Rate this Page
★★★★★

torch.fmax#

torch.fmax(input,other,*,out=None)→Tensor#

Computes the element-wise maximum ofinput andother.

This is liketorch.maximum() except it handles NaNs differently:if exactly one of the two elements being compared is a NaN then the non-NaN element is taken as the maximum.Only if both elements are NaN is NaN propagated.

This function is a wrapper around C++’sstd::fmax and is similar to NumPy’sfmax function.

Supportsbroadcasting to a common shape,type promotion, and integer and floating-point inputs.

Parameters

  • input (Tensor) – the input tensor.

  • other (Tensor) – the second input tensor

Keyword Arguments

out (Tensor,optional) – the output tensor.

Example:

>>>a=torch.tensor([9.7,float('nan'),3.1,float('nan')])>>>b=torch.tensor([-2.2,0.5,float('nan'),float('nan')])>>>torch.fmax(a,b)tensor([9.7000, 0.5000, 3.1000,    nan])