Chunk / ImageR

Blame view

test/test_model.py 4.49 KB
Edit Raw Normal View History
be12257b   Chunk   data-feat-model f... 
1
2
 
__author__ = 'chunk'
9371f8fa   Chunk   SVM param engenee... 
3
 
from sklearn import cross_validation
f4fb4381   Chunk   staged. 
4
 
from pyspark.mllib.regression import LabeledPoint
2bf33465   Chunk   staged. 
5
 
from ..common import *
84648488   Chunk   reverted. 
6
 
from ..mdata import ILSVRC, ILSVRC_S
61e78eb3   Chunk   staged. 
7
 
from ..mmodel.svm import SVM
84648488   Chunk   reverted. 
8
2bf33465   Chunk   staged. 
9
 
import gzip
5a469df5   Chunk   staged. 
10
11
12
 
import cPickle
84648488   Chunk   reverted. 
13
 
timer = Timer()
5a469df5   Chunk   staged. 
14
15
 
package_dir = os.path.dirname(os.path.abspath(__file__))
be12257b   Chunk   data-feat-model f... 
16
d2603183   Chunk   staged. 
17
 
def test_SVM_ILSVRC():
84648488   Chunk   reverted. 
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
 
    timer.mark()
    dil = ILSVRC.DataILSVRC(base_dir='/data/hadoop/ImageNet/ILSVRC/ILSVRC2013_DET_val', category='Train_5000_0.05_orig')
    X, Y = dil.load_data(mode='local')  #
    # X, Y = dil.load_data(mode='hbase') #
    # X, Y = dil.load_data(mode='spark') #
    X_train, X_test, Y_train, Y_test = cross_validation.train_test_split(X, Y, test_size=0.4, random_state=0)
    print np.array(Y).shape, np.array(X).shape
    print np.array(X_train).shape, np.array(Y_train).shape
    print np.array(X_test).shape, np.array(Y_test).shape

    timer.report()

    timer.mark()
    msvm = SVM.ModelSVM(toolset='sklearn')  # 4.884247s 0.777853030816
    # msvm = SVM.ModelSVM(toolset='opencv') #
    # msvm = SVM.ModelSVM(toolset='libsvm') #
    # msvm = SVM.ModelSVM(toolset='spark')
    msvm.train(X_train, Y_train)
    timer.report()

    timer.mark()
    print msvm.test(X_test, Y_test)  #
    timer.report()  #

    # timer.mark()
    # print 'or like this:'
    # scores = cross_validation.cross_val_score(msvm.model, X, Y)
    # print scores
    # timer.report()
d0be60e7   Chunk   jpeg update. 
48
49
 

def test_SVM_ILSVRC_HBASE():
84648488   Chunk   reverted. 
50
 
    timer.mark()
d0be60e7   Chunk   jpeg update. 
51
02528074   Chunk   staged. 
52
53
 
    # dil = ILSVRC.DataILSVRC(base_dir='ILSVRC2013_DET_val', category='Train_3')
    # X, Y = dil.load_data(mode='hbase') # pass
84648488   Chunk   reverted. 
54
6d6d75b8   Chunk   spider LOG system. 
55
 
    dils = ILSVRC_S.DataILSVRC_S(base='ILSVRC2013_DET_val', category='Train_1000')
02528074   Chunk   staged. 
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
 
    X, Y = dils.load_data(mode='hbase')  # pass

    # dil = ILSVRC_S.DataILSVRC_S(base='/data/hadoop/ImageNet/ILSVRC/ILSVRC2013_DET_val/', category='Train_5000_0.1_orig')
    # X1, Y1 = dil.load_data(mode='local')

    X_train, X_test, Y_train, Y_test = cross_validation.train_test_split(X, Y, test_size=0.4, random_state=0)
    print Y, np.sum(np.array(Y) == 0), np.sum(np.array(Y) == 1)
    print np.array(Y).shape, np.array(X).shape
    print np.array(X_train).shape, np.array(Y_train).shape
    print np.array(X_test).shape, np.array(Y_test).shape

    timer.report()

    timer.mark()
    msvm = SVM.ModelSVM(toolset='sklearn')  # 4.884247s 0.777853030816
    # msvm = SVM.ModelSVM(toolset='opencv') #
    # msvm = SVM.ModelSVM(toolset='libsvm') #
    # msvm = SVM.ModelSVM(toolset='spark',sc=dils.sparker)
    msvm.train(X_train, Y_train)
    timer.report()

    timer.mark()
    print msvm.test(X_test, Y_test)  #
    timer.report()  #

    # timer.mark()
    # print msvm.test(X1, Y1)  #(0.048868415782094936, 0.4924709948160948, 0.74568774878372401)
    # timer.report()  #

    # timer.mark()
4f36b116   Chunk   staged. 
86
 
    # print 'or like this:'
02528074   Chunk   staged. 
87
 
    # scores = cross_validation.cross_val_score(msvm.model, X, Y)
9371f8fa   Chunk   SVM param engenee... 
88
 
    # print scores
51708346   Chunk   final experiments... 
89
90
 
    # timer.report()
2bd3da3e   Chunk   staged. 
91
84648488   Chunk   reverted. 
92
 
def test_SVM_ILSVRC_TEST():
d2603183   Chunk   staged. 
93
 
    timer.mark()
02528074   Chunk   staged. 
94
9371f8fa   Chunk   SVM param engenee... 
95
96
97
 
    dil = ILSVRC_S.DataILSVRC_S(base='/data/hadoop/ImageNet/ILSVRC/ILSVRC2013_DET_val/', category='Train_5000_0.1_orig')
    X1, Y1 = dil.load_data(mode='local')
    timer.report()
d0be60e7   Chunk   jpeg update. 
98
99
100
 

    timer.mark()
    msvm = SVM.ModelSVM(toolset='sklearn')  # 4.884247s 0.777853030816
4f36b116   Chunk   staged. 
101
 
    timer.report()
51708346   Chunk   final experiments... 
102
9371f8fa   Chunk   SVM param engenee... 
103
 
    timer.mark()
4f36b116   Chunk   staged. 
104
 
    print msvm.test(X1, Y1)  # (0.048868415782094936, 0.4924709948160948, 0.74568774878372401)
9371f8fa   Chunk   SVM param engenee... 
105
 
    timer.report()  #
d0be60e7   Chunk   jpeg update. 
106
107
108
 
    # timer.mark()
    # print 'or like this:'
    # scores = cross_validation.cross_val_score(msvm.model, X, Y)
9371f8fa   Chunk   SVM param engenee... 
109
 
    # print scores
d0be60e7   Chunk   jpeg update. 
110
 
    # timer.report()
6d6d75b8   Chunk   spider LOG system. 
111
51708346   Chunk   final experiments... 
112
113
114
115
 

def test_SVM_ILSVRC_SPARK():
    timer.mark()
    dils = ILSVRC_S.DataILSVRC_S(base='ILSVRC2013_DET_val', category='Train_5000')
e3e7e73a   Chunk   spider standalone... 
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
 
    # rdd_dataset = dils.load_data(mode='spark')  # pass
    X, Y = dils.load_data(mode='hbase')  # pass
    rdd_dataset = dils.sparker.sc.parallelize(zip(Y, X), 30).map(lambda x: LabeledPoint(x[0], x[1]))
    timer.report()

    timer.mark()
    # msvm = SVM.ModelSVM(toolset='sklearn')  #
    # msvm = SVM.ModelSVM(toolset='opencv') #
    # msvm = SVM.ModelSVM(toolset='libsvm') #
    msvm = SVM.ModelSVM(toolset='spark', sc=dils.sparker)
    msvm.train(rdd_dataset)
    timer.report()

    dataset = rdd_dataset.collect()
    length = len(dataset)

    X_test, Y_test = [dataset[i].features for i in range(length)], [dataset[i].label for i in range(length)]

    timer.mark()
2bd3da3e   Chunk   staged. 
135
 
    print msvm.test(dils.sparker.sc.parallelize(X_test), Y_test)  #
02528074   Chunk   staged. 
136
137
138
139
140
141
142
143
144
 
    timer.report()  #


def test_SVM_ILSVRC_S():
    test_SVM_ILSVRC_HBASE()
    # test_SVM_ILSVRC_SPARK()


if __name__ == '__main__':
f4fb4381   Chunk   staged. 
145
146
147
 
    # test_SVM_CV()
    test_SVM_ILSVRC()
    print 'helllo'
02528074   Chunk   staged. 












9371f8fa
 
 Chunk
 
SVM param engenee...
















02528074
 
 Chunk
 
staged.
















9371f8fa
 
 Chunk
 
SVM param engenee...
















02528074
 
 Chunk
 
staged.
















84648488
 
 Chunk
 
reverted.
















02528074
 
 Chunk
 
staged.
















51708346
 
 Chunk
 
final experiments...
















02528074
 
 Chunk
 
staged.
















9371f8fa
 
 Chunk
 
SVM param engenee...
















84648488
 
 Chunk
 
reverted.
















be12257b
 
 Chunk
 
data-feat-model f...
















d0be60e7
 
 Chunk
 
jpeg update.
















be12257b
 
 Chunk
 
data-feat-model f...
















84648488
 
 Chunk
 
reverted.