# chainer.testing.unary_math_function_unittest¶

chainer.testing.unary_math_function_unittest(func, func_expected=None, label_expected=None, make_data=None, is_linear=None, forward_options=None, backward_options=None, double_backward_options=None)[source]

Decorator for testing unary mathematical Chainer functions.

This decorator makes test classes test unary mathematical Chainer functions. Tested are forward and backward, including double backward, computations on CPU and GPU across parameterized shape and dtype.

Parameters
• func (function or Function) – Chainer function to be tested by the decorated test class. Taking Function is for backward compatibility.

• func_expected – Function used to provide expected values for testing forward computation. If not given, a corresponsing numpy function for func is implicitly picked up by its name.

• label_expected (string) – String used to test labels of Chainer functions. If not given, the name of func is implicitly used.

• make_data – Function to customize input and gradient data used in the tests. It takes shape and dtype as its arguments, and returns a tuple of input, gradient and double gradient data. By default, uniform destribution ranged [-1, 1] is used for all of them.

• is_linear – Tells the decorator that func is a linear function so that it wraps func as a non-linear function to perform double backward test. This argument is left for backward compatibility. Linear functions can be tested by default without specifying is_linear in Chainer v5 or later.

• forward_options (dict) – Options to be specified as an argument of chainer.testing.assert_allclose() function. If not given, preset tolerance values are automatically selected.

• backward_options (dict) – Options to be specified as an argument of chainer.gradient_check.check_backward() function. If not given, preset tolerance values are automatically selected depending on dtype.

• double_backward_options (dict) – Options to be specified as an argument of chainer.gradient_check.check_double_backward() function. If not given, preset tolerance values are automatically selected depending on dtype.

The decorated test class tests forward, backward and double backward computations on CPU and GPU across the following parameterize() ed parameters:

• shape: rank of zero, and rank of more than zero

• dtype: numpy.float16, numpy.float32 and numpy.float64

Additionally, it tests the label of the Chainer function.

Chainer functions tested by the test class decorated with the decorator should have the following properties:

• Unary, taking one parameter and returning one value

• dtype of input and output are the same

• Elementwise operation for the supplied ndarray

Example

The following code defines a test class that tests sin() Chainer function, which takes a parameter with dtype of float and returns a value with the same dtype.

>>> import unittest
>>> from chainer import testing
>>> from chainer import functions as F
>>>
>>> @testing.unary_math_function_unittest(F.sin)
... class TestSin(unittest.TestCase):
...     pass


Because the test methods are implicitly injected to TestSin class by the decorator, it is enough to place pass in the class definition.

To customize test data, make_data optional parameter can be used. The following is an example of testing sqrt Chainer function, which is tested in positive value domain here instead of the default input.

>>> import numpy
>>>
>>> def make_data(shape, dtype):
...     x = numpy.random.uniform(0.1, 1, shape).astype(dtype)
...     gy = numpy.random.uniform(-1, 1, shape).astype(dtype)
...     ggx = numpy.random.uniform(-1, 1, shape).astype(dtype)
...     return x, gy, ggx
...
>>> @testing.unary_math_function_unittest(F.sqrt,
...                                       make_data=make_data)
... class TestSqrt(unittest.TestCase):
...     pass


make_data function which returns input, gradient and double gradient data generated in proper value domains with given shape and dtype parameters is defined, then passed to the decorator’s make_data parameter.