Aliases:
tf.math.log_softmax
tf.nn.log_softmax
tf.nn.log_softmax(
logits,
axis=None,
name=None,
dim=None
)
Defined in tensorflow/python/ops/nn_ops.py
.
Computes log softmax activations. (deprecated arguments)
For each batch i
and class j
we have
logsoftmax = logits - log(reduce_sum(exp(logits), axis))
Args:
logits
: A non-emptyTensor
. Must be one of the following types:half
,float32
,float64
.axis
: The dimension softmax would be performed on. The default is -1 which indicates the last dimension.name
: A name for the operation (optional).dim
: Deprecated alias foraxis
.
Returns:
A Tensor
. Has the same type as logits
. Same shape as logits
.
Raises:
InvalidArgumentError
: iflogits
is empty oraxis
is beyond the last dimension oflogits
.