Expm1

This example implements and tests the expm1 operator - ie. element-wise exp(x) - 1, which is equivalent to that defined for numpy.

expm1(x)

Define the expm1 operator by defining the its operator function to be

\[out_{i} = exp(x_{i}) - 1.0\]

Parameters:x – The input tensor Returns:Element-wise exp(x) - 1 Examples:

>>> import numpy as np
>>> from opveclib import evaluate
>>> from opveclib.examples import expm1
>>> a = np.array([1e-10, -1e-10])
>>> evaluate(expm1(a))
array([  1.00000000e-10,  -1.00000000e-10])
>>> np.expm1(a)
array([  1.00000000e-10,  -1.00000000e-10])
>>> ones = np.ones_like(a)
>>> np.exp(a) - ones
array([  1.00000008e-10,  -1.00000008e-10])

expm1_grad(x, grad)

Define the expm1 gradient operator by defining the its operator function to be

\[out_{i} = exp(x_{i}) * grad_{i}\]

Parameters:

• x – The input tensor argument
• grad – The input gradient tensor to the gradient operator

Returns:

Element-wise gradient of the original operator