Logit
Arguments
- self
(Tensor) the input tensor.
- eps
(float, optional) the epsilon for input clamp bound. Default:
None
logit(input, eps=None, *, out=None) -> Tensor
Returns a new tensor with the logit of the elements of input.
input is clamped to [eps, 1 - eps] when eps is not None.
When eps is None and input < 0 or input > 1, the function will yields NaN.
$$ y_{i} = \ln(\frac{z_{i}}{1 - z_{i}}) \\ z_{i} = \begin{array}{ll} x_{i} & \mbox{if eps is None} \\ \mbox{eps} & \mbox{if } x_{i} < \mbox{eps} \\ x_{i} & \mbox{if } \mbox{eps} \leq x_{i} \leq 1 - \mbox{eps} \\ 1 - \mbox{eps} & \mbox{if } x_{i} > 1 - \mbox{eps} \end{array} $$
Examples
if (torch_is_installed()) {
a <- torch_rand(5)
a
torch_logit(a, eps=1e-6)
}
#> torch_tensor
#> 0.9320
#> 0.3958
#> 3.1804
#> -0.0449
#> 0.4256
#> [ CPUFloatType{5} ]