-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathsim_A1.m
263 lines (217 loc) · 7.38 KB
/
sim_A1.m
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
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
function sim_A1(i)
% HBI for models with the same number of parameters: RL(10) vs KF(30)
mode = '';
if nargin==0, mode = 'sumfig'; end
if nargin==1, mode = 'run'; end
getdefaults('addpath');
switch mode
case 'run'
run(i);
case 'sumfig'
sumfig;
otherwise
error('Unknown mode: %s',mode);
end
end
function run(ii)
nsim = 20;
if nargin<1, ii=nsim; end
simcat = 'sim_A';
Nbar = [10 30];
symname = 'sim_rwkf1';
simstr = sprintf('%s%s',symname,mat2str(Nbar));
models = {'model_rw1','model_kf1'};
tasks = {'task_rw1','task_kf1'};
pnames = { {'\alpha','\beta'},{'\omega','\beta'} };
modelnames = {'Reinforcement learning','Kalman filter'};
mnames = {'RL','KF'};
tau1 = 4*[1 1]';
tau2 = 4*[1 1]';
tau = {tau1,tau2};
tx1 = [.1 2]';
tx2 = [2 2]';
logit = @(x)log(x./(1-x));
mu1 = [logit(tx1(1)); log(tx1(2))];
mu2 = log(tx2);
mu = {mu1,mu2};
normx1 = {@(x)1./(1+exp(-x)), @exp};
normx2 = {@exp,@exp};
normx = {normx1,normx2};
sim_sim(ii,simcat,simstr,Nbar,models,tasks,modelnames,mnames,pnames,mu,tau,normx);
if ii(end)~=nsim, return; end
sim_wrap(nsim,simcat,simstr,modelnames,mnames,pnames,normx);
end
function sumfig
simcat = 'sim_A';
simstr = 'rwkf[10 30]';
fsimfit = fullfile(getdefaults('pipedir'),simcat,sprintf('fit_%s.mat',simstr));
simfit = load(fsimfit);
fit = simfit.fit;
config = simfit.config;
mnames = config.mnames;
pnames = config.pnames;
[pxpm,pxps,Nbarm,Nbars,ngm,egm,rmm,rms,mdx,sdx,corrBMS,methods] = sim_stat_A1(fit,config);
% close all;
Nbarm =[Nbarm; .25 .75];
Nbars =[Nbars; 0 0];
Nbarstrs = {'HBI','True'};
sim_plot_A1(pxpm,pxps,Nbarm,Nbars,ngm,egm,rmm,rms,mdx,sdx,corrBMS,Nbarstrs,mnames,pnames,methods);
set(gcf,'name',mfilename);
end
function sim_plot_A1(pxpm,pxps,Nbar,Nbars,mgm,egm,rm,re,mdx,sdx,corrBMS,Nbarstrs,mnames,pnames,methods)
pnames = sim_plot_adjust_pnames(pnames);
[fs,fst,fsl,fsy,fsalpha,fsxt,fsA,fn,fnt,xsA,ysA,fpos0,siz0,colmap,alf,cmaphbi,bw,colmapsim]=fig_plot_properties; %#ok<*ASGLU>
cmaphbi2 = [cmaphbi; colmapsim(2,:)];
fpos = [.6 fpos0(4)];
siz = fpos./fpos0(3:4).*siz0;
xst = .5;
yst = 1.1;
ysr = .1;
bwp = .9*bw;
bwr = .9*bw;
bwi = 1.6*bw;
bws = 1.4*bw;
bwe = bw*.8;
nr = 2;
nc = 3;
figure;
set(gcf,'units','centimeters');
fsiz = get(gcf,'position');
fsiz(3:4) = siz;
fsiz(1:2) = fpos0(1:2).*fsiz(1:2);
set(gcf,'position',fsiz);
set(gcf,'units','normalized');
%---------
pxpm(pxpm<0.01)=0.01;
h(1) = subplot(nr,nc,1);
errorbarKxN(pxpm,pxps,mnames,'',cmaphbi,0,bwp);
set(gca,'fontsize',fs,'fontname',fn);
set(gca,'xticklabel',mnames,'xcolor','k');
% set(gca,'ygrid','on');
alpha(gca,alf);
ylabel(sprintf('PXP'),'fontsize',fsy);
ylim([0 1.05]);
set(gca,'ytick',0:0.2:1);
title(sprintf('Model selection by HBI'),'fontsize',fst,'fontname',fnt);
ax = ancestor(gca, 'axes');
xaxes = get(ax,'XAxis');
set(xaxes,'fontsize',fsxt);
text(xsA,ysA,'A','fontsize',fsA,'Unit','normalized','fontname','Calibri');
% figure;
h(2) = subplot(nr,nc,2);
errorbarKxN(Nbar,Nbars,mnames,'',cmaphbi2,0,bwp);
set(gca,'fontsize',fs,'fontname',fn);
set(gca,'xticklabel',mnames,'xcolor','k');
% set(gca,'ygrid','on');
alpha(gca,alf);
ylim([0 1.05])
ylabel(('Model frequency'),'fontsize',fsy);
% title(sprintf('Estimated frequency of models\n across subjects'),'fontsize',fst,'fontname',fnt);
title(sprintf('Model comparison by HBI'),'fontsize',fst,'fontname',fnt);
ax = ancestor(gca, 'axes');
xaxes = get(ax,'XAxis');
set(xaxes,'fontsize',fsxt);
text(xsA,ysA,'B','fontsize',fsA,'Unit','normalized','fontname','Calibri');
lg = legend(Nbarstrs,'location','northwest','fontsize',fsl,'Orientation','vertical'); hold on;
set(lg,'linewidth',1);
h(3) = subplot(nr,nc,3);
errorbarKxN(rm,re,{'TA','FA'},'',cmaphbi,0,bwr);
set(gca,'fontsize',fs,'fontname',fn);
% set(gca,'xticklabel',{'Correctly assigned','Incorrectly assigned'},'xcolor','k');
% set(gca,'xticklabel',{'True Positive','False Positive'},'xcolor','k','XTickLabelRotation',20);
ylabel('Responsibility','fontsize',fsy);
title(sprintf('Model attribution by HBI'),'fontsize',fst,'fontname',fnt);
ax = ancestor(gca, 'axes');
xaxes = get(ax,'XAxis');
set(xaxes,'fontsize',fsxt);
alpha(gca,alf);
% set(gca,'ygrid','on');
set(gca,'ytick',.5:0.1:1);
% set(gca,'yticklabel',.5:.1:1);
ylim([.5 1.025]);
text(xsA,ysA,'C','fontsize',fsA,'Unit','normalized','fontname','Calibri');
%------------------
basepos = get(gca,'position');
pos = [1 1].*basepos(1:2) + [.5 .57].*(basepos(3:4));
hinset = axes('Position',[pos .1 0.1]);
bar(mgm*100,'Horizontal','on','facecolor',cmaphbi,'EdgeColor','k','linewidth',1,'basevalue',50,'barwidth',bwi); hold on;
plot([mgm-egm mgm+egm]*100,[1 1],'-','color','k','linewidth',2);
xlim([50 100]);
% set(gca,'box','off');
set(hinset,'xtick',50:10:90);
set(hinset,'xticklabel',50:10:90);
set(hinset,'ticklength', [0 0]);
set(hinset,'yticklabel',[],'xcolor','k');
set(hinset,'linewidth',1);
title('Accuracy %','fontsize',fsxt,'fontname',fnt);
alpha(gca,alf);
%--------------------
%--
% colmap = [.2 .2 .2; .5 .5 .5; .9 .9 .9];
% plotbarit(corrBMS,0*corrBMS,methods);
h(4) = subplot(2,3,4);
corrBMS = max(corrBMS,1);
errorbarNxK(corrBMS,corrBMS*0,methods,'',colmap,0,bws);
set(gca,'fontsize',fs,'fontname',fn);
ylabel('Correct selection %','fontsize',fsy);
% title(sprintf('Model selection\nperformance'),'fontsize',fst,'fontname',fnt);
set(gca,'ytick',0:20:100);
ylim([0 110]);
ax = ancestor(gca, 'axes');
xaxes = get(ax,'XAxis');
set(xaxes,'fontsize',fsxt);
text(xsA,ysA+ysr,'D','fontsize',fsA,'Unit','normalized','fontname','Calibri');
alpha(gca,alf);
text(xst,yst,sprintf('Model selection\nperformance'),'fontsize',fst,'fontname',fnt,...
'Unit','normalized','fontweight','bold','Parent',h(4),'HorizontalAlignment','center'); hold on;
get(gca,'Position');
% figure for parameters
EF ={'E','F'};
if ~isempty(mdx)
for k=[1 2]
hk = subplot(2,3,4+k);
pos = get(hk,'Position');
text(xst,yst,sprintf('Parameter estimation\n performance for %s',mnames{k}),'fontsize',fst,'fontname',fnt,...
'Unit','normalized','fontweight','bold','Parent',hk,'HorizontalAlignment','center'); hold on;
set(hk,'visible','off');
text(xsA,ysA+ysr,EF{k},'fontsize',fsA,'Unit','normalized','fontname','Calibri','Parent',hk); hold on;
np = size(mdx{k},2);
dl = .05;
pos1 = pos;
x0 = pos1(1);
lp = 1/np* (pos1(3) -(np-1)*dl);
for i=1:np
pos1 = pos;
pos1(1) = x0+(i-1)*(lp+dl);
pos1(3) = lp;
axes('Position',pos1);
% errorbarKxN(mxf(:,i),exf(:,i),pnamesf(i),'',cmaphbi,0,bw*rbf); hold on;
errorbarKxN(mdx{k}(:,i),sdx{k}(:,i),pnames{k}{i},'',colmap,0,bwe); hold on;
alpha(gca,alf);
set(gca,'fontsize',fs,'fontname',fn);
% xlabel(pnames{k}{i});
ax = ancestor(gca, 'axes');
xaxes = get(ax,'XAxis');
set(xaxes,'Fontsize',fsalpha,'TickLabelInterpreter','latex');
% if ~strcmp(pnames{k}{i}(1:3),'\it')
% set(xaxes,'FontWeight','bold');
% end
if i==1
ylabel('Error','fontsize',fsy);
end
ytick = get(gca,'ytick');
ytick(end) = [];
ytick = ytick(1:2:end);
set(gca,'ytick',ytick);
end
end
end
yl = get(gca,'ylim');
yl(2) = 1.3*yl(2);
ylim(yl);
lg = legend(methods,'location','north','fontsize',fsl,'Orientation','vertical'); hold on;
set(lg,'linewidth',1);
poslg = get(lg,'position');
% poslg(1) = poslg(1)+.04;
set(lg,'position',poslg);
end