Real-world uses

Created for this library

1. 1.

   A computer vision team uses ReLU activations in CNN hidden layers because of fast convergence and simple gradient flow.

2. 2.

   An NLP team uses ReLU in feedforward sublayers as a simple, fast activation function for early experiments.

3. 3.

   A retail forecasting team uses ReLU activations in its hidden layers as a fast, well-understood baseline.