GAMIT/test_GAMIT_Net.m at master · InfantLab/GAMIT · 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
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
function [AllRet, AllPro] = test_GAMIT_Net(seed)
%testing the Simple Recurrent Network version of GAMIT
%this script generates the figures found in
%Addyman & Mareschal (2014) seed = 2014;

if nargin < 1
    rng('shuffle');
else
    % fix a random seed for replication
    rng(seed);
end

params = GAMIT_Params();
testCases = 20:20:1800;
nLearners = 20;
nAges = 20;
nCurves = 50;
nSamples = 50;

%%%%%%%% SIMULATION 1 %%%%%%%%%%
%first lets look at learning..
%initialize storage variables
AllRet = zeros(nLearners,nAges,length(testCases));
AllPro = AllRet;
for learner = 1:nLearners
    learner
    %initialize weights
    wt1 = []; wt2 = [];
    for age = 1:nAges
        age
        %train the network
        [wt1, wt2] = GAMIT_Learning(params,nCurves,nSamples,wt1,wt2);
        %see what it outputs & store results
        AllRet(learner,age,:) = GAMIT_Net(testCases,1,0,0,params,wt1,wt2);
        AllPro(learner,age,:) = GAMIT_Net(testCases,1,1,0,params,wt1,wt2);
    end
end

epochs = 1:nAges;
TotalTrainingItems = nCurves*nSamples*1:nAges;
%find average
RetByAge = squeeze(mean(AllRet,1));
ProByAge = squeeze(mean(AllPro,1));
plotGraphs(testCases,RetByAge,ProByAge,'Retrospective SRN Learning','Prospective SRN Learning','Time Estimate',true,true,epochs);

%%%%%%%%%% SIMULATION 2 %%%%%%%%%%%%%%%
% %now lets look at error in a single trained network
% %can use last network trained in sim 1.
 %or train a new one
 nCurves = 500;
 nSamples = 100;
 [wt1, wt2] = GAMIT_Learning(params,nCurves,nSamples);
nTrials = 20;
AllRet = zeros(nTrials,length(testCases));
AllPro = AllRet;
for i = 1:nTrials
    AllRet(i,:) = GAMIT_Net(testCases,1,0,0,params,wt1,wt2);
    AllPro(i,:) = GAMIT_Net(testCases,1,1,0,params,wt1,wt2);
end
ErrRet = std(AllRet,1);
ErrPro = std(AllPro,1);
plotGraphs(testCases,ErrRet,ErrPro,'Retrospective SD','Prospective SD','Estimates Standard Deviation',false, false,0);

RelErrRet = ErrRet ./ mean(AllRet,1);
RelErrPro = ErrPro ./ mean(AllPro,1);
%least squares linear regression
p = polyfit(testCases,RelErrRet,1);
fittedRet = polyval(p,testCases);
p = polyfit(testCases,RelErrPro,1);
fittedPro = polyval(p,testCases);
plotGraphs(testCases,[RelErrRet;fittedRet],[RelErrPro;fittedPro],'Retrospective Relative Error','Prospective Relative Error','Error / Interval Length',false, false,0);

%%%%%%%%%% SIMULATION 3 %%%%%%%%%%%%%%%
% % %now lets look at cognitive load

%first plot the comparison graph to Block, Hancock and Zakay 2010
GAMIT_Net_Retro_Pro_Interaction(600,20,0.95,1.05,true,false,wt1,wt2);

% then plot lo,med,hi across full range
AllRet = zeros(3,nTrials,length(testCases));
AllPro = AllRet;
for i = 1:nTrials
    %retrospective low, normal high
    AllRet(1,i,:) = GAMIT_Net(testCases,0.9,0,0,params,wt1,wt2);
    AllRet(2,i,:) = GAMIT_Net(testCases,1.0,0,0,params,wt1,wt2);
    AllRet(3,i,:) = GAMIT_Net(testCases,1.1,0,0,params,wt1,wt2);
    %retrospective low, normal high
    AllPro(1,i,:) = GAMIT_Net(testCases,0.9,1,0,params,wt1,wt2);
    AllPro(2,i,:) = GAMIT_Net(testCases,1.0,1,0,params,wt1,wt2);
    AllPro(3,i,:) = GAMIT_Net(testCases,1.1,1,0,params,wt1,wt2);
end

Ret = squeeze(mean(AllRet,2));
Pro = squeeze(mean(AllPro,2));

plotGraphs(testCases,Ret,Pro,'Retrospective Low vs Normal vs High Cog Load','Prospective Low vs Normal vs High Cog Load','Time Estimates',true,false, 0);
end

function plotGraphs(testCases, retroResults, prospResults, retroTitle, prospTitle,ylab,includeDiagonal,showcolorbar,epochs)

    figure;
    [r,~] = size(retroResults);
    cc=hsv(r);
    colormap(cc);
    if showcolorbar
        ticks = 2:2:length(epochs);
    end

    subplot(2,1,1);
    hold on;
    for i=1:r
        plot(testCases,retroResults(i,:),'Color',cc(i,:));
    end
    if includeDiagonal
         plot(testCases,testCases,':k');
    end
    xlabel('Target Time');
    ylabel(ylab);
    title(retroTitle);
    if showcolorbar
        hd=colorbar('location','EastOutside','YTick',ticks,'YTickLabel',epochs(ticks));
        set(get(hd,'title'),'String','Epochs');
    end

    hold off;
    subplot(2,1,2);
    hold on;
    for i=1:r
        plot(testCases,prospResults(i,:),'Color',cc(i,:));
    end
    if includeDiagonal
        plot(testCases,testCases,':k');
    end
    xlabel('Target Time');
    ylabel(ylab);
    title(prospTitle);
    hold off;
    if showcolorbar
        hd= colorbar('location','EastOutside','YTick',ticks,'YTickLabel',epochs(ticks));
        set(get(hd,'title'),'String','Epochs');
    end
    figureHandle = gcf;
    %# make all text in the figure to size 14 and bold
    set(findall(figureHandle,'type','text'),'fontSize',14,'fontWeight','bold')
end