-
Notifications
You must be signed in to change notification settings - Fork 0
/
std_dyn.m
192 lines (187 loc) · 8.26 KB
/
std_dyn.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
% std_dyn.m
% Set up a binary matrix for how long each simulation lasted
T = cumsum(tau')';
tFinal = zeros(1,optimax);
[maxNodes,maxNodes] = size(totalNumNodes);
measures = 1:maxNodes;
if length(gcFrac) ~= maxNodes
measures100 = 50:50:maxNodes;
else
measures100 = measures;
end
numMeasures = zeros(optimax,maxNodes);
[nM100 nM100] = size(gcFrac);
% nM100 = length(measures100);
numMeasures100 = zeros(optimax,nM100);
for i = 1:optimax
tFinal(i) = max(totalNumNodes(i,:));
numMeasures(i,:) = measures < tFinal(i);
numMeasures100(i,:) = measures100 < tFinal(i);
end
fiftyPercent = sum(sum(numMeasures) > optimax/2);
fiftyPercent100 = sum(sum(numMeasures100) > optimax/2);
for iD = 1:maxNodes
median_totalNumNodes(iD) = median(nonzeros(totalNumNodes(:,iD)));
end
% Mean + std of statistics calculated every time step
% Total number of edges
% mean_totalNumEdges = sum(totalNumEdges)./sum(numMeasures);
for iD = 1:maxNodes
mean_totalNumEdges(iD) = median(nonzeros(totalNumEdges(:,iD)));
end
% std_totalNumEdges = sqrt(sum((totalNumEdges - numMeasures.*repmat(mean_totalNumEdges,optimax,1)).^2)./(sum(numMeasures) - 1));
% std_totalNumEdges(isinf(std_totalNumEdges)) = 0;
% std_totalNumEdges(isnan(std_totalNumEdges)) = 0;
% std_totalNumEdges(isinf(mean_totalNumEdges)) = [];
mean_totalNumEdges(isinf(mean_totalNumEdges)) = [];
% Fraction of random edges
% mean_fracRandomEdges = sum(fracRandomEdges)./sum(numMeasures);
fracRandomEdges(isnan(fracRandomEdges)) = 0;
for iD = 1:maxNodes
mean_fracRandomEdges(iD) = median(nonzeros(fracRandomEdges(:,iD)));
end
% std_fracRandomEdges = sqrt(sum((fracRandomEdges - numMeasures.*repmat(mean_fracRandomEdges,optimax,1)).^2)./(sum(numMeasures) - 1));
% std_fracRandomEdges(isinf(std_fracRandomEdges)) = 0;
% std_fracRandomEdges(isnan(std_fracRandomEdges)) = 0;
% std_fracRandomEdges(isinf(mean_fracRandomEdges)) = [];
mean_fracRandomEdges(isinf(mean_fracRandomEdges)) = [];
% Fraction of duplicate edges
% mean_fracDuplicateEdges = sum(fracDuplicateEdges)./sum(numMeasures);
fracDuplicateEdges(isnan(fracDuplicateEdges)) = 0;
for iD = 1:maxNodes
mean_fracDuplicateEdges(iD) = median(nonzeros(fracDuplicateEdges(:,iD)));
end
% std_fracDuplicateEdges = sqrt(sum((fracDuplicateEdges - numMeasures.*repmat(mean_fracDuplicateEdges,optimax,1)).^2)./(sum(numMeasures) - 1));
% std_fracDuplicateEdges(isinf(std_fracDuplicateEdges)) = 0;
% std_fracDuplicateEdges(isnan(std_fracDuplicateEdges)) = 0;
% std_fracDuplicateEdges(isinf(mean_fracDuplicateEdges)) = [];
mean_fracDuplicateEdges(isinf(mean_fracDuplicateEdges)) = [];
% Mean + std of statistics calculated every 100 time steps
% Global clustering coefficient
% mean_gcc = sum(gcc)./sum(numMeasures100);
for iD = 1:nM100
mean_gcc(iD) = median(nonzeros(gcc(:,iD)));
end
% std_gcc = sqrt(sum((gcc - numMeasures100.*repmat(mean_gcc,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_gcc(isinf(std_gcc)) = 0;
% std_gcc(isnan(std_gcc)) = 0;
% std_gcc(isinf(mean_gcc)) = [];
mean_gcc(isinf(mean_gcc)) = [];
% Diameter
% mean_diameter = sum(diameter)./sum(numMeasures100);
diameter(isnan(diameter)) = 0;
diameter(isinf(diameter)) = 0;
for iD = 1:nM100
mean_diameter(iD) = median(nonzeros(diameter(:,iD)));
end
% std_diameter = sqrt(sum((diameter - numMeasures100.*repmat(mean_diameter,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_diameter(isinf(std_diameter)) = 0;
% std_diameter(isnan(std_diameter)) = 0;
% std_diameter(isinf(mean_diameter)) = [];
mean_diameter(isinf(mean_diameter)) = [];
% Average distance
mean_avgCls = sum(avgCls)./sum(numMeasures100);
% std_avgCls = sqrt(sum((avgCls - numMeasures100.*repmat(mean_avgCls,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_avgCls(isinf(std_avgCls)) = 0;
% std_avgCls(isnan(std_avgCls)) = 0;
% std_avgCls(isinf(mean_avgCls)) = [];
mean_avgCls(isinf(mean_avgCls)) = [];
% Fraction of nodes in giant component
% mean_gcFrac = sum(gcFrac)./sum(numMeasures100);
for iD = 1:nM100
mean_gcFrac(iD) = median(nonzeros(gcFrac(:,iD)));
end
% std_gcFrac = sqrt(sum((gcFrac - numMeasures100.*repmat(mean_gcFrac,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_gcFrac(isinf(std_gcFrac)) = 0;
% std_gcFrac(isnan(std_gcFrac)) = 0;
% std_gcFrac(isinf(mean_gcFrac)) = [];
mean_gcFrac(isinf(mean_gcFrac)) = [];
% Second-largest eigenvalue
% mean_E2 = sum(E2)./sum(numMeasures100);
E2(isnan(E2)) = 0;
E2(isinf(E2)) = 0;
for iD = 1:nM100
mean_E2(iD) = median(nonzeros(E2(:,iD)));
end
% std_E2 = sqrt(sum((E2 - numMeasures100.*repmat(mean_E2,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_E2(isinf(std_E2)) = 0;
% std_E2(isnan(std_E2)) = 0;
% std_E2(isinf(mean_E2)) = [];
mean_E2(isinf(mean_E2)) = [];
if exist('frag10','var')
% 10% fragmentation tolerance
mean_frag10 = sum(frag10)./sum(numMeasures100);
% std_frag10 = sqrt(sum((frag10 - numMeasures100.*repmat(mean_frag10,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_frag10(isinf(std_frag10)) = 0;
% std_frag10(isnan(std_frag10)) = 0;
% std_frag10(isinf(mean_frag10)) = [];
mean_frag10(isinf(mean_frag10)) = [];
end
if exist('Q','var')
% Fraction of nodes in giant component ignoring microcomponents
mean_gcFrac_le10 = sum(gcFrac_le10)./sum(numMeasures100);
% std_gcFrac_le10 = sqrt(sum((gcFrac_le10 - numMeasures100.*repmat(mean_gcFrac_le10,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_gcFrac_le10(isinf(std_gcFrac_le10)) = 0;
% std_gcFrac_le10(isnan(std_gcFrac_le10)) = 0;
% std_gcFrac_le10(isinf(mean_gcFrac_le10)) = 1;
mean_gcFrac_le10(isinf(mean_gcFrac_le10)) = 1;
% Fraction of nodes in giant component ignoring orphan nodes
% mean_gcFrac_nz = sum(gcFrac_nz)./sum(numMeasures100);
gcFrac_nz(isnan(gcFrac_nz)) = 0;
gcFrac_nz(isinf(gcFrac_nz)) = 0;
for iD = 1:nM100
mean_gcFrac_nz(iD) = median(nonzeros(gcFrac_nz(:,iD)));
end
% std_gcFrac_nz = sqrt(sum((gcFrac_nz - numMeasures100.*repmat(mean_gcFrac_nz,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_gcFrac_nz(isinf(std_gcFrac_nz)) = 0;
% std_gcFrac_nz(isnan(std_gcFrac_nz)) = 0;
% std_gcFrac_nz(isinf(mean_gcFrac_nz)) = [];
mean_gcFrac_nz(isinf(mean_gcFrac_nz)) = [];
% Fraction of HBLC (high betweenness low connectivity) nodes
mean_HBLC = sum(HBLC)./sum(numMeasures100);
% std_HBLC = sqrt(sum((HBLC - numMeasures100.*repmat(mean_HBLC,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_HBLC(isinf(std_HBLC)) = 0;
% std_HBLC(isnan(std_HBLC)) = 0;
% std_HBLC(isinf(mean_HBLC)) = [];
mean_HBLC(isinf(mean_HBLC)) = [];
% Modularity
% mean_Q = sum(Q)./sum(numMeasures100);
Q(isnan(Q)) = 0;
Q(isinf(Q)) = 0;
for iQ = 1:nM100
mean_Q(iQ) = median(nonzeros(Q(:,iQ)));
end
% std_Q = sqrt(sum((Q - numMeasures100.*repmat(mean_Q,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_Q(isinf(std_Q)) = 0;
% std_Q(isnan(std_Q)) = 0;
% std_Q(isinf(mean_Q)) = [];
mean_Q(isinf(mean_Q)) = [];
end
if exist('gcFrac_1','var')
% Fraction of nodes in giant component ignoring orphan nodes
% mean_gcFrac_1 = sum(gcFrac_1)./sum(numMeasures100);
gcFrac_1(isnan(gcFrac_1)) = 0;
gcFrac_1(isinf(gcFrac_1)) = 0;
for iD = 1:nM100
mean_gcFrac_1(iD) = median(nonzeros(gcFrac_1(:,iD)));
end
% std_gcFrac_1 = sqrt(sum((gcFrac_1 - numMeasures100.*repmat(mean_gcFrac_1,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_gcFrac_1(isinf(std_gcFrac_1)) = 0;
% std_gcFrac_1(isnan(std_gcFrac_1)) = 0;
% std_gcFrac_1(isinf(mean_gcFrac_1)) = [];
mean_gcFrac_1(isinf(mean_gcFrac_1)) = [];
% Fraction of nodes in giant component ignoring orphan nodes
mean_gcFrac_5 = sum(gcFrac_5)./sum(numMeasures100);
% std_gcFrac_5 = sqrt(sum((gcFrac_5 - numMeasures100.*repmat(mean_gcFrac_5,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_gcFrac_5(isinf(std_gcFrac_5)) = 0;
% std_gcFrac_5(isnan(std_gcFrac_5)) = 0;
% std_gcFrac_5(isinf(mean_gcFrac_5)) = [];
mean_gcFrac_5(isinf(mean_gcFrac_5)) = [];
% Fraction of nodes in giant component ignoring orphan nodes
mean_gcFrac_10 = sum(gcFrac_10)./sum(numMeasures100);
% std_gcFrac_10 = sqrt(sum((gcFrac_10 - numMeasures100.*repmat(mean_gcFrac_10,optimax,1)).^2)./(sum(numMeasures100) - 1));
% std_gcFrac_10(isinf(std_gcFrac_10)) = 0;
% std_gcFrac_10(isnan(std_gcFrac_10)) = 0;
% std_gcFrac_10(isinf(mean_gcFrac_10)) = [];
mean_gcFrac_10(isinf(mean_gcFrac_10)) = [];
end