wolffd@0: function [out, stats, features, individual] = test_generic_display_results(file, get_features, show)
wolffd@0: % [out, stats] = test_generic_display_results([file], get_features)
wolffd@0: % 
wolffd@0: % [out, stats, features, individual] = test_generic_display_results([file], get_features)
wolffd@0: % 
wolffd@0: % displays the finalresults mat file and enables 
wolffd@0: % further analysis and duiagnostics of the individual runs
wolffd@0: 
wolffd@0: features = [];
wolffd@0: 
wolffd@0: if nargin < 3
wolffd@0:     show = 1;
wolffd@0: end
wolffd@0: if nargin < 2
wolffd@0:     get_features = 1;
wolffd@0: end
wolffd@0: 
wolffd@0: global comparison;
wolffd@0: global comparison_ids;
wolffd@0:         
wolffd@0: if nargin < 1 || isempty(file) || isnumeric(file)
wolffd@0:     u = dir();
wolffd@0:     u = {u.name};
wolffd@0:     [idx, strpos] = substrcellfind(u, '_finalresults.mat', 1);
wolffd@0:     
wolffd@0:     if numel(idx) < 1
wolffd@0:         error 'This directory contains no valid test data';
wolffd@0:     end
wolffd@0:     
wolffd@0:     if exist('file','var') && isnumeric(file)
wolffd@0:         file = u{idx(file)};
wolffd@0:     else
wolffd@0:         if numel(idx) > 1
wolffd@0:             file = u{idx(ask_dataset())};
wolffd@0:         else
wolffd@0:             file = u{idx(1)};
wolffd@0:         end
wolffd@0:     end
wolffd@0: end
wolffd@0: 
wolffd@0: 
wolffd@0: % ---
wolffd@0: % LOAD THE RESULT DATA
wolffd@0: % We have:
wolffd@0: % Y
wolffd@0: % out.fparams
wolffd@0: %     trainparams
wolffd@0: %     dataPartition
wolffd@0: %     mean_ok_test
wolffd@0: %     var_ok_test
wolffd@0: %     mean_ok_train
wolffd@0: % ---
wolffd@0: load(file);
wolffd@0: 
wolffd@0: % compability
wolffd@0: if isfield(out, 'mlrparams')
wolffd@0:     for i = 1:numel(out)
wolffd@0:         out(i).trainparams = out(i).mlrparams;
wolffd@0:     end
wolffd@0: end
wolffd@0: 
wolffd@0: 
wolffd@0: % ---
wolffd@0: % % get statistics for feature parameters
wolffd@0: % Visualise the accuracy and variance
wolffd@0: % ---
wolffd@0: if isfield(out, 'inctrain') && show
wolffd@0:     for i = 1:numel(out)
wolffd@0:         
wolffd@0:        figure;
wolffd@0:        boxplot([out(i).inctrain.mean_ok_test], sqrt([out(i).inctrain.var_ok_test]), [out(i).inctrain.mean_ok_train]);
wolffd@0:        set(gca,'XTick',1:numel(out(i).inctrain.trainfrac), ...
wolffd@0:         'XTickLabel', out(i).inctrain.trainfrac* 100);
wolffd@0: 
wolffd@0:        xlabel ('fraction of training data');
wolffd@0:        title (sprintf('increasing training size test, config %d',i));
wolffd@0:        legend('train', 'train weighted', 'test', 'test weighted');
wolffd@0:        
wolffd@0:     end
wolffd@0: end
wolffd@0: 
wolffd@0: 
wolffd@0: if numel([out.mean_ok_test]) > 1 && show
wolffd@0:     
wolffd@0:     % plot means  % plot std = sqrt(var) % plot training results
wolffd@0:     figure;
wolffd@0:     boxplot([out.mean_ok_test], sqrt([out.var_ok_test]), [out.mean_ok_train]);
wolffd@0:     title (sprintf('Performance for all configs'));
wolffd@0: end
wolffd@0:     
wolffd@0:     
wolffd@0: % --- 
wolffd@0: % write max. test success
wolffd@0: % ---
wolffd@0:     mean_ok_test = [out.mean_ok_test];
wolffd@0:     [val, idx] = max(mean_ok_test(1,:));
wolffd@0: if show
wolffd@0:     fprintf(' --- Maximal test set success: nr. %d, %3.2f percent. --- \n', idx, val * 100)
wolffd@0: end
wolffd@0: 
wolffd@0: % ---
wolffd@0: % display parameter statistics
wolffd@0: % ---
wolffd@0: stats = test_generic_display_param_influence(out, show);
wolffd@0: 
wolffd@0: 
wolffd@0: if nargout < 3 
wolffd@0:     return; 
wolffd@0: end
wolffd@0: % ---
wolffd@0: % display statistics and get features
wolffd@0: %  for run with best test success
wolffd@0: % ---
wolffd@0: [resfile, featfile] = get_res_filename(out, idx);
wolffd@0: 
wolffd@0: % ---
wolffd@0: % import features:
wolffd@0: % 1. reset databse
wolffd@0: % 2. import features
wolffd@0: % 3. assign to clip ids as in ranking
wolffd@0: % ---
wolffd@0: if get_features
wolffd@0:     type = MTTAudioFeatureDBgen.import_type(featfile);
wolffd@0:     db_name = MTTAudioFeatureDBgen.db_name(type);
wolffd@0:     eval(sprintf('global %s', db_name));
wolffd@0:     eval(sprintf('%s.reset();', db_name));
wolffd@0:     eval(sprintf('features = %s.import(featfile);', db_name));
wolffd@0: 
wolffd@0:     if isfield(out,'clip_ids')
wolffd@0:         clips = MTTClip(out(1).clip_ids);
wolffd@0:         features = clips.features(type);
wolffd@0:     end
wolffd@0: end
wolffd@0: 
wolffd@0: % ---
wolffd@0: % Display Metric Stats
wolffd@0: % tmp = test_mlr_display_metric_stats(individual.out, individual.diag, features);
wolffd@0: % ---
wolffd@0: 
wolffd@0: if nargout < 4 
wolffd@0:     return; 
wolffd@0: end
wolffd@0: individual = load(resfile);
wolffd@0: for i = 1:numel(out)
wolffd@0:  
wolffd@0:         [resfile, featfile] = get_res_filename(out, i);
wolffd@0:         
wolffd@0:         if get_features
wolffd@0:             % reset db and load testing features
wolffd@0:             eval(sprintf('global %s', db_name));
wolffd@0:             eval(sprintf('%s.reset();', db_name));
wolffd@0:             eval(sprintf('%s.import(featfile);', db_name));
wolffd@0:         end
wolffd@0:     
wolffd@0:         % load individual results
wolffd@0:         if i == 1;
wolffd@0:             
wolffd@0:             individual = load(resfile);
wolffd@0:         else
wolffd@0:             
wolffd@0:             individual(i) = load(resfile);
wolffd@0:         end
wolffd@0: end
wolffd@0: end
wolffd@0: 
wolffd@0: function out = ask_dataset()
wolffd@0: % ---
wolffd@0: % displays the parameters of the datasets, 
wolffd@0: % and asks for the right one to display
wolffd@0: % ---
wolffd@0: clc;
wolffd@0: u = dir();
wolffd@0: u = {u.name};
wolffd@0: [idx, strpos] = substrcellfind(u, '_params.mat', 1);
wolffd@0: 
wolffd@0: for i = 1:numel(idx)
wolffd@0:     file = u{idx(i)};
wolffd@0:     fprintf('------------ Dataset nr. %d --------------\n',i);
wolffd@0:     fprintf('Filename: %s\n',file);
wolffd@0:     type(file);
wolffd@0: end
wolffd@0: 
wolffd@0: out = (input('Please choose the dataset number: '));
wolffd@0: end
wolffd@0: 
wolffd@0: 
wolffd@0: function [resfile, featfile] = get_res_filename(out, i)
wolffd@0: % get filename given test results and index
wolffd@0: 
wolffd@0:     paramhash = hash(xml_format(out(i).fparams),'MD5');
wolffd@0:     
wolffd@0:     paramhash_mlr = hash(xml_format(out(i).trainparams),'MD5');
wolffd@0: 
wolffd@0:     featfile = sprintf('runlog_%s_feat.mat', paramhash);
wolffd@0: 
wolffd@0:     resfile = sprintf('runlog_%s.%s_results.mat',...
wolffd@0:             paramhash, paramhash_mlr);
wolffd@0: end
wolffd@0: 
wolffd@0: 
wolffd@0: function boxplot(mean, std, train);
wolffd@0: 
wolffd@0:     bar([train; mean]', 1.5);
wolffd@0:     hold on;
wolffd@0:     errorbar(1:size(mean,2), mean(1,:), std(1,:),'.');
wolffd@0: %     plot(train,'rO');
wolffd@0:     colormap(spring);
wolffd@0:     axis([0 size(mean,2)+1 max(0, min(min([train mean] - 0.1))) max(max([train mean] + 0.1))]);
wolffd@0: end