-
Notifications
You must be signed in to change notification settings - Fork 0
/
HiddenMarkovModels_UPonly_all_train.m
152 lines (122 loc) · 4.26 KB
/
HiddenMarkovModels_UPonly_all_train.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
% HiddenMarkovModels.m
clear all;
ThPer = 60;
States = 2 ; %2-20
dataset = '8482_15p';
Repetitions = 10; %10
Run_Num = 10; %for the save filename
Epoch_Select = 3; % select which sleep epoch to analyze (1 = sleep1; 2 =
% sleep2; 3 = sleep3)
%UPdir = 'Z:\soroush_UPstate\Template Matching\Data\7165_16p'
output_dir = sprintf('E:/HMM - UP&Down/Soroush/Data/New_Result/%s', dataset);
% % % % data_file = sprintf('E:/HMM - UP&Down/Soroush/Data/New_Result/%s/sleep3_UP_epochs_run505.mat', dataset);
data_file = sprintf('E:/HMM - UP&Down/Soroush/Data/New_Result/%s/sleep3_UP_epochs_run505_Th%dMotionless.mat', dataset, ThPer);
cd(output_dir);
% Load data and UP/DOWN periods:
%***************************************
switch Epoch_Select
case 1
load('sleep1_HMM_data.mat')
load(data_file, 'UP_epochs')
sleep = 1;
case 2
load('sleep2_HMM_data.mat')
load(data_file, 'UP_epochs')
sleep = 2;
case 3
load('sleep3_HMM_data.mat')
load(data_file, 'UP_epochs')
sleep = 3;
otherwise
error('Wrong Value for Epoch_select')
end
filename = ['E:\HMM - UP&Down\Soroush\Data\New_Result\',num2str(dataset),'\HMM_UPonly_all_train_',num2str(States),'states_',num2str(Repetitions),'_rep_run',num2str(Run_Num),'_sleep',num2str(sleep),'_Th',num2str(ThPer),'Motionless.mat'];
% load('sleep3_HMM_data.mat')
% filename = 'data0round.txt';
% [ids times]=textread(filename,'%f %f');
UP_train = {};
for i = 1:size(UP_epochs,1)
idx = find(times > UP_epochs(i,1) & times < UP_epochs(i,2));
tt = times(idx)-UP_epochs(i,1);
T = UP_epochs(i,2)-UP_epochs(i,1)+1;
n = ids(idx);
UP_train{i} = sleepOBS(tt,n,T);
end
numNeurons=max(ids);
numEpochs=length(UP_epochs(:,1));
Vit = cell(1,numEpochs);
Vit_all = cell(1,numEpochs);
BB = cell(1,numEpochs);
tic
[HMM,EMIS_all,TRANS_all,Likelihood]=sleepHMMTraining_new(UP_train,States,numNeurons,Repetitions,200);
toc
tic
for ep=1:numEpochs
TrainStart=UP_epochs(ep,1);
TrainEnd=UP_epochs(ep,2);
T=TrainEnd-TrainStart+1;
% data in Train set:
i=find(times<TrainEnd & times>TrainStart);
tt=times(i);
n=ids(i);
tt=tt-TrainStart;
% Build observations
Obs=sleepOBS(tt,n,T);
%Training:
%**********************************************************************
% sleepHMMTraining(Obs,States,numNeurons,Repetitions,Iterations)
% [HMM,EMIS_all,TRANS_all,Likelihood]=sleepHMMTraining_new(Obs,States,numNeurons,Repetitions,200);
%
%
TRANS=HMM.TRANSITION;
EMIS=HMM.EMISSION;
HMmodels(ep).TRANS= TRANS;
HMmodels(ep).EMIS=EMIS;
HMmodels(ep).TRANS_all=TRANS_all;
HMmodels(ep).EMIS_all=EMIS_all;
HMmodels(ep).Likelihood = Likelihood;
HMmodels(ep).tt = tt;
HMmodels(ep).n = n;
HMmodels(ep).T = T;
HMmodels(ep).Obs = Obs;
HMmodels(ep).StartTime = TrainStart;
HMmodels(ep).EndTime = TrainEnd;
%Decoding stage (Viterbi)
%**********************************************************************
try
S = hmmviterbi(Obs, TRANS, EMIS); % S: most likely sequence of states
S_all = {};
for ii = 1:Repetitions
S_all{ii} = hmmviterbi(Obs, TRANS_all{ii}, EMIS_all{ii});
end
Vit{ep}=S;
Vit_all{ep} = S_all;
catch e
continue;
end
% Figures:
%**********************************************************************
% set(figure,'Position',[250 100 1000 400],'Color','w');
%
% subplot(3,1,2:3)
%
% plot(tt,n/numNeurons,'k.')
% hold on
% plot(1.2*S-1.3)
%
% tx1=2*floor(T/7);
% tx2=tx1+3*floor(T/7);
% % tx1 = 0;
% % tx2 = TrainEnd;
% % set(gca,'XColor','w','YColor','w','YLim',[-.3 1.5],'XLim',[tx1 tx2])
% set(gca,'YLim',[-.3 1.5],'XLim',[tx1 tx2])
% title(['epoch ' num2str(ep)],'FontSize',16)
% % xlim([0 TrainEnd])
%
% pause(0);
end
toc
UP_test = UP_epochs;
% save HMMSTATES Vit Vit_all
% save MarkovModels HMmodels
save(filename, 'Vit', 'Vit_all', 'BB', 'HMmodels', 'States', 'numNeurons','Repetitions', 'numNeurons', 'numEpochs', 'data_file', 'UP_train','UP_test')