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
|
"""Tests for module 1d Wasserstein solver"""
# Author: Adrien Corenflos <adrien.corenflos@aalto.fi>
# Nicolas Courty <ncourty@irisa.fr>
#
# License: MIT License
import numpy as np
import pytest
import ot
from ot.lp import wasserstein_1d
from ot.backend import get_backend_list
from scipy.stats import wasserstein_distance
backend_list = get_backend_list()
def test_emd_1d_emd2_1d_with_weights():
# test emd1d gives similar results as emd
n = 20
m = 30
rng = np.random.RandomState(0)
u = rng.randn(n, 1)
v = rng.randn(m, 1)
w_u = rng.uniform(0., 1., n)
w_u = w_u / w_u.sum()
w_v = rng.uniform(0., 1., m)
w_v = w_v / w_v.sum()
M = ot.dist(u, v, metric='sqeuclidean')
G, log = ot.emd(w_u, w_v, M, log=True)
wass = log["cost"]
G_1d, log = ot.emd_1d(u, v, w_u, w_v, metric='sqeuclidean', log=True)
wass1d = log["cost"]
wass1d_emd2 = ot.emd2_1d(u, v, w_u, w_v, metric='sqeuclidean', log=False)
wass1d_euc = ot.emd2_1d(u, v, w_u, w_v, metric='euclidean', log=False)
# check loss is similar
np.testing.assert_allclose(wass, wass1d)
np.testing.assert_allclose(wass, wass1d_emd2)
# check loss is similar to scipy's implementation for Euclidean metric
wass_sp = wasserstein_distance(u.reshape((-1,)), v.reshape((-1,)), w_u, w_v)
np.testing.assert_allclose(wass_sp, wass1d_euc)
# check constraints
np.testing.assert_allclose(w_u, G.sum(1))
np.testing.assert_allclose(w_v, G.sum(0))
@pytest.mark.parametrize('nx', backend_list)
def test_wasserstein_1d(nx):
from scipy.stats import wasserstein_distance
rng = np.random.RandomState(0)
n = 100
x = np.linspace(0, 5, n)
rho_u = np.abs(rng.randn(n))
rho_u /= rho_u.sum()
rho_v = np.abs(rng.randn(n))
rho_v /= rho_v.sum()
xb = nx.from_numpy(x)
rho_ub = nx.from_numpy(rho_u)
rho_vb = nx.from_numpy(rho_v)
# test 1 : wasserstein_1d should be close to scipy W_1 implementation
np.testing.assert_almost_equal(wasserstein_1d(xb, xb, rho_ub, rho_vb, p=1),
wasserstein_distance(x, x, rho_u, rho_v))
# test 2 : wasserstein_1d should be close to one when only translating the support
np.testing.assert_almost_equal(wasserstein_1d(xb, xb + 1, p=2),
1.)
# test 3 : arrays test
X = np.stack((np.linspace(0, 5, n), np.linspace(0, 5, n) * 10), -1)
Xb = nx.from_numpy(X)
res = wasserstein_1d(Xb, Xb, rho_ub, rho_vb, p=2)
np.testing.assert_almost_equal(100 * res[0], res[1], decimal=4)
def test_wasserstein_1d_type_devices(nx):
rng = np.random.RandomState(0)
n = 10
x = np.linspace(0, 5, n)
rho_u = np.abs(rng.randn(n))
rho_u /= rho_u.sum()
rho_v = np.abs(rng.randn(n))
rho_v /= rho_v.sum()
for tp in nx.__type_list__:
print(nx.dtype_device(tp))
xb = nx.from_numpy(x, type_as=tp)
rho_ub = nx.from_numpy(rho_u, type_as=tp)
rho_vb = nx.from_numpy(rho_v, type_as=tp)
res = wasserstein_1d(xb, xb, rho_ub, rho_vb, p=1)
nx.assert_same_dtype_device(xb, res)
def test_emd_1d_emd2_1d():
# test emd1d gives similar results as emd
n = 20
m = 30
rng = np.random.RandomState(0)
u = rng.randn(n, 1)
v = rng.randn(m, 1)
M = ot.dist(u, v, metric='sqeuclidean')
G, log = ot.emd([], [], M, log=True)
wass = log["cost"]
G_1d, log = ot.emd_1d(u, v, [], [], metric='sqeuclidean', log=True)
wass1d = log["cost"]
wass1d_emd2 = ot.emd2_1d(u, v, [], [], metric='sqeuclidean', log=False)
wass1d_euc = ot.emd2_1d(u, v, [], [], metric='euclidean', log=False)
# check loss is similar
np.testing.assert_allclose(wass, wass1d)
np.testing.assert_allclose(wass, wass1d_emd2)
# check loss is similar to scipy's implementation for Euclidean metric
wass_sp = wasserstein_distance(u.reshape((-1,)), v.reshape((-1,)))
np.testing.assert_allclose(wass_sp, wass1d_euc)
# check constraints
np.testing.assert_allclose(np.ones((n,)) / n, G.sum(1))
np.testing.assert_allclose(np.ones((m,)) / m, G.sum(0))
# check G is similar
np.testing.assert_allclose(G, G_1d, atol=1e-15)
# check AssertionError is raised if called on non 1d arrays
u = np.random.randn(n, 2)
v = np.random.randn(m, 2)
with pytest.raises(AssertionError):
ot.emd_1d(u, v, [], [])
def test_emd1d_type_devices(nx):
rng = np.random.RandomState(0)
n = 10
x = np.linspace(0, 5, n)
rho_u = np.abs(rng.randn(n))
rho_u /= rho_u.sum()
rho_v = np.abs(rng.randn(n))
rho_v /= rho_v.sum()
for tp in nx.__type_list__:
print(nx.dtype_device(tp))
xb = nx.from_numpy(x, type_as=tp)
rho_ub = nx.from_numpy(rho_u, type_as=tp)
rho_vb = nx.from_numpy(rho_v, type_as=tp)
emd = ot.emd_1d(xb, xb, rho_ub, rho_vb)
emd2 = ot.emd2_1d(xb, xb, rho_ub, rho_vb)
nx.assert_same_dtype_device(xb, emd)
nx.assert_same_dtype_device(xb, emd2)
|