Log1p

Usage

torch_log1p(self)

Arguments

self

(Tensor) the input tensor.

Note

This function is more accurate than torch_log for small values of input

log1p(input, out=NULL) -> Tensor

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

$$y_i={\log }_e(x_i+1)$$

Examples

if (torch_is_installed()) {

a = torch_randn(c(5))
a
torch_log1p(a)
}
#> torch_tensor
#>     nan
#>  0.9079
#> -0.6770
#>  0.4305
#>     nan
#> [ CPUFloatType{5} ]