log(1 / mean(1 / exp(input_tensor))).
input_tensor along the dimensions given in
keepdims is true, the rank of the tensor is reduced by 1 for each entry in
keepdims is true, the reduced dimensions are retained with length
axis has no entries, all dimensions are reduced, and a tensor with a
single element is returned.
This function is more numerically stable than
log(1 / mean(1 - exp(input))).
It avoids overflows caused by taking the exp of large inputs and underflows
caused by taking the log of small inputs.
The tensor to reduce. Should have numeric type.
The dimensions to reduce. If
None (the default), reduces all
dimensions. Must be in the range
Boolean. Whether to keep the axis as singleton dimensions.
False (i.e., squeeze the reduced dimensions).
str or list ofstr
axis names to additionally
reduce over. ProvidingNone
will not reduce over any axes.
Allow using anall_gather
under TensorFlow when computing the distributed maximum. This fallback is
only efficient whenaxis
reduces away most of the dimensions ofinput_tensor
name prefixed to Ops created by this function.
The reduced tensor.