-
Notifications
You must be signed in to change notification settings - Fork 0
/
test_script.py
345 lines (274 loc) · 11.9 KB
/
test_script.py
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
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
import numpy as np
from numpy.testing import *
import pandas as pd
import pandas.util.testing as pdt
import phase as ph
import nose
class TestCircle(object):
def setup(self):
self.circ = ph.circle(8, 8, 4, 64)
def test_center_val(self):
assert_equal(self.circ[8,8], 1.)
def test_edge_vals(self):
assert_equal(self.circ[8,4], .5)
assert_equal(self.circ[8,12], .5)
assert_equal(self.circ[4,8], .5)
assert_equal(self.circ[12,8], .5)
class TestCrop(object):
def setup(self):
self.arr = np.random.rand(16, 16)
self.x0, self.y0 = 6, 6
self.s = 4
def test_aligned_even(self):
assert_array_equal(self.arr[4:8,4:8],
ph.crop(self.arr, self.x0, self.y0, self.s))
def test_aligned_odd(self):
assert_array_equal(self.arr[4:9,4:9],
ph.crop(self.arr, self.x0, self.y0, self.s+1))
def test_nonaligned_even(self):
assert_array_equal(self.arr[4:8,4:8],
ph.crop(self.arr, self.x0+.25, self.y0+.25, self.s))
assert_array_equal(self.arr[5:9,4:8],
ph.crop(self.arr, self.x0+.25, self.y0+.75, self.s))
def test_nonaligned_odd(self):
assert_array_equal(self.arr[4:9,4:9],
ph.crop(self.arr, self.x0+.25, self.y0+.25, self.s+1))
assert_array_equal(self.arr[5:10,4:9],
ph.crop(self.arr, self.x0+.25, self.y0+.75, self.s+1))
class TestCSF(object):
def setup(self):
self.y, self.x = np.indices((16, 16))
def test_peak_scalar(self):
assert_equal(ph.CSF(0, 0, 4), 1)
def test_peak_array(self):
assert_equal(ph.CSF(self.x-8, self.y-8, 4)[8, 8], 1)
def test_minima_scalar(self):
assert_almost_equal(ph.CSF(0, 4, 4), 0)
assert_almost_equal(ph.CSF(0, -4, 4), 0)
assert_almost_equal(ph.CSF(4, 0, 4), 0)
assert_almost_equal(ph.CSF(-4, 0, 4), 0)
def test_minima_array(self):
assert_almost_equal(ph.CSF(self.x-8, self.y-8, 4)[8, 4], 0)
assert_almost_equal(ph.CSF(self.x-8, self.y-8, 4)[8, 12], 0)
assert_almost_equal(ph.CSF(self.x-8, self.y-8, 4)[4, 8], 0)
assert_almost_equal(ph.CSF(self.x-8, self.y-8, 4)[12, 8], 0)
class TestTotalPower(object):
def setup(self):
N = 512
self.bg = 1250.
amp = 32000.
s = 10.
y, x = np.indices((N,N))
dx, dy = 5.27, -3.78
r2 = (x - (N / 2 - dx)) ** 2 + (y - (N / 2 - dy)) ** 2
gauss = amp * np.exp(-r2 / 2 / s ** 2)
self.totp = gauss.sum()
img = self.bg + gauss
self.result = ph.total_power(img, N/2 - dx, N/2 - dy,
r1=100, r2=200, r3=300, x=x, y=y)
def test_background(self):
assert_almost_equal(self.result[3], self.bg)
def test_total_power(self):
assert_almost_equal(self.result[0], self.totp)
class TestLocatePeak(object):
def setup(self):
Nx, Ny = 32, 64
self.amp = 32000.
s = 10.
y, x = np.indices((Ny, Nx))
self.res = 16
max_delta = 3.
self.x0 = Nx/2 + (np.random.rand() - .5) * max_delta
self.y0 = Ny/2 + (np.random.rand() - .5) * max_delta
r_sq = (x - self.x0) ** 2 + (y -self.y0) ** 2
self.img = self.amp * np.exp(-r_sq / 2 / s ** 2)
self.result = ph.locate_peak(self.img, self.res)
def test_x(self):
assert_array_less(np.abs(self.x0 - self.result[0]), 1./self.res)
def test_y(self):
assert_array_less(np.abs(self.y0 - self.result[1]), 1./self.res)
def test_amp(self):
assert_approx_equal(self.result[2], self.amp, significant=3)
class TestAnalyzePeaks(object):
def setup(self):
N = 512
n = 3
window = 32
self.res = 16
r1 = 100
r2 = 200
r3 = 300
amp0 = 20000.
bg0 = 1200.
s0 = 10.
max_delta_pos = 30.
max_delta_amp = 10000.
max_delta_bg = 500.
max_delta_s = 5.
y, x = np.indices((N, N))
self.imgs = np.ones((n, N, N), dtype=np.float64)
self.reference_df = pd.DataFrame(index=range(n),
columns=['x0', 'y0', 'amp', 'bg',
'tot power', 'norm amp'],
dtype=np.float64)
for i in range(n):
x0 = N/2 + (np.random.rand() - .5) * max_delta_pos
y0 = N/2 + (np.random.rand() - .5) * max_delta_pos
r_sq = (x - x0) ** 2 + (y -y0) ** 2
amp = amp0 + (np.random.rand() - .5) * max_delta_amp
s = s0 + (np.random.rand() - .5) * max_delta_s
bg = bg0 + (np.random.rand() - .5) * max_delta_bg
gauss = amp * np.exp(-r_sq / 2 / s ** 2)
totp = gauss.sum()
self.imgs[i] = gauss + bg
self.reference_df.iloc[i] = x0, y0, amp, bg, totp, amp / totp
self.result_df = ph.analyze_peaks(self.imgs, window, self.res,
r1, r2, r3, print_output=False)[0]
def test_pos(self):
assert_array_less(np.abs(self.result_df.loc[:,'x0':'y0'] -
self.reference_df.loc[:,'x0':'y0']).as_matrix(),
1./self.res)
def test_amp(self):
pdt.assert_almost_equal(self.result_df['amp'], self.reference_df['amp'],
check_less_precise=3)
def test_bg(self):
pdt.assert_almost_equal(self.result_df['bg'], self.reference_df['bg'],
check_less_precise=3)
def test_totp(self):
pdt.assert_almost_equal(self.result_df['tot power'],
self.reference_df['tot power'],
check_less_precise=3)
def test_norm(self):
pdt.assert_almost_equal(self.result_df['norm amp'],
self.reference_df['norm amp'],
check_less_precise=3)
class TestErrfFFTW(object):
def setup(self):
if not ph.FFTW_LOADED:
raise nose.SkipTest('FFTW not loaded.')
F = np.random.rand(4, 256, 256)
Z = np.random.rand(4)
wl = np.random.rand()
ph0 = np.random.rand(256, 256)
errf_numpy = ph.Errf(Z[0], Z[1:], F[0], F[1:], wl)
errf_fftw = ph.Errf_FFTW(Z[0], Z[1:], F[0], F[1:], wl)
self.E_numpy, self.dE_numpy = errf_numpy(ph0)
self.E_fftw, self.dE_fftw = errf_fftw(ph0)
def test_E(self):
assert_approx_equal(self.E_fftw, self.E_numpy)
def test_dE(self):
assert_allclose(self.dE_fftw, self.dE_numpy)
class TestErrfCUDA(object):
def setup(self):
if not ph.CUDA_LOADED:
raise nose.SkipTest('CUDA not loaded.')
F = np.random.rand(4, 256, 256)
Z = np.random.rand(4)
wl = np.random.rand()
ph0 = np.random.rand(256, 256)
errf_numpy = ph.Errf(Z[0], Z[1:], F[0], F[1:], wl)
errf_cuda = ph.Errf_CUDA(Z[0], Z[1:], F[0], F[1:], wl)
self.E_numpy, self.dE_numpy = errf_numpy(ph0)
self.E_cuda, self.dE_cuda = errf_cuda(ph0)
def test_E(self):
assert_approx_equal(self.E_cuda, self.E_numpy)
def test_dE(self):
assert_allclose(self.dE_cuda, self.dE_numpy)
class TestTransformsFFTW(object):
def setup(self):
if not ph.FFTW_LOADED:
raise nose.SkipTest('FFTW not loaded.')
N = 256
self.U = np.random.rand(N, N) + 1.j * np.random.rand(N, N)
self.z = np.random.rand()
self.wl = np.random.rand()
self.trans_numpy = ph.Transforms(N)
self.trans_fftw = ph.Transforms_FFTW(N)
def test_fft(self):
fft_numpy = self.trans_numpy.fft(self.U)
fft_fftw = self.trans_fftw.fft(self.U)
assert_allclose(fft_fftw, fft_numpy)
def test_ifft(self):
ifft_numpy = self.trans_numpy.ifft(self.U)
ifft_fftw = self.trans_fftw.ifft(self.U)
assert_allclose(ifft_fftw, ifft_numpy)
def test_fraun(self):
fraun_numpy_pos = self.trans_numpy.fraun(self.U, self.z, self.wl)
fraun_fftw_pos = self.trans_fftw.fraun(self.U, self.z, self.wl)
fraun_numpy_neg = self.trans_numpy.fraun(self.U, -self.z, self.wl)
fraun_fftw_neg = self.trans_fftw.fraun(self.U, -self.z, self.wl)
assert_allclose(fraun_fftw_pos, fraun_numpy_pos)
assert_allclose(fraun_fftw_neg, fraun_numpy_neg)
def test_asp(self):
raise nose.SkipTest('To be fixed.')
asp_numpy_pos = self.trans_numpy.asp(self.U, self.z, self.wl)
asp_fftw_pos = self.trans_fftw.asp(self.U, self.z, self.wl)
asp_numpy_neg = self.trans_numpy.asp(self.U, -self.z, self.wl)
asp_fftw_neg = self.trans_fftw.asp(self.U, -self.z, self.wl)
assert_allclose(asp_fftw_pos, asp_numpy_pos)
assert_allclose(asp_fftw_neg, asp_numpy_neg)
class TestTransformsCUDA(object):
def setup(self):
if not ph.CUDA_LOADED:
raise nose.SkipTest('CUDA not loaded.')
N = 256
self.U = np.random.rand(N, N) + 1.j * np.random.rand(N, N)
self.z = np.random.rand()
self.wl = np.random.rand()
self.trans_numpy = ph.Transforms(N)
self.trans_cuda = ph.Transforms_CUDA(N)
def test_fft(self):
fft_numpy = self.trans_numpy.fft(self.U)
fft_cuda = self.trans_cuda.fft(self.U)
assert_allclose(fft_cuda, fft_numpy)
def test_ifft(self):
ifft_numpy = self.trans_numpy.ifft(self.U)
ifft_cuda = self.trans_cuda.ifft(self.U)
assert_allclose(ifft_cuda, ifft_numpy)
def test_fraun(self):
fraun_numpy_pos = self.trans_numpy.fraun(self.U, self.z, self.wl)
fraun_cuda_pos = self.trans_cuda.fraun(self.U, self.z, self.wl)
fraun_numpy_neg = self.trans_numpy.fraun(self.U, -self.z, self.wl)
fraun_cuda_neg = self.trans_cuda.fraun(self.U, -self.z, self.wl)
assert_allclose(fraun_cuda_pos, fraun_numpy_pos)
assert_allclose(fraun_cuda_neg, fraun_numpy_neg)
def test_asp(self):
asp_numpy_pos = self.trans_numpy.asp(self.U, self.z, self.wl)
asp_cuda_pos = self.trans_cuda.asp(self.U, self.z, self.wl)
asp_numpy_neg = self.trans_numpy.asp(self.U, -self.z, self.wl)
asp_cuda_neg = self.trans_cuda.asp(self.U, -self.z, self.wl)
assert_allclose(asp_cuda_pos, asp_numpy_pos)
assert_allclose(asp_cuda_neg, asp_numpy_neg)
class TestZernike(object):
def setup(self):
self.N = 512
self.a = 150.4
v, u = np.indices((self.N, self.N))
r = np.sqrt((u - self.N / 2.) ** 2 + (v - self.N / 2.) ** 2) / self.a
p = np.arctan2(v - self.N / 2., u - self.N / 2.)
self.R = ph.circle(self.N / 2., self.N / 2., self.a, self.N)
self.Z = np.ones((8, self.N, self.N), dtype=np.float64)
self.Z[1] = 2. * r * np.cos(p)
self.Z[2] = 2. * r * np.sin(p)
self.Z[3] = np.sqrt(3.) * (2. * r ** 2 - 1.)
self.Z[4] = np.sqrt(6.) * r ** 2 * np.sin(2. * p)
self.Z[5] = np.sqrt(6.) * r ** 2 * np.cos(2. * p)
self.Z[6] = np.sqrt(8.) * (3. * r ** 3 - 2. * r) * np.sin(p)
self.Z[7] = np.sqrt(8.) * (3. * r ** 3 - 2. * r) * np.cos(p)
self.C = np.random.rand(8) * .1
self.W = np.sum(self.C[:,None,None] * self.Z, axis=0)
self.zern = ph.Zernike(self.N, 8)
self.zern.make_zernikes(self.N / 2., self.N / 2., self.a)
def test_poly(self):
assert_allclose(self.zern.Z, self.Z)
def test_wf(self):
assert_allclose(self.zern(self.C, self.N / 2., self.N / 2., self.a),
self.W)
assert_allclose(self.zern(self.C), self.W)
def test_fit(self):
assert_allclose(self.zern.fit(self.W, self.N / 2., self.N / 2., self.a),
self.C, atol=1e-4)
assert_allclose(self.zern.fit(self.W), self.C, atol=1e-4)
# Run the tests!
if __name__ == '__main__':
result = nose.run(argv=['ignored', '-v'])