Deep Learning is the hottest thing in machine learning right now. But what’s behind all the buzz? Well to begin with, deep learning is well… deep!

Deep learning makes use of multiple layers of nodes or neurons to extract complex relations between data.  Each layer processes the data and passes it onto the next all the way to the output layer. The reason deep learning is such a popular model is because of the versatility of layers, which allows them to be used across a wide range in anything from classifying dog photos, to sequence analysis or extracting user sentiment from text.  

So let’s jump right in ! I’m assuming you already have python installed. If not, go right ahead and download python from the official website. 

  1. Go to Command Prompt (Terminal in Mac) 

We’ll be looking at Keras, one of the most popular deep learning frameworks that is also very user friendly. So first we’ll need to install Keras and a few backend libraries that support it. Type in-

pip install keras
pip install numpy
pip install tensorflow
pip install jupyter notebook

2.Open Jupyter notebook

If you notice we installed something called a Jupyter notebook. This is a very popular integrated development environment (IDE) for python that is favoured by the machine learning community due to its clean look, easy customizability and ease with which graphs and plots can be displayed. So go ahead and type into the terminal –

jupyter notebook

3.Create your model

You should see the jupyter notebook tab open up in your browser. Go ahead and create a new file. You’re all set to create your first model. First let’s import Keras –  

Let’s go ahead and define our model – 

That’s it ! You’ve built your first deep learning model.  Don’t worry if you don’t understand all the code you’re writing yet. The drawback of deep learning is that due to the complexity of the model it has a lot of parameters to customize. However the default parameters work well enough for simple cases.

Check out the Keras tutorial if you want a deeper look at Keras and want to start feeding data into your model.