tf.nn.elu(features, name=None)

tf.nn.elu(features, name=None)

See the guide: Neural Network > Activation Functions

Computes exponential linear: exp(features) - 1 if < 0, features otherwise.

See Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)


  • features: A Tensor. Must be one of the following types: float32, float64, int32, int64, uint8, int16, int8, uint16, half.
  • name: A name for the operation (optional).


A Tensor. Has the same type as features.

Defined in tensorflow/python/ops/