eigenfish/test.py at master · pens/eigenfish · GitHub

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
import numpy
import os
import shutil
import unittest
from eigenfish import *
from process.process import *
from classify.classify import *
from util import *


#TODO fix occasional ARPACK errors
class EmptyProcessor(Processor):
    def process(self, img_mat, shape):
        return img_mat

class EmptyClassifier(Classifier):
    def __init__(self):
        pass

    def train(self, data, labels):
        pass

    def classify(self, data):
        return ["none"]

    def cross_validate(self, data, labels):
        return 0.0

    def save(self, filename):
        pass

    def load(self, filename):
        pass

class TestEigenfish(unittest.TestCase):
    def setUp(self):
        self.shape = (10, 10)
        self.mat = numpy.hstack((numpy.ones((100, 10), 'F'),
                                 numpy.zeros((100, 10), 'F')))
        self.labels = (["ones" for i in range(10)] +
                       ["zeroes" for i in range(10)])
        self.test_mat = numpy.hstack((numpy.zeros((100, 3), 'F'),
                                      numpy.ones((100, 3), 'F')))
        self.test_labels = (["zeroes" for i in range(3)] +
                            ["ones" for i in range(3)])
        self.ef = Eigenfish(self.shape)
        os.mkdir("test/")

    def tearDown(self):
        shutil.rmtree("test/")

    def test_train_classify(self):
        self.ef.train(self.mat, self.labels)
        res = self.ef.classify(self.test_mat)
        for i in range(len(self.test_labels)):
            self.assertEqual(self.test_labels[i], res[i])

    def test_cross_validate(self):
        self.ef.train(self.mat, self.labels)
        pct_correct = self.ef.cross_validate(self.test_mat, self.test_labels)
        self.assertEqual(pct_correct, 1.0)

    def test_save_load(self):
        self.ef.train(self.mat, self.labels)
        res1 = self.ef.classify(self.test_mat)

        self.ef.save('test/temp.pkl')
        self.ef = Eigenfish(self.shape)
        self.ef.load('test/temp.pkl')

        res2 = self.ef.classify(self.test_mat)
        for i in range(len(res1)):
            self.assertEqual(res1[i], res2[i],
                             'Results not equal after save/load')
        ef = Eigenfish(self.shape, 'test/temp.pkl')
        res3 = ef.classify(self.test_mat)
        for i in range(len(res1)):
            self.assertEqual(res1[i], res3[i],
                             'Constructor training file load failed')

    def test_custom_modules(self):
        self.ef = Eigenfish(self.shape, None, EmptyProcessor, EmptyClassifier)
        self.ef.train(self.mat, self.labels)
        res = self.ef.classify(self.test_mat)
        self.assertEqual(["none"], res)

    def test_load_img_mat(self):
        filenames = ["test_data/%d.jpg" % i for i in range(4)]
        img_mat, shape = load_img_mat(filenames)
        self.assertEqual(img_mat.shape, (shape[0] * shape[1], 4))


class TestClassify(unittest.TestCase):
    def setUp(self):
        self.mat = numpy.hstack((numpy.ones((100, 10), 'F'),
                                 numpy.zeros((100, 10), 'F')))
        self.labels = (["ones" for i in range(10)] +
                       ["zeroes" for i in range(10)])
        self.test_mat = numpy.hstack((numpy.zeros((100, 3), 'F'),
                                      numpy.ones((100, 3), 'F')))
        self.test_labels = (["zeroes" for i in range(3)] +
                            ["ones" for i in range(3)])
        self.classifier = Classifier()
        os.mkdir("test/")

    def tearDown(self):
        shutil.rmtree("test/")

    def test_train_classify(self):
        self.classifier.train(self.mat, self.labels)
        res = self.classifier.classify(self.test_mat)
        for i in range(len(self.test_labels)):
            self.assertEqual(self.test_labels[i], res[i])

    def test_cross_validate(self):
        self.classifier.train(self.mat, self.labels)
        pct_correct = (
            self.classifier.cross_validate(self.test_mat, self.test_labels))
        self.assertEqual(pct_correct, 1.0)

    def test_save_load(self):
        self.classifier.train(self.mat, self.labels)
        res1 = self.classifier.classify(self.test_mat)

        self.classifier.save('test/temp.pkl')
        self.classifier = Classifier()
        self.classifier.load('test/temp.pkl')

        res2 = self.classifier.classify(self.test_mat)
        for i in range(len(res1)):
            self.assertEqual(res1[i], res2[i])


class TestProcess(unittest.TestCase):
    def setUp(self):
        self.shape = (10, 10)
        self.mat = numpy.random.random_sample((100, 10))

    def test_process(self):
        processor = Processor()
        proc_mat = processor.process(self.mat, self.shape)

    def test_rpca(self):
        l, s = rpca(self.mat)
        mat2 = l + s
        for i in range(self.mat.size):
            self.assertAlmostEqual(self.mat.flat[i], mat2.flat[i], 2)

    def test_fft2_series(self):
        fft = fft2_series(self.mat, self.shape)

if __name__ == "__main__":
    unittest.main()