Base class for Univariate Random Sampler classes. More...
#include <pdf1d_sampler.h>
Public Member Functions | |
pdf1d_sampler () | |
Dflt ctor. | |
virtual | ~pdf1d_sampler () |
Destructor. | |
const pdf1d_pdf & | model () const |
PDF of which this is an instance. | |
virtual void | set_model (const pdf1d_pdf &) |
Set model for which this is an instance. | |
virtual double | sample ()=0 |
Draw random sample from distribution. | |
virtual void | get_samples (vnl_vector< double > &x) |
Fill x with samples drawn from distribution. | |
virtual void | regular_samples (vnl_vector< double > &x) |
Fill x with samples possibly chosen so as to represent the distribution. | |
virtual void | regular_samples_and_prob (vnl_vector< double > &x, vnl_vector< double > &p) |
Fill x with samples possibly chosen so as to represent the distribution. | |
virtual void | reseed (unsigned long)=0 |
Reseeds the internal random number generator. | |
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 pdf1d_sampler * | clone () const =0 |
Create a copy on the heap and return base class pointer. | |
virtual void | print_summary (vcl_ostream &os) const |
Print class to os. | |
Protected Attributes | |
const pdf1d_pdf * | pdf_model_ |
Base class for Univariate Random Sampler classes.
This is really a random number generator whose outputs have the PDF of the relevant pdf1d_pdf. Sampler objects should only exist as long as their pdf object, and are not meant to be persistent.
Definition at line 23 of file pdf1d_sampler.h.
pdf1d_sampler::pdf1d_sampler | ( | ) |
Dflt ctor.
Definition at line 18 of file pdf1d_sampler.cxx.
pdf1d_sampler::~pdf1d_sampler | ( | ) | [virtual] |
Destructor.
Definition at line 26 of file pdf1d_sampler.cxx.
virtual pdf1d_sampler* pdf1d_sampler::clone | ( | ) | const [pure virtual] |
Create a copy on the heap and return base class pointer.
Implemented in pdf1d_exponential_sampler, pdf1d_flat_sampler, pdf1d_gaussian_sampler, pdf1d_mixture_sampler, pdf1d_epanech_kernel_pdf_sampler, pdf1d_gaussian_kernel_pdf_sampler, and pdf1d_weighted_epanech_kernel_sampler.
void pdf1d_sampler::get_samples | ( | vnl_vector< double > & | x | ) | [virtual] |
Fill x with samples drawn from distribution.
Definition at line 45 of file pdf1d_sampler.cxx.
vcl_string pdf1d_sampler::is_a | ( | ) | const [virtual] |
Name of the class.
Reimplemented in pdf1d_exponential_sampler, pdf1d_flat_sampler, pdf1d_gaussian_sampler, pdf1d_mixture_sampler, pdf1d_epanech_kernel_pdf_sampler, pdf1d_gaussian_kernel_pdf_sampler, and pdf1d_weighted_epanech_kernel_sampler.
Definition at line 82 of file pdf1d_sampler.cxx.
bool pdf1d_sampler::is_class | ( | vcl_string const & | s | ) | const [virtual] |
Does the name of the class match the argument?.
Reimplemented in pdf1d_exponential_sampler, pdf1d_flat_sampler, pdf1d_gaussian_sampler, pdf1d_mixture_sampler, pdf1d_epanech_kernel_pdf_sampler, pdf1d_gaussian_kernel_pdf_sampler, and pdf1d_weighted_epanech_kernel_sampler.
Definition at line 91 of file pdf1d_sampler.cxx.
const pdf1d_pdf & pdf1d_sampler::model | ( | ) | const |
PDF of which this is an instance.
Definition at line 38 of file pdf1d_sampler.cxx.
void pdf1d_sampler::print_summary | ( | vcl_ostream & | os | ) | const [virtual] |
Print class to os.
Definition at line 101 of file pdf1d_sampler.cxx.
void pdf1d_sampler::regular_samples | ( | vnl_vector< double > & | x | ) | [virtual] |
Fill x with samples possibly chosen so as to represent the distribution.
Generate a set of pseudo-random samples, chosen so as to be suitable to represent the distribution. This is meant to be used for estimating continuous integrals with sampled approximations. Where there are multiple peaks (e.g. kernel or mixture models), it is preferred that the number of samples from each component is roughly proportional to the weight for the component. When small numbers are requested, this can be done explicitly. The default is simply to call sample() for each element of x
By default simply randomly sample from distribution
Reimplemented in pdf1d_exponential_sampler, pdf1d_flat_sampler, pdf1d_gaussian_sampler, pdf1d_epanech_kernel_pdf_sampler, pdf1d_gaussian_kernel_pdf_sampler, and pdf1d_weighted_epanech_kernel_sampler.
Definition at line 55 of file pdf1d_sampler.cxx.
void pdf1d_sampler::regular_samples_and_prob | ( | vnl_vector< double > & | x, |
vnl_vector< double > & | p | ||
) | [virtual] |
Fill x with samples possibly chosen so as to represent the distribution.
As regular_samples(x), but p[i] is set to p(x[i])
By default simply randomly sample from distribution
Definition at line 65 of file pdf1d_sampler.cxx.
virtual void pdf1d_sampler::reseed | ( | unsigned | long | ) | [pure virtual] |
Reseeds the internal random number generator.
To achieve quasi-random initialisation use;
#include <vcl_ctime.h>
..
sampler.reseed(vcl_time(0));
Implemented in pdf1d_exponential_sampler, pdf1d_flat_sampler, pdf1d_gaussian_sampler, pdf1d_mixture_sampler, pdf1d_epanech_kernel_pdf_sampler, pdf1d_gaussian_kernel_pdf_sampler, and pdf1d_weighted_epanech_kernel_sampler.
virtual double pdf1d_sampler::sample | ( | ) | [pure virtual] |
Draw random sample from distribution.
Implemented in pdf1d_mixture_sampler, pdf1d_exponential_sampler, pdf1d_flat_sampler, pdf1d_gaussian_sampler, pdf1d_epanech_kernel_pdf_sampler, pdf1d_gaussian_kernel_pdf_sampler, and pdf1d_weighted_epanech_kernel_sampler.
void pdf1d_sampler::set_model | ( | const pdf1d_pdf & | model | ) | [virtual] |
Set model for which this is an instance.
Reimplemented in pdf1d_mixture_sampler, pdf1d_exponential_sampler, pdf1d_flat_sampler, and pdf1d_gaussian_sampler.
Definition at line 32 of file pdf1d_sampler.cxx.
const pdf1d_pdf* pdf1d_sampler::pdf_model_ [protected] |
Definition at line 26 of file pdf1d_sampler.h.