From the course: Training Neural Networks in Python

Unlock this course with a free trial

Join today to access over 25,600 courses taught by industry experts.

A simple model of a neuron

A simple model of a neuron - Python Tutorial

From the course: Training Neural Networks in Python

Start my 1-month free trial Buy for my team

A simple model of a neuron

“

- [Instructor] So based on the roles of the dendrites, nucleus, and axons, let's just assign those roles to a set of inputs a calculation unit, and an output respectively. To come up with the block diagram, you can see at the right, we'll call the inputs X0 through Xn minus 1 for a neuron with n inputs. Now, biological neurons seem to react in different sensitivities to different incoming signals, so we'll model that mathematically by assigning a weight to each input channel. Let's call them W0 through Wn minus 1. Next, the computation performed in order to fire or not is a simple weighted sum as shown inside the circle and that's exactly what our output will be. Here's an example of how a neuron would help us classify. Let's suppose we have a two input neuron and we feed data with two features into it. These features are shown in the plot at the right, which shows the risk of getting some medical condition X based…

Contents