vpdfl_kernel_pdf Class Reference

#include <vpdfl_kernel_pdf.h>

Inheritance diagram for vpdfl_kernel_pdf:

Detailed Description

Multi-variate kernel PDF.

Distribution is the sum of a set of kernel functions placed on the training samples.

Essentially a special case of a mixture model. It differs from the former in that a) every component has the same shape, though possibly a different width b) every component has the same weight c) Usually one uses different training algorithms.

Definition at line 24 of file vpdfl_kernel_pdf.h.

Public Member Functions

vpdfl_kernel_pdf ()

Dflt ctor.

virtual ~vpdfl_kernel_pdf ()

Destructor.

void set_centres (const vnl_vector< double > *x, int n, double width)

Initialise so all kernels have the same width.

void set_centres (const vnl_vector< double > *x, int n, const vnl_vector< double > &width)

Initialise so all kernels have given width.

const vcl_vector< vnl_vector<
double > > & centre () const

Position of kernel centres.

const vnl_vector< double > & width () const

Width of each kernel.

bool all_same_width () const

True if all kernels have the same width.

short version_no () const

Version number for I/O.

virtual vcl_string is_a () const

Name of the class.

virtual bool is_class (vcl_string const &s) const

Does the name of the class match the argument?.

virtual void print_summary (vcl_ostream &os) const

Print class to os.

virtual void b_write (vsl_b_ostream &bfs) const

Save class to binary file stream.

virtual void b_read (vsl_b_istream &bfs)

Load class from binary file stream.

const vnl_vector< double > & mean () const

Mean of distribution.

const vnl_vector< double > & variance () const

Variance of each dimension.

int n_dims () const

Number of dimensions.

virtual int n_peaks () const

Number of peaks of distribution.

virtual const vnl_vector<
double > & peak (int) const

Position of the i'th peak.

virtual double log_p (const vnl_vector< double > &x) const =0

Log of probability density at x.

virtual double operator() (const vnl_vector< double > &x) const

Probability density at x.

virtual void gradient (vnl_vector< double > &g, const vnl_vector< double > &x, double &p) const =0

Gradient and value of PDF at x.

virtual void gradient_logp (vnl_vector< double > &g, const vnl_vector< double > &x) const

Gradient and value of log(p(x)) at x.

virtual vpdfl_sampler_base * new_sampler () const =0

Create a sampler object on the heap.

virtual double log_prob_thresh (double pass_proportion) const

Compute threshold for PDF to pass a given proportion.

virtual void nearest_plausible (vnl_vector< double > &x, double log_p_min) const =0

Compute nearest point to x which has a density above a threshold.

virtual bool is_valid_pdf () const

Return true if the object represents a valid PDF.

virtual vpdfl_pdf_base * clone () const =0

Create a copy on the heap and return base class pointer.

Protected Member Functions

void calc_mean_var ()

Compute mean/variance given current centres and widths.

void set_mean (const vnl_vector< double > &m)

void set_variance (const vnl_vector< double > &v)

Protected Attributes

vcl_vector< vnl_vector< double > > x_

Position of kernel centres.

vnl_vector< double > width_

Width of each kernel.

bool all_same_width_

True if all kernels have the same width.

Constructor & Destructor Documentation

vpdfl_kernel_pdf::vpdfl_kernel_pdf ( )

Dflt ctor.

Definition at line 20 of file vpdfl_kernel_pdf.cxx.

vpdfl_kernel_pdf::~vpdfl_kernel_pdf ( ) [virtual]

Destructor.

Definition at line 28 of file vpdfl_kernel_pdf.cxx.

Member Function Documentation

bool vpdfl_kernel_pdf::all_same_width ( ) const [inline]

True if all kernels have the same width.

Definition at line 63 of file vpdfl_kernel_pdf.h.

void vpdfl_kernel_pdf::b_read ( vsl_b_istream & bfs ) [virtual]

Load class from binary file stream.

Implements vpdfl_pdf_base.
Reimplemented in vpdfl_gaussian_kernel_pdf.
Definition at line 144 of file vpdfl_kernel_pdf.cxx.

void vpdfl_kernel_pdf::b_write ( vsl_b_ostream & bfs ) const [virtual]

Save class to binary file stream.

Implements vpdfl_pdf_base.
Reimplemented in vpdfl_gaussian_kernel_pdf.
Definition at line 132 of file vpdfl_kernel_pdf.cxx.

void vpdfl_kernel_pdf::calc_mean_var ( ) [protected]

Compute mean/variance given current centres and widths.

Definition at line 33 of file vpdfl_kernel_pdf.cxx.

const vcl_vector<vnl_vector<double> >& vpdfl_kernel_pdf::centre ( ) const [inline]

Position of kernel centres.

Definition at line 57 of file vpdfl_kernel_pdf.h.

virtual vpdfl_pdf_base* vpdfl_pdf_base::clone ( ) const [pure virtual, inherited]

Create a copy on the heap and return base class pointer.

Implemented in vpdfl_axis_gaussian, vpdfl_gaussian, vpdfl_gaussian_kernel_pdf, vpdfl_mixture, and vpdfl_pc_gaussian.

virtual void vpdfl_pdf_base::gradient ( vnl_vector< double > & g,

const vnl_vector< double > & x,

double & p

) const [pure virtual, inherited]

Gradient and value of PDF at x.
Computes gradient of PDF at x, and returns the prob at x in p
Implemented in vpdfl_axis_gaussian, vpdfl_gaussian, vpdfl_gaussian_kernel_pdf, and vpdfl_mixture.

void vpdfl_pdf_base::gradient_logp ( vnl_vector< double > & g,

const vnl_vector< double > & x

) const [virtual, inherited]

Gradient and value of log(p(x)) at x.
Computes gradient df/dx of f(x)=log(p(x)) at x. Result is vector of same dimensionality as x. Default baseclass implementation uses gradient() to compute grad/p
Reimplemented in vpdfl_axis_gaussian.
Definition at line 127 of file vpdfl_pdf_base.cxx.

vcl_string vpdfl_kernel_pdf::is_a ( ) const [virtual]

Name of the class.

Reimplemented from vpdfl_pdf_base.
Reimplemented in vpdfl_gaussian_kernel_pdf.
Definition at line 94 of file vpdfl_kernel_pdf.cxx.

bool vpdfl_kernel_pdf::is_class ( vcl_string const & s ) const [virtual]

Does the name of the class match the argument?.

Reimplemented from vpdfl_pdf_base.
Reimplemented in vpdfl_gaussian_kernel_pdf.
Definition at line 104 of file vpdfl_kernel_pdf.cxx.

bool vpdfl_pdf_base::is_valid_pdf ( ) const [virtual, inherited]

Return true if the object represents a valid PDF.
This will return false, if n_dims() is 0, for example just ofter default construction.
Reimplemented in vpdfl_mixture.
Definition at line 141 of file vpdfl_pdf_base.cxx.

virtual double vpdfl_pdf_base::log_p ( const vnl_vector< double > & x ) const [pure virtual, inherited]

Log of probability density at x.

Implemented in vpdfl_axis_gaussian, vpdfl_gaussian, vpdfl_gaussian_kernel_pdf, vpdfl_mixture, and vpdfl_pc_gaussian.

double vpdfl_pdf_base::log_prob_thresh ( double pass_proportion ) const [virtual, inherited]

Compute threshold for PDF to pass a given proportion.

Reimplemented in vpdfl_axis_gaussian, and vpdfl_gaussian.
Definition at line 40 of file vpdfl_pdf_base.cxx.

const vnl_vector<double>& vpdfl_pdf_base::mean ( ) const [inline, inherited]

Mean of distribution.

Definition at line 45 of file vpdfl_pdf_base.h.

int vpdfl_pdf_base::n_dims ( ) const [inline, inherited]

Number of dimensions.

Definition at line 51 of file vpdfl_pdf_base.h.

virtual int vpdfl_pdf_base::n_peaks ( ) const [inline, virtual, inherited]

Number of peaks of distribution.

Definition at line 54 of file vpdfl_pdf_base.h.

virtual void vpdfl_pdf_base::nearest_plausible ( vnl_vector< double > & x,

double log_p_min

) const [pure virtual, inherited]

Compute nearest point to x which has a density above a threshold.
If log_p(x)>log_p_min then x unchanged. Otherwise x is moved (typically up the gradient) until log_p(x)>=log_p_min.
Parameters:

x This may be modified to the nearest plausible position.

log_p_min lower threshold for log_p(x)

Implemented in vpdfl_axis_gaussian, vpdfl_gaussian, vpdfl_gaussian_kernel_pdf, and vpdfl_mixture.

virtual vpdfl_sampler_base* vpdfl_pdf_base::new_sampler ( ) const [pure virtual, inherited]

Create a sampler object on the heap.
Caller is responsible for deletion.
Implemented in vpdfl_axis_gaussian, vpdfl_gaussian, vpdfl_gaussian_kernel_pdf, and vpdfl_mixture.

double vpdfl_pdf_base::operator() ( const vnl_vector< double > & x ) const [virtual, inherited]

Probability density at x.

Reimplemented in vpdfl_gaussian_kernel_pdf, and vpdfl_mixture.
Definition at line 34 of file vpdfl_pdf_base.cxx.

virtual const vnl_vector<double>& vpdfl_pdf_base::peak ( int ) const [inline, virtual, inherited]

Position of the i'th peak.

Definition at line 57 of file vpdfl_pdf_base.h.

void vpdfl_kernel_pdf::print_summary ( vcl_ostream & os ) const [virtual]

Print class to os.

Implements vpdfl_pdf_base.
Reimplemented in vpdfl_gaussian_kernel_pdf.
Definition at line 122 of file vpdfl_kernel_pdf.cxx.

void vpdfl_kernel_pdf::set_centres ( const vnl_vector< double > * x,

int n,

const vnl_vector< double > & width

)

Initialise so all kernels have given width.
width[i] is essentially the sd on kernel i
Definition at line 78 of file vpdfl_kernel_pdf.cxx.

void vpdfl_kernel_pdf::set_centres ( const vnl_vector< double > * x,

int n,

double width

)

Initialise so all kernels have the same width.
width is essentially the sd on the kernels
Definition at line 65 of file vpdfl_kernel_pdf.cxx.

void vpdfl_pdf_base::set_mean ( const vnl_vector< double > & m ) [inline, protected, inherited]

Reimplemented in vpdfl_gaussian.
Definition at line 34 of file vpdfl_pdf_base.h.

void vpdfl_pdf_base::set_variance ( const vnl_vector< double > & v ) [inline, protected, inherited]

Definition at line 35 of file vpdfl_pdf_base.h.

const vnl_vector<double>& vpdfl_pdf_base::variance ( ) const [inline, inherited]

Variance of each dimension.

Definition at line 48 of file vpdfl_pdf_base.h.

short vpdfl_kernel_pdf::version_no ( ) const

Version number for I/O.

Reimplemented from vpdfl_pdf_base.
Reimplemented in vpdfl_gaussian_kernel_pdf.
Definition at line 113 of file vpdfl_kernel_pdf.cxx.

const vnl_vector<double>& vpdfl_kernel_pdf::width ( ) const [inline]

Width of each kernel.

Definition at line 60 of file vpdfl_kernel_pdf.h.

Member Data Documentation

bool vpdfl_kernel_pdf::all_same_width_ [protected]

True if all kernels have the same width.

Definition at line 34 of file vpdfl_kernel_pdf.h.

vnl_vector<double> vpdfl_kernel_pdf::width_ [protected]

Width of each kernel.

Definition at line 31 of file vpdfl_kernel_pdf.h.

vcl_vector<vnl_vector<double> > vpdfl_kernel_pdf::x_ [protected]

Position of kernel centres.

Definition at line 28 of file vpdfl_kernel_pdf.h.

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

contrib/mul/vpdfl/vpdfl_kernel_pdf.h
contrib/mul/vpdfl/vpdfl_kernel_pdf.cxx

Generated on Thu Jan 10 14:43:34 2008 for contrib/mul/vpdfl by

1.4.4


Public Member Functions
	vpdfl_kernel_pdf ()
	Dflt ctor.
virtual	~vpdfl_kernel_pdf ()
	Destructor.
void	set_centres (const vnl_vector< double > *x, int n, double width)
	Initialise so all kernels have the same width.
void	set_centres (const vnl_vector< double > *x, int n, const vnl_vector< double > &width)
	Initialise so all kernels have given width.
const vcl_vector< vnl_vector< double > > &	centre () const
	Position of kernel centres.
const vnl_vector< double > &	width () const
	Width of each kernel.
bool	all_same_width () const
	True if all kernels have the same width.
short	version_no () const
	Version number for I/O.
virtual vcl_string	is_a () const
	Name of the class.
virtual bool	is_class (vcl_string const &s) const
	Does the name of the class match the argument?.
virtual void	print_summary (vcl_ostream &os) const
	Print class to os.
virtual void	b_write (vsl_b_ostream &bfs) const
	Save class to binary file stream.
virtual void	b_read (vsl_b_istream &bfs)
	Load class from binary file stream.
const vnl_vector< double > &	mean () const
	Mean of distribution.
const vnl_vector< double > &	variance () const
	Variance of each dimension.
int	n_dims () const
	Number of dimensions.
virtual int	n_peaks () const
	Number of peaks of distribution.
virtual const vnl_vector< double > &	peak (int) const
	Position of the i'th peak.
virtual double	log_p (const vnl_vector< double > &x) const =0
	Log of probability density at x.
virtual double	operator() (const vnl_vector< double > &x) const
	Probability density at x.
virtual void	gradient (vnl_vector< double > &g, const vnl_vector< double > &x, double &p) const =0
	Gradient and value of PDF at x.
virtual void	gradient_logp (vnl_vector< double > &g, const vnl_vector< double > &x) const
	Gradient and value of log(p(x)) at x.
virtual vpdfl_sampler_base *	new_sampler () const =0
	Create a sampler object on the heap.
virtual double	log_prob_thresh (double pass_proportion) const
	Compute threshold for PDF to pass a given proportion.
virtual void	nearest_plausible (vnl_vector< double > &x, double log_p_min) const =0
	Compute nearest point to x which has a density above a threshold.
virtual bool	is_valid_pdf () const
	Return true if the object represents a valid PDF.
virtual vpdfl_pdf_base *	clone () const =0
	Create a copy on the heap and return base class pointer.
Protected Member Functions
void	calc_mean_var ()
	Compute mean/variance given current centres and widths.
void	set_mean (const vnl_vector< double > &m)
void	set_variance (const vnl_vector< double > &v)
Protected Attributes
vcl_vector< vnl_vector< double > >	x_
	Position of kernel centres.
vnl_vector< double >	width_
	Width of each kernel.
bool	all_same_width_
	True if all kernels have the same width.