Normalizes the given input across the specified dimension to unit length.
tf.contrib.layers.unit_norm( inputs, dim, epsilon=1e-07, scope=None )
Note that the rank of
input must be known.
Tensorof arbitrary size.
dim: The dimension along which the input is normalized.
epsilon: A small value to add to the inputs to avoid dividing by zero.
scope: Optional scope for variable_scope.
ValueError: If dim is smaller than the number of dimensions in 'inputs'.