-
Notifications
You must be signed in to change notification settings - Fork 372
Expand file tree
/
Copy pathtest_barycenters.py
More file actions
166 lines (140 loc) · 5.8 KB
/
test_barycenters.py
File metadata and controls
166 lines (140 loc) · 5.8 KB
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
import numpy as np
import tslearn.barycenters
from tslearn.utils import to_time_series
__author__ = 'Romain Tavenard romain.tavenard[at]univ-rennes2.fr'
def test_set_weights():
w = tslearn.barycenters._set_weights(None, 3)
np.testing.assert_allclose(w, np.ones((3, )))
w = tslearn.barycenters._set_weights([.5, .25, .25], 3)
np.testing.assert_allclose(w, [.5, .25, .25])
w = tslearn.barycenters._set_weights([.5, .25, .25], 2)
np.testing.assert_allclose(w, np.ones((2, )))
def test_euclidean_barycenter():
n, sz, d = 15, 10, 3
rng = np.random.RandomState(0)
time_series = rng.randn(n, sz, d)
bar = tslearn.barycenters.euclidean_barycenter(time_series)
np.testing.assert_allclose(bar, time_series.mean(axis=0))
weights = rng.rand(n, )
weights /= np.sum(weights)
bar = tslearn.barycenters.euclidean_barycenter(time_series,
weights=weights)
np.testing.assert_allclose(bar, np.average(time_series,
axis=0, weights=weights))
def test_dba():
n, sz, d = 15, 10, 3
rng = np.random.RandomState(0)
time_series = rng.randn(n, sz, d)
# Equal length, 0 iterations -> Euclidean
euc_bar = tslearn.barycenters.euclidean_barycenter(time_series)
dba_bar = tslearn.barycenters.dtw_barycenter_averaging_petitjean(
time_series,
max_iter=0)
np.testing.assert_allclose(euc_bar, dba_bar)
# Equal length, >0 iterations
dba_bar = tslearn.barycenters.dtw_barycenter_averaging_petitjean(
time_series,
max_iter=5)
ref = np.array([[0.33447722, 0.0418787, -0.03953774],
[-0.75757987, -0.26841384, -0.22418874],
[-0.0473153, 0.41030073, 0.06069343],
[0.36250957, -0.79033572, 0.02300398],
[0.02783764, -0.05039364, -0.79595523],
[0.38139685, 0.37661911, 0.09506468],
[0.48628337, 0.17192078, -1.16404917],
[-0.40263459, 0.59364783, -0.6843561],
[0.67493146, -0.37714421, 0.16604165],
[-0.32249566, 0.09109832, 0.55489214]])
np.testing.assert_allclose(dba_bar, ref, atol=1e-6)
# With n_jobs
dba_bar = tslearn.barycenters.dtw_barycenter_averaging_petitjean(
time_series,
max_iter=5,
n_jobs=-1
)
np.testing.assert_allclose(dba_bar, ref, atol=1e-6)
dba_bar_mm = tslearn.barycenters.dtw_barycenter_averaging(
time_series,
max_iter=5)
np.testing.assert_allclose(dba_bar, dba_bar_mm)
# With n_jobs
dba_bar_mm = tslearn.barycenters.dtw_barycenter_averaging(
time_series,
max_iter=5,
n_jobs=-1
)
np.testing.assert_allclose(dba_bar, dba_bar_mm)
weights = rng.rand(n)
dba_bar = tslearn.barycenters.dtw_barycenter_averaging_petitjean(
time_series,
weights=weights,
max_iter=5)
dba_bar_mm = tslearn.barycenters.dtw_barycenter_averaging(
time_series,
weights=weights,
max_iter=5)
np.testing.assert_allclose(dba_bar, dba_bar_mm)
# With an init of different size
init_barycenter = rng.randn(sz-1, d)
ref = np.array([[0.421127, 0.054492, -0.124027],
[-0.498396, -0.200649, 0.029864],
[0.431492, -0.397721, -0.585897],
[-0.906793, 0.021529, -0.514742],
[0.629140, -0.152363, -0.434802],
[0.609915, 0.348230, 0.447403],
[-0.041696, 0.760246, -0.361783],
[-0.042272, -0.642698, -0.662480],
[0.261962, 0.111322, 0.389292]])
dba_bar = tslearn.barycenters.dtw_barycenter_averaging_petitjean(
time_series,
init_barycenter=init_barycenter,
max_iter=5,
verbose=1
)
dba_bar_mm = tslearn.barycenters.dtw_barycenter_averaging(
time_series,
init_barycenter=init_barycenter,
max_iter=5,
verbose=1
)
np.testing.assert_allclose(dba_bar, ref, atol=1e-6)
np.testing.assert_allclose(dba_bar_mm, ref, atol=1e-6)
# Subgradient averaging, 0 iter -> init
init_barycenter = [0]
np.testing.assert_array_equal(
tslearn.barycenters.dtw_barycenter_averaging_subgradient(
time_series,
init_barycenter=init_barycenter,
max_iter=0
),
to_time_series(init_barycenter)
)
# Subgradient averaging with a single input returns it
time_series = rng.randn(1, sz, d)
np.testing.assert_array_equal(
tslearn.barycenters.dtw_barycenter_averaging_subgradient(
time_series
),
time_series.reshape(sz, d)
)
def test_softdtw_barycenter():
n, sz, d = 15, 10, 3
rng = np.random.RandomState(0)
time_series = rng.randn(n, sz, d)
# Equal length, 0 iterations -> Euclidean
euc_bar = tslearn.barycenters.euclidean_barycenter(time_series)
sdtw_bar = tslearn.barycenters.softdtw_barycenter(time_series, max_iter=0)
np.testing.assert_allclose(euc_bar, sdtw_bar)
# Equal length, >0 iterations
sdtw_bar = tslearn.barycenters.softdtw_barycenter(time_series, max_iter=5)
ref = np.array([[0.28049395, -0.01190817, -0.06228361],
[-0.67097059, -0.10737132, -0.33867808],
[0.29380813, 0.0474172, 0.32718516],
[0.14438242, -0.56877605, -0.14563386],
[0.15620728, -0.04473494, -0.63912905],
[0.35989018, 0.42118863, -0.2127066],
[0.16831249, 0.65420655, -0.53587191],
[-0.20737107, 0.15301328, -0.74052802],
[0.53149515, -0.24839857, -0.03430969],
[-0.17690603, 0.07217633, 0.58071408]])
np.testing.assert_allclose(sdtw_bar, ref, atol=1e-6)