Repository navigation
Expand file tree
/
Copy pathmain.m
More file actions
151 lines (138 loc) · 5.51 KB
/
Copy pathmain.m
File metadata and controls
151 lines (138 loc) · 5.51 KB
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
clc;clear;close all;
addpath model/ utils/ matlab_decoding/ training/
% data = data_setup(); % load data
% opt = opt_Setup(data); % set options
% s = rng();
ps = pyenv;
if ~ (ps.Status == "Loaded")
pyenv("Version", "C:\Environments\python_envs\py311_IJCNN_2603\python.exe")
end
pyfunc = py.importlib.import_module('apis_for_matlab');
py.importlib.reload(pyfunc);
% PolicyGradient_LinearLatentMapping_tVAE(data,opt,predict,s);
DataNameList = ["data/rat008_1206.mat","data/rat009_0711.mat","data/rat010_0704.mat",...
"data/rat011_0823.mat","data/rat025_0923.mat","data/rat028_1030.mat"];
for dataIdx = 6 % 5:6
DataName = DataNameList(dataIdx);
data = data_setup(DataName); % load data
opt = opt_Setup(data); % set options
opt.modelName = 'TC10_'; % save TC results to results/TC_...
rng('default');
for testFold = 5 % 1:opt.foldNum
opt.testFold = testFold;
trainFolds = 1:opt.foldNum; trainFolds(testFold) = [];
opt.trainTrials = cell2mat(opt.folds(trainFolds));
opt.NumberOfTrainTrials = length(opt.trainTrials);
opt.testTrials = opt.folds{testFold};
opt.NumberOfTestTrials = length(opt.testTrials);
opt.NumberOfAllTrials = opt.NumberOfTrainTrials + opt.NumberOfTestTrials;
mPFC_model = pyfunc.get_model(data.rat, testFold-1, opt.latent_dim, 'mPFC', size(data.mPFC, 1));
M1_model = pyfunc.get_model(data.rat, testFold-1, opt.latent_dim, 'M1', size(data.M1, 1));
% for i = 1:opt.
s = rng;
% SupervisedLearning(data,opt,pyfunc,mPFC_model,M1_model,s);
PolicyGradient_LinearLatentMapping_tVAE(data,opt,pyfunc,mPFC_model,M1_model,s)
% opt.Mode = 'train';
% opt.DataLoaderCursor = 1;
% [batchInput,batchM1_truth,batchActions,batchTrials,opt] = DataLoader_tVAE(data,opt);
% res = pyfunc.neural2latent(mPFC_model, batchInput, opt.latent_dim);
% latent_mu = double(res{1}); latent_mu = latent_mu(1:length(batchInput), :);
%
% ref_mPFC_train = load(['latent_models_backup\tVAE_' ...
% data.rat '_mPFC_' num2str(testFold-1) '_1En1De_' num2str(opt.latent_dim) 'latent_decoder_pos_train_set.mat']);
%
% figure()
% for lIdx = 1:opt.latent_dim
% subplot(opt.latent_dim, 1, lIdx)
% plot(latent_mu(101+(1:3e3), lIdx), 'r', 'LineWidth',1.5)
% hold on
% plot(ref_mPFC_train.latent_mu(1:3e3, lIdx), 'k', 'LineWidth',1.5)
% hold off
% end
%
% opt.Mode = 'test';
% [testInput,testM1_truth,testActions,testTrials,opt] = DataLoader_tVAE(data,opt);
% res = pyfunc.neural2latent(mPFC_model, testInput, opt.latent_dim);
% latent_mu = double(res{1}); latent_mu = latent_mu(1:length(testInput), :);
%
% ref_mPFC_test = load(['latent_models_backup\tVAE_' ...
% data.rat '_mPFC_' num2str(testFold-1) '_1En1De_' num2str(opt.latent_dim) 'latent_decoder_pos.mat']);
%
% figure()
% for lIdx = 1:opt.latent_dim
% subplot(opt.latent_dim, 1, lIdx)
% plot(latent_mu(101+(1:3e3), lIdx), 'r', 'LineWidth',1.5)
% hold on
% plot(ref_mPFC_test.latent_mu(1:3e3, lIdx), 'k', 'LineWidth',1.5)
% hold off
% end
% opt.Mode = 'train';
% opt.DataLoaderCursor = 1;
% [batchInput,batchM1_truth,batchActions,batchTrials,opt] = DataLoader_tVAE(data,opt);
% res = pyfunc.neural2latent(M1_model, batchM1_truth, opt.latent_dim);
% latent_mu = double(res{1}); latent_mu = latent_mu(1:length(batchM1_truth), :);
%
% ref_M1_train = load(['latent_models_backup\tVAE_' ...
% data.rat '_M1_' num2str(testFold-1) '_1En1De_' num2str(opt.latent_dim) 'latent_decoder_pos_train_set.mat']);
%
% figure()
% for lIdx = 1:opt.latent_dim
% subplot(opt.latent_dim, 1, lIdx)
% plot(latent_mu(101+(1:3e3), lIdx), 'r', 'LineWidth',1.5)
% hold on
% plot(ref_M1_train.latent_mu(1:3e3, lIdx), 'k', 'LineWidth',1.5)
% hold off
% end
%
% opt.Mode = 'test';
% [testInput,testM1_truth,testActions,testTrials,opt] = DataLoader_tVAE(data,opt);
% res = pyfunc.neural2latent(M1_model, testM1_truth, opt.latent_dim);
% latent_mu = double(res{1}); latent_mu = latent_mu(1:length(testM1_truth), :);
%
% ref_M1_test = load(['latent_models_backup\tVAE_' ...
% data.rat '_M1_' num2str(testFold-1) '_1En1De_' num2str(opt.latent_dim) 'latent_decoder_pos.mat']);
%
% figure()
% for lIdx = 1:opt.latent_dim
% subplot(opt.latent_dim, 1, lIdx)
% plot(latent_mu(101+(1:3e3), lIdx), 'r', 'LineWidth',1.5)
% hold on
% plot(ref_M1_test.latent_mu(1:3e3, lIdx), 'k', 'LineWidth',1.5)
% hold off
% end
%
% disp(['----', opt.DataIndex,' ',Sch,' fold ',num2str(opt.testFold),' Train start----']);
% % parfor(i = 1:opt.ReTrainTimes,8) % limited memory orz
% disp(['parfor loop ', num2str(i)])
% s = rng;
% % if (constraint==0)
% PolicyGradient_inRwd_Entropy_Manifold(data,opt,s,inputEnsemble,M1_truth,Actions,Trials);
% % else
% % SupervisedLearning(data,opt,s,inputEnsemble,M1_truth,Actions,Trials);
% % end
% % end
%
% paths = dir(['results/temp/',data.DataIndex,'_',Sch,'_',num2str(opt.testFold),'_','*.mat']);
% testSuc = zeros(1,length(paths));
% for i=1:length(paths)
% load(['results/temp/',paths(i).name],"MaxReward","testSucRate")
% testSuc(i) = testSucRate;
% end
% [bestTestSuc,I] = max(testSuc);
% source = ['results/temp/',paths(I).name];
% destination = [source([1:8 14:end-20]), '.mat'];
% copyfile(source, destination)
% disp([destination, ' Best testSuc: ', num2str(bestTestSuc)])
% end
end
end
%
% %%
% function Sch = getScheduleName(constraint)
% if constraint==0
% Sch = 'RL';
% else
% Sch = 'Cons';
% end
%
% end