tensorplay.functional.log1p

tensorplay.functional.log1p(input, *, out=None) → Tensor[source]

Returns a new tensor with the natural logarithm of (1 + input).

$$y_i = \log_{e} (x_i + 1)$$

Note

This function is more accurate than tensorplay.log() for small values of input

Parameters:

input (Tensor) – the input tensor.

Keyword Arguments:

out (Tensor, optional) – the output tensor.

Example:

>>> a = tensorplay.randn(5)
>>> a
tensor([-1.0090, -0.9923,  1.0249, -0.5372,  0.2492])
>>> tensorplay.log1p(a)
tensor([    nan, -4.8653,  0.7055, -0.7705,  0.2225])