-
Notifications
You must be signed in to change notification settings - Fork 14
/
test_core.py
63 lines (52 loc) · 1.8 KB
/
test_core.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
import unittest
import numpy as np
import ptstat as stat
import torch
import numpy.testing as npt
from torch.autograd import Variable
# Make sure tests are deterministic:
cuda = False # torch.cuda.is_available()
np.random.seed(1)
torch.manual_seed(1)
if cuda:
torch.cuda.manual_seed(1)
def to_np(x):
if cuda:
return x.data.cpu().numpy()
else:
return x.data.numpy()
class TestRandomVariables(unittest.TestCase):
def setUp(self):
batch_size = 2
rv_dimension = 5
p = torch.normal(torch.zeros(batch_size, rv_dimension), torch.ones(batch_size, rv_dimension))
p_pos = torch.abs(torch.normal(torch.zeros(batch_size, rv_dimension), torch.ones(batch_size, rv_dimension)))
p_pos = torch.clamp(p_pos, 0.1, 0.9)
if cuda:
p = p.cuda()
p_pos = p_pos.cuda()
p = Variable(p)
p_pos = Variable(p_pos)
self.rv = [
stat.Normal(size=(batch_size, rv_dimension), cuda=cuda),
stat.Normal(p, p_pos),
stat.Categorical(size=(batch_size, rv_dimension), cuda=cuda),
stat.Categorical(p_pos / torch.sum(p_pos, 1).expand_as(p_pos)),
stat.Bernoulli(size=(batch_size, rv_dimension), cuda=cuda),
stat.Bernoulli(p_pos),
stat.Uniform(size=(batch_size, rv_dimension), cuda=cuda)
]
def test_entropy(self):
mc_samples = 5000
for rv in self.rv:
entropy_avg = 0
x_avg = 0
for n in range(mc_samples):
x = rv.sample()
x_avg += x
entropy_avg += - rv.log_pdf(x)
entropy_avg /= mc_samples
x_avg /= mc_samples
npt.assert_allclose(to_np(entropy_avg), to_np(rv.entropy()), 0.02)
if __name__ == '__main__':
unittest.main()