> A **Rectified Linear Unit**(ReLU) is a ```non-linear``` activation function that performs on multi-layer neural networks. (e.g., **f(x) = max(0,x)** where *x = input value*).
  
# What happens in ReLU layer?
In this layer we remove every negative value from the filtered image and replace it with zero. This function only activates when the node input is above a certain quantity. So, when the input is below zero the output is zero.
However, when the input rises above a certain ```threshold``` it has *linear relationship* with the dependent variable. This means that it is able to *accelerate the speed of* a training data set in a deep neural network that is faster than other activation functions -- this is done to avoid summing up with zero.

## A ReLU layer:
* performs ```element-wise``` operations.
* has an output that is a rectified feature map.
# Examples of ReLU
| x |f(x)=x| F(X)|
| - | - | - |
|-110 |f(-110)=0| 0  |
| -15|f(-15)=0| 0  |
|  15|f(15)=15 | 15  |
| 25 |f(25)=25 | 25 |

![](/api/edpresso/shot/4775123253460992/image/6094824000716800)


What is a ReLU layer?

A ReLU layer removes negative values by setting them to zero and accelerates training speed with a linear relationship when inputs are positive.

x	f(x)=x	F(X)
-110	f(-110)=0	0
-15	f(-15)=0	0
15	f(15)=15	15
25	f(25)=25	25