torch.logcumsumexp#

torch.logcumsumexp(input,dim,*,out=None)→Tensor#

Returns the logarithm of the cumulative summation of the exponentiation ofelements ofinput in the dimensiondim.

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

$$\text{logcumsumexp}(x{)}_{ij}=\log \sum _{k=0}^j\exp (x_{ik})$$logcumsumexp(x)ij​=logk=0∑j​exp(xik​)

Parameters

• input (Tensor) – the input tensor.

• dim (int) – the dimension to do the operation over

Keyword Arguments

out (Tensor,optional) – the output tensor.

Example:

>>>a=torch.randn(10)>>>torch.logcumsumexp(a,dim=0)tensor([-0.42296738, -0.04462666,  0.86278635,  0.94622083,  1.05277811,         1.39202815,  1.83525007,  1.84492621,  2.06084887,  2.06844475]))