View source on GitHub
|
Snake activation to learn periodic functions.
tfa.activations.snake(
x: tfa.types.TensorLike,
frequency: tfa.types.Number = 1
) -> tf.Tensor
Computes snake activation:
\[ \mathrm{snake}(x) = \mathrm{x} + \frac{1 - \cos(2 \cdot \mathrm{frequency} \cdot x)}{2 \cdot \mathrm{frequency} }. \]
See Neural Networks Fail to Learn Periodic Functions and How to Fix It.
Usage:
x = tf.constant([-1.0, 0.0, 1.0])tfa.activations.snake(x)<tf.Tensor: shape=(3,), dtype=float32, numpy=array([-0.29192656, 0. , 1.7080734 ], dtype=float32)>
Args | |
|---|---|
x
|
A Tensor.
|
frequency
|
A scalar, frequency of the periodic part. |
Returns | |
|---|---|
A Tensor. Has the same type as x.
|