Changeset 1054

Show
Ignore:
Timestamp:
06/07/10 17:25:57 (14 years ago)
Author:
smidl
Message:

doc

Location:
applications/bdmtoolbox
Files:
3 added
1 removed
11 modified
14 moved

Legend:

Unmodified
Added
Removed

  • applications/bdmtoolbox/build_doc.sh

    r1049 r1054  
    55this_dir=`pwd` 
    66cd doc/from_bdm 
    7 ./make_all 
     7./make_all.sh 
    88 
    99cd $this_dir 

  • applications/bdmtoolbox/doc/from_bdm/make_all.sh

    r1053 r1054  
    66./process_class.sh bdm::DS 
    77./process_class.sh bdm::logger 
     8./process_class.sh bdm::Controller 
    89 
    910cp -u ../../../../library/doc/html/*.png ../html/bdm_doc 

  • applications/bdmtoolbox/doc/from_bdm/process_class.sh

    r1052 r1054  
    2323 
    2424cat tmp__classes | xargs ./filter_annotated.pl 
    25 cp ../html/bdm_doc/annotated.html ../html/bdm_doc/annotated_$CLASS.html 
     25 
     26CLS_DEF=`echo $CLASS | sed "s/::/_/g"` 
     27cp ../html/bdm_doc/annotated.html ../html/bdm_doc/annotated_$CLS_DEF.html 

  • applications/bdmtoolbox/doc/local/001wrappers.dox

    r1044 r1054  
    11/*! 
    2 \page bdt_wrappers Elementary functions operating on descriptive matlab structures 
     2\page bdt_wrappers Use Case #2: Matlab wrappers for BDM functions 
    33 
    44A wrapper is a mex file that does the following: 
    … …  
    88  
    99The use of these functions is inefficient, however, it is very usefull for: 
    10  - operations on parts of the experiment configurations for scenarios, see example in \ref some_ref 
     10 - operations on parts of the experiment configurations for scenarios, see ... 
    1111 - insight into the results, e.g. mean values of non-standard densities  
    1212 - algorithms written in Matlab using these functions can be trivially re-written into C++ 
    13   
    1413  
    1514General notation of wrappers is: <b> \<class\>_\<function\>.m </b> 
    … …  
    6665 
    6766The follwing wrappers are provided: 
    68  - \b epdf_mean returning mean value of any epdf function in \ref annotated_epdf. 
     67 - \b epdf_mean returning mean value of any epdf function in <a href="annotated_bdm_epdf.html"> list </a>. 
    6968 \code  
    7069 >> G.class='egamma'; 
    … …  
    7776    0.7500  
    7877 \endcode 
    79  - \b epdf_variance returning variance of any epdf function in \ref annotated_epdf. 
     78 - \b epdf_variance returning variance of any epdf function in <a href="annotated_bdm_epdf.html"> list </a>.. 
    8079 \code 
    8180 >> epdf_variance(G) 
    … …  
    103102\endcode 
    104103 
    105 For up-to-date list of wrappers see files in \ref files.html 
     104For up-to-date list of wrappers see files in <a href="files.html"> files.html</a> 
    106105  
    107106 \section bdt_wrp_off Wrappers of offsprings 

  • applications/bdmtoolbox/doc/local/002mex_classes.dox

    r1044 r1054  
    11/*! 
    2 \page mex_bdm  BDMToolbox Development - combining Matlab classes and BDM classes 
     2\page mex_bdm  Use Case #3: combining Matlab classes and BDM classes 
    33 
    44The classes written in Matlab can be combined with standard BDM classes via corresponding C++ classes with equal names, i.e. mexBM.m is accepted by C++ class mexBM 

  • applications/bdmtoolbox/doc/local/01userguide_pdf.dox

    r1044 r1054  
    4141\endcode 
    4242 
    43 Other distributions are created analogously, see \ref bdm_doc/annotated_epdf.html 
     43Other distributions are created analogously, see <a href="annotated_bdm_epdf.html"> list </a>. 
    4444 
    4545\section ug_pdf_marg Marginalization and conditioning 
    … …  
    104104Compulsory fields \c g.dim and \c g.dimc are used to check correct dimension of inputs and outputs of the function.  
    105105 
    106 List of functions is in \ref bdm_doc/annotated_fnc 
     106List of <a href="annotated_bdm_epdf.html"> functions </a>. 
    107107 
    108108\section ug_pdf_mex Using Matlab classes of pdfs 
    … …  
    117117See \ref mex_bdm, and files: mex/mex_classes/mexEpdf.m 
    118118 
    119 If you wish to write your own Matlab classes see \ref ug_dev_mat. 
     119If you wish to write your own Matlab classes see \ref devguide_mat. 
    120120 
    121 For list of all available pdf objects, see \ref annotated_epdf.html , \ref annotated_pdf.html and \ref annotated.html 
     121For list of all available pdf objects, see <a href="annotated_bdm_epdf.html"> BDM epdfs </a>, <a href="annotated_bdm_pdf.html"> BDM pdfs </a> and <a href="annotated.html"> Matlab classes </a>. 
    122122*/ 

  • applications/bdmtoolbox/doc/local/02userguide_sim.dox

    r1044 r1054  
    22\page userguide_sim BDM Use - System, Data, Simulation 
    33 
    4 This section serves as introduction to the scenario of data simulation. Since it is the simplest of all scenarios defined in \ref userguide0 it also serves as introduction to configuration of an experiment (see \ref ui) and basic decision making objects (bdm::RV and bdm::DS). 
     4This section serves as introduction to the scenario of data simulation. Since it is the simplest of all scenarios it also serves as introduction to configuration of an experiment (see \ref ui) and basic decision making objects (bdm::RV and bdm::DS). 
    55 
    66All experiments are demonstrated on mex file \c simulator, which is also available as a standalone application. 
    … …  
    7474 
    7575No further specification, e.g. if the data are pre-recorded or computed on-the-fly, are given. 
    76 For a list of available DataSources, see ... 
     76For a list of available DataSources, see <a href="annotated_bdm_DS.html"> list </a>. 
    7777 
    7878 
    … …  
    131131A standard Data source has two levels, \c logdt and \c logut which means "log all outputs, dt" and "log all inputs, ut". 
    132132Readers familiar with Simulink environment may look at the RV as being unique identifiers of inputs and outputs of simulation blocks. The inputs are connected automatically with the outputs with matching RV. This view is however, very incomplete, RV have more roles than this. 
     133 
     134List is available <a href="annotated_bdm_logger.html"> loggers </a>. 
    133135 
    134136\section ug_pdfds How to create a simulator from pdfs  
    … …  
    250252\endcode 
    251253 
    252 For list of all DataSources and loggers, see \ref app_base 
     254List of all <a href="annotated_bdm_DS.html"> DataSources </a> and <a href="annotated_bdm_logger.html">loggers </a>. 
    253255*/ 

  • applications/bdmtoolbox/doc/local/03userguide_estim.dox

    r1044 r1054  
    4343Implementation of these operations is heavily dependent on the specific class of prior pdf, or its approximations. We can identify only a few principal approaches to this problem. For example, analytical estimation which is possible within sufficient the Exponential Family, or estimation when both prior and posterior are approximated by empirical densities.  
    4444These approaches are first level of descendants of class \c BM, classes bdm::BMEF and bdm::PF, respectively. 
     45 
     46List of all available <a href="annotated_bdm_BM.html"> Bayesian Models </a>. 
    4547 
    4648\section ug2_arx_basic Estimation of ARX models 
    … …  
    185187In order to create a new extension of an estimator, copy file with class mexLaplaceBM.m and redefine the methods therein. If needed create new classes for pdfs by inheriting from mexEpdf, it the same way as in the mexLaplace.m example class. 
    186188 
    187 For list of all estimators, see \ref app_base. 
     189For list of all Matlab estimators, see <a href="annotated.html"> list </a>. 
     190 
    188191*/ 

  • applications/bdmtoolbox/doc/local/04userguide_ctrl.dox

    r1044 r1054  
    4242variants and approximations of dynamic programming (or optimal control), see []. 
    4343 
     44List of available <a href="annotated_bdm_Controller.html"> Controllers</a>. 
    4445TODO... 
    4546 

  • applications/bdmtoolbox/doc/local/20devguide_matlab.dox

    r1044 r1054  
    1919See relevant documentation in Matlab. 
    2020 
    21 For list of existing mex* classes see \ref annotated.html 
     21For list of existing mex* classes see <a href="annotated.html"> list </a>. 
    2222 
    2323\section dev_mat_pdf Creating your own probability density 

  • applications/bdmtoolbox/mex/mex_classes/mexBM.m

    r944 r1054  
     1%> @file mexEpdf.m 
     2%> @brief File mapping root class of BM from BDM 
     3% ====================================================================== 
     4%> @brief Abstract class of Bayesian Model (estimator), bdm::BM 
     5% 
     6%> This class provides a bridge between bdm::BM and Matlab 
     7% ====================================================================== 
    18classdef mexBM 
    29    properties 
    3         % description of internal variables 
     10        %> description of internal variables of parameters 
    411        rv=RV 
     12        %> description of internal variables of data in condition 
    513        rvc=RV 
     14        %> description of internal variables of observed data 
    615        rvy=RV 
    7         % log of evidence (marginal likelihood) potentially computed by the 
     16        %> log of evidence (marginal likelihood) potentially computed by the 
    817        % bayes rule for one step 
    918        log_evidence 
    10         % posterior density - offspring of mexEpdf! 
     19        %> posterior density - offspring of mexEpdf! 
    1120        apost_pdf 
    1221        % 
    … …  
    1423 
    1524    methods 
     25        %> check consistency of the object and fill defaults 
    1626        function p=validate(p) 
    1727            % checks if all paramateres match 
    1828        end 
     29        %> dimensionality of the class: dims = [size_of_posterior size_of_data size_of_condition] 
    1930        function dims=dimensions(p) 
    2031            %please fill 
    … …  
    2233            dims = [0,0,0] % 
    2334        end 
     35        %> Performs Bayesian update of the internal posterior using data \a dt and condition \a cond. 
    2436        function obj=bayes(obj,dt,cond) 
    2537            % transform old estimate into new estimate 
    2638        end 
     39        %> Computes predictor of the observed data in the next step 
    2740        function p=epredictor(obj,cond) 
    2841            % return posterior density 
    … …  
    3952            r=obj.rvy; 
    4053        end 
     54        %> Evidence of the last data, \f[f(y_t|y_0\ldots y_t-1, cond_0\ldots cond_t\f] 
    4155        function ev = logevidence(obj) 
    4256            ev = obj.log_evidence; 
    4357        end 
     58        %> Posterior pdf 
    4459        function post=posterior(obj); 
    4560            post = obj.apost_pdf; 

  • applications/bdmtoolbox/mex/mex_classes/mexDirac.m

    r983 r1054  
     1%> @file mexDirec.m 
     2%> @brief Matlab implementation of Dirac density  
     3% ====================================================================== 
     4%> @brief Unconditional Dirac density 
     5% 
     6%> \f[ f(x| x_i) = \delta(x-x_i)\f] 
     7% ====================================================================== 
    18classdef mexDirac < mexEpdf 
    29    % Dirac delta probability distribution 

  • applications/bdmtoolbox/mex/mex_classes/mexLaplace.m

    r944 r1054  
     1%> @file mexLaplace.m 
     2%> @brief Matrlab implemnetation of Laplace density 
     3% ====================================================================== 
     4%> @brief Unconditional Laplace density 
     5% 
     6%> \f[ f(x|\mu,b) \propto \exp(-|x-\mu|/b)\f] 
     7% ====================================================================== 
    18classdef mexLaplace < mexEpdf 
    29    properties