Logcumsumexp

Usage

torch_logcumsumexp(self, dim)

Arguments

self

(Tensor) the input tensor.

dim

(int) the dimension to do the operation over

logcumsumexp(input, dim, *, out=None) -> Tensor

Returns the logarithm of the cumulative summation of the exponentiation of elements of input in the dimension dim.

For summation index $j$ given by dim and other indices $i$, the result is

$$\text{logcumsumexp}(x{)}_{ij}=\log \sum _{j=0}^i\exp (x_{ij})$$

Examples

if (torch_is_installed()) {

a <- torch_randn(c(10))
torch_logcumsumexp(a, dim=1)
}
#> torch_tensor
#> -0.6432
#>  0.8586
#>  1.0778
#>  1.1589
#>  2.0787
#>  2.2299
#>  2.2579
#>  2.8005
#>  2.9865
#>  3.1512
#> [ CPUFloatType{10} ]