From the course: Machine Learning and AI in Cybersecurity by Pearson
Join today to access over 25,500 courses taught by industry experts.
Neural networks with TensorFlow
From the course: Machine Learning and AI in Cybersecurity by Pearson
Neural networks with TensorFlow
Now, let's get a little more specific, although we've already seen a code sample. Let's take a look at specific neural networks with TensorFlow. The simplest neural network is called a perceptron. It's an algorithm that takes an input vector of some values, input features, and then outputs either a 1, yes, or 0, no. You have a vector of weights, the weighting between the neural network node connections. You have what's called the dot product. Now that's something from linear algebra, and if you're not comfortable with linear algebra, you'll be able to complete this course, certainly, and you'll be able to do a great deal of machine learning. But at some point, I recommend you do learn some basic linear algebra. We see some basic elementary geometry being used to define a boundary that changes position according to values. Now, this is some sample code for a very simple perceptron. We have our NumPy array. We have a perceptron using the sigmoid activation function. We go through a…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.