Congratulations! You have now built your own perceptron from scratch.

Let’s step back and think about what you just accomplished and see if there are any limits to a single perceptron.

Earlier, the data points in the training set were linearly separable i.e. a single line could easily separate the two dissimilar sets of points.

What would happen if the data points were scattered in such a way that a line could no longer classify the points? A single perceptron with only two inputs wouldn’t work for such a scenario because it cannot represent a non-linear decision boundary.

That’s when more perceptrons and features come into play!

By increasing the number of features and perceptrons, we can give rise to the Multilayer Perceptrons, also known as Neural Networks, which can solve much more complicated problems.

With a solid understanding of perceptrons, you are now ready to dive into the incredible world of Neural Networks!