions[max_j-1]-bin_positions[i];
        for(int j= min_j; j < max_j-1; ++j)
        {
            sum_weights+= gauss_cum(bin_positions[j+1], mu, sigma) -
                gauss_cum(bin_positions[j], mu, sigma);
        }
        for(int j= min_j; j < max_j-1; ++j)
            smoothed_function[i]+= ( gauss_cum(bin_positions[j+1], mu, sigma) 
                                     - gauss_cum(bin_positions[j], mu, sigma)
                ) 
                * source_function[j] / sum_weights;
    }

    return 0.0; //dummy - FIXME - xsm



/*
  if(bin_positions.length() != 0 && source_function.length() != bin_positions.length()-1)
  PLERROR("in LimitedGaussianSmoother::smooth  There must be one more bin_positions than the "
  "number of source_function points.");
  //if no bin_positions given, assume positions are 0, 1, 2, ..., n
  if(bin_positions.length() == 0)
  {
  int n= source_function.length()+1;
  bin_positions.resize(n);
  for(int i= 0; i < n; ++i)
  bin_positions[i]= i;
  }
  //if no dest_bin_positions given, assume same as bin_positions
  if(dest_bin_positions.length() == 0)
  dest_bin_positions= bin_positions;

  smoothed_function.resize(dest_bin_positions.length()-1);
  smoothed_function.fill(0.0);
  real window_size, mu, sigma, sum_weights;
  int n= smoothed_function.length();
  for(int i= 0; i < n; ++i)
  {
  sum_weights= 0.0;
  int nj= source_function.length();
  for(int j= 0; j < nj; ++j)
  {
  mu= 0.5*(bin_positions[i+1]+bin_positions[i]);
  sigma= bin_positions[i+1]-bin_positions[i];
  window_size= window_size_wrt_sigma * sigma;
  real p1= mu - 0.5*window_size,
  p2= mu + 0.5*window_size;

  Vec::iterator it = find_if(options.begin(), options.end(),
  bind2nd(mem_fun(&OptionBase::isOptionNamed), optionname));

          

  }
      


  / *
  int min_j= i-static_cast<int>(window_size), max_j= i+static_cast<int>(window_size);
  if(min_j < 0) min_j= 0;
  if(max_j > smoothed_function.length()) max_j= smoothed_function.length();
  * /
  real sum_weights= 0.0;
  for(int j= min_j; j < max_j-1; ++j)
  {
  sum_weights+= gauss_cum(bin_positions[j+1], mu, sigma) -
  gauss_cum(bin_positions[j], mu, sigma);
  }
  for(int j= min_j; j < max_j-1; ++j)
  smoothed_function[i]+= ( gauss_cum(bin_positions[j+1], mu, sigma) 
  - gauss_cum(bin_positions[j], mu, sigma)
  ) 
  * source_function[j] / sum_weights;
  }
*/
}

Here is the call graph for this function:


Member Data Documentation

Reimplemented from PLearn::Smoother.

Definition at line 113 of file LimitedGaussianSmoother.h.

Definition at line 65 of file LimitedGaussianSmoother.h.

Referenced by smooth().

Definition at line 65 of file LimitedGaussianSmoother.h.

Referenced by smooth().


The documentation for this class was generated from the following files:
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines