tensorflow::ops::LogSoftmax

#include <nn_ops.h>

Computes log softmax activations.

Summary

For each batch i and class j we have 

logsoftmax[i, j] = logits[i, j] - log(sum(exp(logits[i])))

Args:

  • scope: A Scope object
  • logits: 2-D with shape [batch_size, num_classes].

Returns:

  • Output: Same shape as logits.

Constructors and Destructors
LogSoftmax(const ::tensorflow::Scope & scope, ::tensorflow::Input logits)

Public attributes
logsoftmax
operation

Public functions
node() const
::tensorflow::Node *
operator::tensorflow::Input() const
operator::tensorflow::Output() const

Public attributes

logsoftmax

::tensorflow::Output logsoftmax

operation

Operation operation

Public functions

LogSoftmax

 LogSoftmax(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input logits
)

node

::tensorflow::Node * node() const

operator::tensorflow::Input

 operator::tensorflow::Input() const

operator::tensorflow::Output

 operator::tensorflow::Output() const