-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathmain.m
More file actions
194 lines (156 loc) · 7.21 KB
/
main.m
File metadata and controls
194 lines (156 loc) · 7.21 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
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
%%
% This is the main file to run the tests. If you don't want to run all
% the different algorithms, set the options you want in this section, run
% it, then run the section corresponding to the algorithm you care about.
% Each section of this code will call singlecomparison() to get plot data
% for the algorithm of choice. Refer to singlecomparison.m for more
% details.
n = 10;
lowerlimit = 1000;
upperlimit = 11000;
step = 10000;
exponents = '{2.1,2.1,2.1,2.1,2.1,2.1,2.1,2.1,2.1,2.1}';
seed = 42;
rng(seed);
numberofruns = 10;
sizes = lowerlimit:step:upperlimit;
addpath('third_party')
%% -------------------- FastICA with pow3 nonlinearity --------------------
% Set up empy arrays to hold the data
amaridataundampened = zeros(numberofruns,length(sizes));
frobeniusdataundampened = zeros(numberofruns,length(sizes));
amaridatadampened = zeros(numberofruns,length(sizes));
frobeniusdatadampened = zeros(numberofruns,length(sizes));
% Run the comparison algorithm to get the data, storing it
% each time
for i = 1:numberofruns
% First, without dampening
disp(['Starting run ' int2str(i)]);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', ...
'exponents', exponents, ...
'seed', seed, ...
'run', i);
amaridataundampened(i,:) = result(1,:);
frobeniusdataundampened(i,:) = result(2,:);
% With dampening
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'damp', 'true', ...
'regenerate_samples', 'false', ...
'exponents', exponents, ...
'seed', seed, ...
'run', i);
amaridatadampened(i,:) = result(1,:);
frobeniusdatadampened(i,:) = result(2,:);
end
myicaplot(amaridataundampened, frobeniusdataundampened, ...
amaridatadampened, frobeniusdatadampened, sizes, 'FastICA - pow3')
%% -------------------- FastICA with tanh nonlinearity --------------------
amaridataundampened = zeros(numberofruns,length(sizes));
frobeniusdataundampened = zeros(numberofruns,length(sizes));
amaridatadampened = zeros(numberofruns,length(sizes));
frobeniusdatadampened = zeros(numberofruns,length(sizes));
for i = 1:numberofruns
disp(['Starting run ' int2str(i)]);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'algorithm', 'tanh');
amaridataundampened(i,:) = result(1,:);
frobeniusdataundampened(i,:) = result(2,:);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'damp', 'true', 'algorithm', 'tanh', ...
'regenerate_samples', 'false');
amaridatadampened(i,:) = result(1,:);
frobeniusdatadampened(i,:) = result(2,:);
end
myicaplot(amaridataundampened, frobeniusdataundampened, ...
amaridatadampened, frobeniusdatadampened, sizes, 'FastICA - tanh')
%% -------------------------- Fourier PCA ---------------------------------
amaridataundampened = zeros(numberofruns,length(sizes));
frobeniusdataundampened = zeros(numberofruns,length(sizes));
amaridatadampened = zeros(numberofruns,length(sizes));
frobeniusdatadampened = zeros(numberofruns,length(sizes));
for i = 1:numberofruns
% With dampening
disp(['Starting run ' int2str(i)]);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'algorithm', 'fpca');
amaridataundampened(i,:) = result(1,:);
frobeniusdataundampened(i,:) = result(2,:);
% Without dampening
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'dampen', 'true', 'algorithm', 'fpca');
amaridatadampened(i,:) = result(1,:);
frobeniusdatadampened(i,:) = result(2,:);
end
myicaplot(amaridataundampened, frobeniusdataundampened, ...
amaridatadampened, frobeniusdatadampened, sizes, 'Fourier PCA')
%% -------------------------------- SOBI ----------------------------------
amaridataundampened = zeros(numberofruns,length(sizes));
frobeniusdataundampened = zeros(numberofruns,length(sizes));
amaridatadampened = zeros(numberofruns,length(sizes));
frobeniusdatadampened = zeros(numberofruns,length(sizes));
for i = 1:numberofruns
disp(['Starting run ' int2str(i)]);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'algorithm', 'sobi');
amaridataundampened(i,:) = result(1,:);
frobeniusdataundampened(i,:) = result(2,:);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'dampen', 'true', 'algorithm', 'sobi');
amaridatadampened(i,:) = result(1,:);
frobeniusdatadampened(i,:) = result(2,:);
end
myicaplot(amaridataundampened, frobeniusdataundampened, ...
amaridatadampened, frobeniusdatadampened, sizes, 'SOBI')
%% -------------------------------- JADE ----------------------------------
amaridataundampened = zeros(numberofruns,length(sizes));
frobeniusdataundampened = zeros(numberofruns,length(sizes));
for i = 1:numberofruns
disp(['Starting run ' int2str(i)]);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'algorithm', 'jade');
amaridataundampened(i,:) = result(1,:);
frobeniusdataundampened(i,:) = result(2,:);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'dampen', 'true', 'algorithm', 'jade', ...
'regenerate_samples', 'false');
amaridatadampened(i,:) = result(1,:);
frobeniusdatadampened(i,:) = result(2,:);
end
myicaplot(amaridataundampened, frobeniusdataundampened, ...
amaridatadampened, frobeniusdatadampened, sizes, 'JADE')
%% -------------------------------- SIMPLE --------------------------------
amaridataundampened = zeros(numberofruns,length(sizes));
frobeniusdataundampened = zeros(numberofruns,length(sizes));
for i = 1:numberofruns
disp(['Starting run ' int2str(i)]);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'algorithm', 'simple');
amaridataundampened(i,:) = result(1,:);
frobeniusdataundampened(i,:) = result(2,:);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'dampen', 'true', 'algorithm', 'simple', ...
'regenerate_samples', 'false');
amaridatadampened(i,:) = result(1,:);
frobeniusdatadampened(i,:) = result(2,:);
end
myicaplot(amaridataundampened, frobeniusdataundampened, ...
amaridatadampened, frobeniusdatadampened, sizes, 'Simple')
%% -------------------------------- SANITY CHECK --------------------------
% Calls singlecomparison with the directive to use uniform points from the
% hypercube
for i = 1:numberofruns
disp(['Starting run ' int2str(i)]);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'dampen', 'false', 'algorithm', 'sobi', ...
'sanity', 'true');
amaridataundampened(i,:) = result(1,:);
frobeniusdataundampened(i,:) = result(2,:);
result = singlecomparison(n, lowerlimit, upperlimit, step, ...
'verbose', 'true', 'dampen', 'true', 'algorithm', 'sobi', ...
'sanity', 'true');
amaridatadampened(i,:) = result(1,:);
frobeniusdatadampened(i,:) = result(2,:);
end
myicaplot(amaridataundampened, frobeniusdataundampened, ...
amaridatadampened, frobeniusdatadampened, sizes, 'Sanity Check')