Zcells/evalutation_script.m at master · Lab513/Zcells · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
% CD to the Zcells directory and then:
addpath(genpath('.'))

%% Start by creating (or loading) a (pre-existing) training set with the GUI:
training_set = createOrLoadTrainingSet();

training_set.parameters.parallel_processing.status = false; % Parallel training in evaluations not supported for now
%% Parameters for evaluation:

% General parameters:
test_percent = 1;
displayConfMat = false;
displayStackPred = true;

% Parameters to evaluate against:
evaluate = {}; % Leave empty if you just want to evaluate the current training_set

% evaluate(1) = {
%                {'parameters.feature_extraction.nbcomponents', 30}
%                };


evaluate(1) = {
               {}
               };
evaluate(2) = {   {'parameters.frame_processing' ,...
                        {struct('name','Intensity')}
                   }
               };
evaluate(3) = {
                {'parameters.class_specific.default.SVM.MinLeafSize', 3}
               };

evaluate(4) = {
                {'parameters.class_specific.default.SVM.NumTrees', 100}
               };
evaluate(5) = {
                {'parameters.class_specific.default.SVM.NumTrees', 30}
               };
evaluate(6) = {
                {'parameters.focus_shifting.radius', 3}
               };
evaluate(7) = {
                {'parameters.feature_extraction.nbcomponents', 5}
               };
evaluate(8) = {
                {'parameters.feature_extraction.nbcomponents', 8}
               };
evaluate(9) = {
                {'parameters.feature_extraction.nbcomponents', 30}
               };
evaluate(10) = {
                {'parameters.frames_subselection.type', 'log';
                'parameters.frames_subselection.nbframes_linlog', 40;
                'parameters.feature_extraction.nbcomponents', 20}
               };
evaluate(11) = {
                {'parameters.class_specific.default.SVM.MinLeafSize', 5}
               };
evaluate(12) = {
                {'parameters.class_specific.default.SVM.NumTrees', 50}
               };
evaluate(13) = {
                {'parameters.class_specific.default.SVM.NumTrees', 75}
                };
evaluate(14) = {
                {'parameters.class_specific.default.SVM.NumTrees', 75}
                };

% Run each evaluation against specific stacks:
predstack = {
                '/home/jeanbaptiste/data/Zcells/google_drive/CuratedSets/Mammalian/Stacks/Stack_13.mat', ...
                '/home/jeanbaptiste/data/Zcells/google_drive/CuratedSets/Mammalian/Stacks/Stack_14.mat', ...
                '/home/jeanbaptiste/data/Zcells/google_drive/CuratedSets/Mammalian/Stacks/Stack_15.mat'};


%% Run evalutation:

for ind1 = 1:numel(evaluate)

    fprintf("%s: Starting evaluation for parameters set #%d\n",datestr(now),ind1)
    eval_set = evalparams(training_set,evaluate,ind1);

    [data, eval_set.feature_extraction, eval_data] = LoadAndExtract(eval_set,'test_percent',test_percent);
    eval_labels{ind1} = eval_data.label;

    fprintf("%s: Launching training\n",datestr(now))
    Classifier = training(eval_set, data);

    fprintf("%s: Launching prediction\n",datestr(now))
    tic
    eval_results{ind1} = RunPrediction([], ...
		        eval_set.feature_extraction.coeff, ...
                eval_set.feature_extraction.mu, ...
		        Classifier, ...
		        'isProcessed', 1, ...
		        'processedMat', eval_data.components ...
		        );
	prediction_time(ind1) = toc;

    if displayConfMat
        eval_y_hat = format_results(eval_results{ind1}); % Only reliable for non-hierarchical cases
        eval_cm = confusionmat(eval_data.label, eval_y_hat);
        figure('Name', sprintf('Confusion Matrix #%d', ind1));
        plotConfMat(eval_cm, eval_set.classnames);
    end

    if displayStackPred
        tic
        for ind2 = 1:numel(predstack)
            tmp = matfile(predstack{ind2});
            midframe = tmp.stack(1,1);
            midframe = midframe{round(numel(midframe)/2)};
            stack_results = RunPrediction(tmp, ...
                eval_set.feature_extraction.coeff, ...
                eval_set.feature_extraction.mu, ...
                Classifier, ...
                'use_features', eval_set.parameters.frame_processing, ...
                'FramesSelection', eval_set.parameters.frames_subselection.frames...
                );
            [I, ~] = show_classes(stack_results, size(midframe),'Colors',eval_set.rgbmap);
            figure('Name', sprintf('Stack #%d, Prediction #%d', ind2, ind1));
            imshow(I);
        end
        stacks_time(ind1) = toc;
    end

    fprintf("%s: Done!\n\n",datestr(now))
    drawnow
end