Hey Community,

Today we are going on with the second part.

You can find all previous posts in this list:

Part	Headline	Steemit Link:
I	Introduction	https://steemit.com/neuronalnetwork/@haggy2k3/neuronal-networks-part-1-introduction
II	Units	Current Document
III	Connections	https://steemit.com/science/@haggy2k3/neuronal-networks-part-iii-connections
IV	Inputs	https://steemit.com/science/@haggy2k3/neuronal-networks-part-iv-inputs
V	Activities and Outputs
VI	Training and Test
VII	Metrics
VIII	Learning Rules
IX	Types of NN
X	Features of NN
XI	Use Cases
XII	Bonus: Getting started with Membrain

Units

Neural networks consist of several neurons. These neurons are also referred to
as units, units or nodes. They are used to collect information from the
environment or from other neurons and to forward it to other units or the
environment in a modified form.

In this picture you can see the three different types of neurons:

Input units (the red ones in the picture):

• units that can receive signals (stimuli, patterns) from the outside world.

Hidden units (the yellow ones in the picture):

• units that are located between input and output units and contain an
  internal modified representation of the outside world.

Output units (the green ones in the picture):

• units that transmit signals to the outside world.

nodes within the same vertical order (for example, the two output units on the
right in the figure) are grouped together as a layer.

If you like this post and you want to read more please upvote, resteem and/or follow my blog.

Best regards

Haggy2k3