/
erfcinv.py
63 lines (48 loc) · 1.66 KB
/
erfcinv.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
try:
from scipy import special
available_cpu = True
except ImportError as e:
available_cpu = False
_import_error = e
import math
import chainer
from chainer.backends import cuda
from chainer import function_node
from chainer import utils
from chainer.utils import type_check
BACKWORDC = math.pi ** 0.5 / 2
class ErfcInv(function_node.FunctionNode):
@property
def label(self):
return 'erfcinv'
def check_type_forward(self, in_types):
type_check.expect(in_types.size() == 1)
type_check.expect(in_types[0].dtype.kind == 'f')
def forward_cpu(self, x):
if not available_cpu:
raise ImportError('SciPy is not available. Forward computation'
' of erfcinv in CPU cannot be done. ' +
str(_import_error))
self.retain_outputs((0,))
return utils.force_array(special.erfcinv(x[0]), dtype=x[0].dtype),
def forward_gpu(self, x):
self.retain_outputs((0,))
return cuda.elementwise(
'T x', 'T y',
'y = erfcinv(x)',
'elementwise_erfcinv',
)(x[0]),
def backward(self, indexes, gy):
y, = self.get_retained_outputs()
return -BACKWORDC * chainer.functions.exp(y ** 2) * gy[0],
def erfcinv(x):
"""Elementwise inverse function of complementary error function.
.. note::
Forward computation in CPU cannot be done if
`SciPy <https://www.scipy.org/>`_ is not available.
Args:
x (:class:`~chainer.Variable` or :ref:`ndarray`): Input variable.
Returns:
~chainer.Variable: Output variable.
"""
return ErfcInv().apply((x,))[0]