Changeset 983

Timestamp:

05/25/10 22:33:29 (14 years ago)

Author:

smidl

Message:

doc - extensions

Files:

• applications/bdmtoolbox/mex/mex_classes/mexDirac.m (modified) (3 diffs)
• applications/bdmtoolbox/mex/mex_classes/mexLaplaceBM.m (modified) (2 diffs)
• library/doc/tutorial/000install.dox (modified) (6 diffs)
• library/doc/tutorial/005userguide0.dox (modified) (1 diff)
• library/doc/tutorial/01userguide_sim.dox (modified) (1 diff)
• library/doc/tutorial/02userguide_estim.dox (modified) (3 diffs)
• library/doc/tutorial/030userguide3.dox (modified) (1 diff)
• library/doc/tutorial/040devguide_matlab.dox (added)
• library/doc/tutorial/045devguide_cpp.dox (added)

applications/bdmtoolbox/mex/mex_classes/mexDirac.m

@@ -1 +1 @@
 classdef mexDirac < mexEpdf
     % Dirac delta probability distribution
-    properties
-        point % point of the support
+    properties                                                 % DATA structures
+        point                                                  % point of the support
     end
     methods
-        function m=mean(obj)
+        function m=mean(obj)                                   % compute mean values of the density
             m = obj.point;
         end
-        function obj=validate(obj)
+        function obj=validate(obj)                             % check if data structures are consistent
             % point should be a column
             if (size(obj.point,2)>1)
@@ -18 +18 @@
             end
         end
-        function dim=dimension(obj)
+        function dim=dimension(obj)                            % inform others about your dimensions
             dim = size(obj.point,1);
         end
-        function v=variance(obj)
+        function v=variance(obj)                               % compute variance
             v=zeros(size(obj.point));
         end
-        function l=evallog(obj,x)
+        function l=evallog(obj,x)                              % return logarithm of your density at point x
             if obj.point==x
                 l = inf;
@@ -31 +31 @@
             end
         end
-        function s=sample(obj);
+        function s=sample(obj);                                % return random sample from your density
             s = obj.point;
         end

applications/bdmtoolbox/mex/mex_classes/mexLaplaceBM.m

@@ -3 +3 @@
     % Maximum likelihood approximation of the Bayes rule is used, posterior is in the form of dirac.
     properties
-        max_window_length = 10;   % max window length
-        data_window =[];          % estimate of mu is median of observations
+        max_window_length = 10;                               % max window length (default = 10)
+        data_window =[];                                      % sliding window of data 
     end
     methods
-        function obj=validate(obj)
-            % prepare all internal objects for use
+        function obj=validate(obj)                            % prepare all internal objects for use
             obj.apost_pdf = mexDirac;
             obj.apost_pdf.point = [0;0];
-            obj.log_evidence = 0; % evidence is not computed!
+            obj.log_evidence = 0;                             % evidence is not computed!
             disp('val');
         end
 
         function dims=dimensions(obj)
-            %please fill
-            %dims = [size_of_posterior size_of_data size_of_condition]
-            dims = [2,1,0] % we have: [2d parameters, 1d observations, 0d condition]
+            %please fill: dims = [size_of_posterior size_of_data size_of_condition]
+            dims = [2,1,0]                                    % we have: [2d parameters, 1d observations, 0d condition]
         end
-        function obj=bayes(obj,dt,cond)
+        function obj=bayes(obj,dt,cond)                       % approximate bayes rule
             if size(obj.data_window,2)>=obj.max_window_length
                 obj.data_window = [dt obj.data_window(1:end-1)];
@@ -29 +27 @@
             m_hat = mean(obj.data_window);
             b_hat = sum(abs(obj.data_window-m_hat))/ size(obj.data_window,2);
-            obj.apost_pdf.point = [m_hat; b_hat];
+            obj.apost_pdf.point = [m_hat; b_hat];             % store result in psoterior pdf
         end
-        function p=epredictor(obj,cond)
+        function p=epredictor(obj,cond)                       % when predictive density is needed approximate it by Laplace with point estimates
             % return predictive density (max likelihood)
             p = mexLaplace;

library/doc/tutorial/000install.dox

@@ -6 +6 @@
 
 BDM comes in two basic versions:
-\li \b binary   with precompiled executables and matlab mex files that can be configured by external text files
-\li \b source \b code  that can be compiled into any required form. 
+ - \b binary   with precompiled executables and matlab mex files that can be configured by external text files
+ - \b source \b code  that can be compiled into any required form. 
 
 \section in_bin Binaries for BDMtoolbox
@@ -13 +13 @@
 
 Installation:
-\li svn checkout http://mys.utia.cas.cz:1800/svn/mixpp/applications/bdmtoolbox/ into your local directory of your choice (this will be denoted \<toolbox_dir\>)
-\li or download a prepared zip file from http://staff.utia.cas.cz/smidl/Public/bdmtoolbox_win32.zip and extract it to any directory (will be denoted \<toolbox_dir\>)
-\li open matlab and type:
+ - svn checkout http://mys.utia.cas.cz:1800/svn/mixpp/applications/bdmtoolbox/ into your local directory of your choice (this will be denoted \<toolbox_dir\>)
+ - or download a prepared zip file from http://staff.utia.cas.cz/smidl/Public/bdmtoolbox_win32.zip and extract it to any directory (will be denoted \<toolbox_dir\>)
+ - open matlab and type:
 \code 
 >> addpath <toolbox_dir>/mex; 
@@ -29 +29 @@
 \endcode
 For linux, the default build dir is \<toolbox_dir\>/mex, hence only me and mex_classes need to be in the path.
-\li the toolbox is ready to be used, test it e.g. by: 
+ - the toolbox is ready to be used, test it e.g. by: 
 \code >> cd tutorial/userguide; 
 >> pdfds_example \endcode
-\li if no error is given, the toolbox is installed correctly
-\li proceed to tutorial at http://mys.utia.cas.cz:1800/trac/bdm/doxygen/pages.html to learn how to use it
+ - if no error is given, the toolbox is installed correctly
+ - proceed to tutorial at http://mys.utia.cas.cz:1800/trac/bdm/doxygen/pages.html to learn how to use it
 
 \section src Source code
@@ -47 +47 @@
 
 You need to have the following tools available on your system:
-\li working C++ compiler, e.g. gcc on Linux, or Visual Studio on windows
-\li CMake, a cross-platform build system, http://www.cmake.org
+ - working C++ compiler, e.g. gcc on Linux, or Visual Studio on windows
+ - CMake, a cross-platform build system, http://www.cmake.org
  
 \section lin Linux
@@ -89 +89 @@
 Compilation of BDM:
 
- * When everything is set-up, make a copy of library\system\win32\run_cmake.bat, edit it 
+ - When everything is set-up, make a copy of library\system\win32\run_cmake.bat, edit it 
   to reflect your environment (i.e. change paths to cmake and your C++ target environment) 
   and run it.
 
- * Alternatively, you can run CMake GUI and specify address to your \<bdm_dir\>
+ - Alternatively, you can run CMake GUI and specify address to your \<bdm_dir\>
 
 
- * It should generate build files for your compiler in the main BDM 
+ - It should generate build files for your compiler in the main BDM 
   directory
 
@@ -108 +108 @@
 
 Right now, Cmake configuration is hard-coded for one testing machine. Look for the paths in SetBDMEnv.cmake and correct them.
-Open CMake GUI, clisk Configure followed by Create.
+Open CMake GUI, click Configure followed by Create.
 Configuration files for your favourite build environment should be created in the BDM directory.
 

library/doc/tutorial/005userguide0.dox

@@ -41 +41 @@
  - \ref userguide_pdf
  - \ref userguide_sim
-- \ref userguide_estim
+ - \ref userguide_estim
 
 */

library/doc/tutorial/01userguide_sim.dox

@@ -70 +70 @@
  -# return data observed at time \f$ t \f$, 
  -# perform one a time step.
- -# describe what these data are, via class RV, introduced in ref \ug_pdf_marg,
+ -# describe what these data are, via class RV, introduced in ref \ref ug_pdf_marg,
  -# store log of its activity into dedicated logger.
 

library/doc/tutorial/02userguide_estim.dox

@@ -28 +28 @@
 \li Result Logger is an object (class logger) dedicated to storing important data from the experiment.
 
-Since objects  datasource and the logger has already been introduced in section \ref userguide, it remains to introduce 
+Since objects  datasource and the logger has already been introduced in section \ref userguide_sim, it remains to introduce 
 object \c Bayesian \c Model (bdm::BM).
 
@@ -62 +62 @@
  - logbounds - store also lower and upper bounds on estimates (obtained by calling BM::posterior().qbounds()),
  - logevidence - store also evidence of each step of the Bayes rule.
-These values are stored in given logger (\ref ug_loggers). By default, only mean values of the estimate are stored.
+These values are stored in given logger (\ref ug_store). By default, only mean values of the estimate are stored.
 
 Storing of the evidence is useful, e.g. in model selection task when two models are compared.
@@ -179 +179 @@
 Similarly to the extension of pdf, the estimators (or filters) can be extended via prepared class \c mexBM in directory bdmtoolbox/mex/mex_classes.
 
-An example of such class is mexLaplaceBM in \<toolbox_dir\>\tutorial\userguide\laplace_example.m
+An example of such class is mexLaplaceBM in \<toolbox_dir\>/tutorial/userguide/laplace_example.m
 
 Note that matlab-extended classes of mexEpdf, specifically, mexDirac and mexLaplace are used as outputs of methods posterior and epredictor, respectively.

library/doc/tutorial/030userguide3.dox

@@ -2 +2 @@
 \page userguide_ctrl BDM Use - Control and Design of Decision Making Strategy
 
-Bayesian estimation as introduced in Section \ref userguide2 is a special case of Decision Strategy, see. ....
+Bayesian estimation as introduced in Section \ref userguide_estim is a special case of Decision Strategy, see. ....
 A more demanding scenario is to design a control strategy that will influence the system operating under uncertainty, this is implemented as scenario \c closedloop.