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
|
'''
Runs unit tests for dimensionality reduction algorithms.
To run the unit tests, type the following from the system command line:
# python -m spectral.tests.dimensionality
'''
from __future__ import absolute_import, division, print_function, unicode_literals
import numpy as np
import spectral as spy
from spectral.tests.spytest import SpyTest, test_method
class DimensionalityTest(SpyTest):
'''Tests various math functions.'''
def setup(self):
self.data = spy.open_image('92AV3C.lan').load()
def test_mnf_all_equals_data(self):
'''Test that MNF transform with all components equals original data.'''
data = self.data
signal = spy.calc_stats(data)
noise = spy.noise_from_diffs(data[117: 137, 85: 122, :])
mnfr = spy.mnf(signal, noise)
denoised = mnfr.denoise(data, num=data.shape[-1])
assert(np.allclose(denoised, data))
def test_ppi(self):
'''Tests that ppi function runs'''
data = self.data
p = spy.ppi(data, 4)
def test_ppi_threshold(self):
'''Tests that ppi function runs with threshold arg'''
data = self.data
p = spy.ppi(data, 4, 10)
def test_ppi_continues(self):
'''Tests that running ppi with initial indices works as expected.'''
data = self.data
s = np.random.get_state()
p = spy.ppi(data, 4)
np.random.set_state(s)
p2 = spy.ppi(data, 2)
p2 = spy.ppi(data, 2, start=p2)
assert(np.all(p == p2))
def test_ppi_centered(self):
'''Tests that ppi with mean-subtracted data works as expected.'''
data = self.data
s = np.random.get_state()
p = spy.ppi(data, 4)
np.random.set_state(s)
data_centered = data - spy.calc_stats(data).mean
p2 = spy.ppi(data_centered, 4)
assert(np.all(p == p2))
def test_smacc_minimal(self):
'''Tests smacc correctness on minimal example.'''
H = np.array([
[1.0, 0.0, 0.0],
[0.0, 1.0, 0.0],
[1.0, 1.0, 0.0],
[0.0, 1.0, 1.0]
])
S, F, R = spy.smacc(H)
assert(np.allclose(np.matmul(F, S) + R, H))
assert(np.min(F) == 0.0)
expected_S = np.array([
# First two longer ones.
[1., 1., 0.],
[0., 1., 1.],
# First of the two shorted ones. Other can be expressed other 3.
[1., 0., 0.],
])
assert(np.array_equal(S, expected_S))
def test_smacc_runs(self):
'''Tests that smacc runs without additional arguments.'''
# Without scaling numeric errors accumulate.
scaled_data = self.data / 10000
S, F, R = spy.smacc(scaled_data)
data_shape = scaled_data.shape
H = scaled_data.reshape(data_shape[0] * data_shape[1], data_shape[2])
assert(np.allclose(np.matmul(F, S) + R, H))
assert(np.min(F) == 0.0)
assert(len(S.shape) == 2 and S.shape[0] == 9 and S.shape[1] == 220)
def test_smacc_min_endmembers(self):
'''Tests that smacc runs with min_endmember argument.'''
# Without scaling numeric errors accumulate.
scaled_data = self.data / 10000
S, F, R = spy.smacc(scaled_data, 10)
data_shape = scaled_data.shape
H = scaled_data.reshape(data_shape[0] * data_shape[1], data_shape[2])
assert(np.allclose(np.matmul(F, S) + R, H))
assert(np.min(F) == 0.0)
assert(len(S.shape) == 2 and S.shape[0] == 10 and S.shape[1] == 220)
def test_smacc_max_residual_norm(self):
'''Tests that smacc runs with max_residual_norm argument.'''
# Without scaling numeric errors accumulate.
scaled_data = self.data / 10000
S, F, R = spy.smacc(scaled_data, 9, 0.8)
data_shape = scaled_data.shape
H = scaled_data.reshape(data_shape[0] * data_shape[1], data_shape[2])
assert(np.allclose(np.matmul(F, S) + R, H))
assert(np.min(F) == 0.0)
residual_norms = np.einsum('ij,ij->i', R, R)
assert(np.max(residual_norms) <= 0.8)
def test_pca_runs(self):
'''Should be able to compute PCs and transform data.'''
data = self.data
xdata = spy.principal_components(data).transform(data)
def test_pca_runs_from_stats(self):
'''Should be able to pass image stats to PCA function.'''
data = self.data
stats = spy.calc_stats(data)
xdata = spy.principal_components(stats).transform(data)
def test_orthogonalize(self):
'''Can correctly create an orthogonal basis from vectors.'''
x = np.linspace(0, np.pi, 1001)
# Create sin and cos vectors of unit length
sin_h = np.sin(x)
sin_h /= np.linalg.norm(sin_h)
cos_h = np.cos(x)
cos_h /= np.linalg.norm(cos_h)
X = np.array([50 * sin_h, 75 * cos_h])
Y = spy.orthogonalize(X)
assert(np.allclose(Y.dot(Y.T), np.array([[1, 0], [0, 1]])))
assert(np.allclose(X.dot(Y.T), np.array([[50, 0], [0, 75]])))
def test_orthogonalize_subset(self):
'''Can correctly create an orthogonal basis from vector subset.'''
x = np.linspace(0, np.pi, 1001)
# Create sin and cos vectors of unit length
sin_h = np.sin(x)
sin_h /= np.linalg.norm(sin_h)
cos_h = np.cos(x)
cos_h /= np.linalg.norm(cos_h)
# First vector in X will already be a unit vector
X = np.array([sin_h, 75 * cos_h])
Y = spy.orthogonalize(X, start=1)
assert(np.allclose(Y.dot(Y.T), np.array([[1, 0], [0, 1]])))
assert(np.allclose(X.dot(Y.T), np.array([[1, 0], [0, 75]])))
def run():
print('\n' + '-' * 72)
print('Running dimensionality tests.')
print('-' * 72)
test = DimensionalityTest()
test.run()
if __name__ == '__main__':
from spectral.tests.run import parse_args, reset_stats, print_summary
parse_args()
reset_stats()
run()
print_summary()
|
