-
Notifications
You must be signed in to change notification settings - Fork 36
/
RunLinearSVMRWS.m
130 lines (78 loc) · 4.65 KB
/
RunLinearSVMRWS.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
function RunLinearSVMRWS(DataSetStartIndex, DataSetEndIndex)
% first 2 values are '.' and '..' - UCR Archive 2018 version has 128 datasets
dir_struct = dir('/rigel/dsi/users/ikp2103/VLDBGRAIL/UCR2018/');
Datasets = {dir_struct(3:130).name};
% Sort Datasets
[Datasets, DSOrder] = sort(Datasets);
Results = zeros(length(Datasets),11);
addpath(genpath('LibLinear/matlab/.'));
distcomp.feature( 'LocalUseMpiexec', false )
%rng(ceil(DataSetStartIndex*100))
%pause(100*rand);
poolobj = gcp('nocreate');
delete(poolobj);
parpool(22);
rng('default')
for i = 1:length(Datasets)
if (i>=DataSetStartIndex && i<=DataSetEndIndex)
disp(['Dataset being processed: ', char(Datasets(i))]);
DS = LoadUCRdataset(char(Datasets(i)));
[Thebestcost2,Thebestacc2,Thebestiming2] = GridSearchLinearSVM2(-10,0.1,20,DS.TrainInstancesCount,DS.TrainClassLabels,Datasets,i);
ZRep = dlmread( strcat( 'RWSREPRESENTATIONS','/',char(Datasets(i)),'/','RWS_Supervised', '.Zrep') );
ZRepTrain = ZRep(1:DS.TrainInstancesCount,:);
ZRepTest = ZRep(DS.TrainInstancesCount+1:end,:);
ZRepTrain = sparse(ZRepTrain);
ZRepTest = sparse(ZRepTest);
tic;
cmd = ['-e 0.001 -s 2 -c ', num2str(2^Thebestcost2)];
model_precomputed = train(DS.TrainClassLabels, ZRepTrain, cmd);
ModelTrainingRuntime = toc;
tic;
[predict_label_P, accuracy_P, dec_values_P] = predict(DS.TestClassLabels, ZRepTest, model_precomputed);
PredictionRuntime = toc;
Results(i,1) = 0;
Results(i,2) = 0;
Results(i,3) = Thebestcost2;
%Results(i,4) = Thebestcost3;
Results(i,4) = 0;
Results(i,5) = 0;
Results(i,6) = Thebestacc2*0.01;
%Results(i,7) = Thebestacc3*0.01;
Results(i,7) = 0;
%Results(i,8) = Thebestiming1+Thebestiming2+Thebestiming3;
Results(i,8) = Thebestiming2;
Results(i,9) = accuracy_P(1)*0.01;
Results(i,10) = ModelTrainingRuntime;
Results(i,11) = PredictionRuntime;
dlmwrite( strcat('RunLinearSVMRWS/','RunLinearSVMRWS', '_Dataset_', num2str(i)) , Results, 'delimiter', '\t');
end
end
poolobj = gcp('nocreate');
delete(poolobj);
end
function [Thebestcost,Thebestacc,Thebestiming] = GridSearchLinearSVM2(GridStart,GridStep,GridEnd,TrainInstancesCount,TrainClassLabels,Datasets,DatasetsNumber)
% Tuning Parameters
log2cTmp = GridStart:GridStep:GridEnd;
bestacc = zeros(1,length(log2cTmp));
bestcost = zeros(1,length(log2cTmp));
besttiming = zeros(1,length(log2cTmp));
ZRep = dlmread( strcat( 'RWSREPRESENTATIONS','/',char(Datasets(DatasetsNumber)),'/','RWS_Supervised', '.Zrep') );
ZRepTrain = ZRep(1:TrainInstancesCount,:);
ZRepTrain = sparse(ZRepTrain);
% grid search
parfor log2cNEW = 1:length(log2cTmp)
log2cNEW
tic;
log2c = log2cTmp(log2cNEW);
cmd = ['-q -e 0.001 -s 2 -v ' num2str(10) ' -c ', num2str(2^log2c)];
cv = train(TrainClassLabels, ZRepTrain, cmd);
bestacc(log2cNEW) = cv;
bestcost(log2cNEW) = log2c;
besttiming(log2cNEW) = toc;
end
[Maxbestacc,~] = max(bestacc);
Posbestacc = find(bestacc==Maxbestacc,1,'last');
Thebestiming = sum(besttiming);
Thebestcost = bestcost(Posbestacc);
Thebestacc = Maxbestacc;
end