flax.linen.activation.relu#
- flax.linen.activation.relu = <jax._src.custom_derivatives.custom_jvp object>[source]#
Rectified linear unit activation function.
Computes the element-wise function:
\[\mathrm{relu}(x) = \max(x, 0)\]except under differentiation, we take:
\[\nabla \mathrm{relu}(0) = 0\]For more information see Numerical influence of ReLU’(0) on backpropagation.
- Parameters
x – input array
- Returns
An array.
Example
>>> jax.nn.relu(jax.numpy.array([-2., -1., -0.5, 0, 0.5, 1., 2.])) Array([0. , 0. , 0. , 0. , 0.5, 1. , 2. ], dtype=float32)
See also