DEV Community

Fawaz Siddiqi for IBM Developer

Posted on • Updated on

Your Path to Deep Learning


The majority of data in the world is unlabeled and unstructured, for instance, images, sound, and text data. Shallow neural networks cannot easily capture relevant structures within this type of data, but deep networks are capable of discovering the hidden structures. With the help of various frameworks we can build neural networks that allow us to demystify various use cases based on various industries and predict outcomes, thus they have now become an integral part of our day to day lives.

As a follow-up to our Your Path to AI series which was conducted in 2020, we are now conducting Your Path to Deep Learning. A series which is mainly focused on the fundamentals of deep learning and with this series you will get an understanding of deep learning concepts, deep learning architectures, a comparison of deep learning frameworks and you will build various deep learning models throughout the series for linear and logistic regression, recurrent neural networks and Restricted Boltzmann Machines.

This series consists of 4 workshops that really dive deep into the essentials and working of various types of neural networks.

Workshop 1: User Reviews Sentiment Analysis using Logistic Regression

In this first workshop of the Your Path to Deep Learning series, we talk about how to use Natural language processing to classify user reviews from the IMDB Dataset. Developers can leverage user reviews for a given product or service from social media sites and customer reviews, and use Deep Learning models & identify patterns to predict sentiments, or other statistics to be used for business decisions. In this workshop, the user reviews will be classified using Logistic Regression to predict whether the review is positive or negative.

Workshop 1 Resources

Workshop 2: Identify Handwritten Digits using Convolutional Neural Networks with TensorFlow

Recognising handwritten numbers is a piece of cake for humans, but it’s a non-trivial task for machines. Currently, however, with the advancement of machine learning, people have made machines more capable of performing this task. 

In this second workshop of the Your Path to Deep Learning series, we will used the MNIST Dataset to build two Neural Networks capable to perform handwritten digits classification. The first Network is a simple Multi-layer Perceptron (MLP) and the second one is a Convolutional Neural Network (CNN). We will show you the step by step guide to create a simple handwritten digit recognizer in Watson Studio with TensorFlow.

Workshop 2 Resources

Workshop 3: Language Processing using Recurrent Neural Networks with TensorFlow

Language modelling is the task of assigning probabilities to sequences of words, and is one of the most important tasks in natural language processing. Given the context of one word or a sequence of words in the language that the language model was trained on, the model should provide the next most probable words or sequence of words that follows from the given sequence of words in the sentence.

In this third workshop of the Your Path to Deep Learning series, we will talk about how to perform language modeling on the Penn Treebank data set by creating a Recurrent Neural Network using long short-term memory (LSTM) units in a Jupyter notebook. 

Workshop 3 Resources

Workshop 4: Personalized Recommendation Engines with TensorFlow

Recommendation engines are an integral part of our lives and touch us with nearly every digital service we use these days. They allow users to gain a personalised experience based on their behaviour and are used by most popular media & e-commerce platforms. 

In this fourth and last workshop of the Your Path to Deep Learning series, you will learn how to build a Restricted Boltzmann Machine using TensorFlow that will give you recommendations based on movies that have been watched. The data sets used in the workshops are from GroupLens, and contain movies, users, and movie ratings. 

You'll learn to use a sigmoid activation function for the neural network, and the recommendations returned are based on the recommendation score generated by a Restricted Boltzmann Machine (RBM) using Collaborative Filtering. 

Workshop 4 Resources

For us, we really enjoyed conducting this series as we got to interact with different people within the industry having different viewpoints.

Did I forget to mention? You can also get a badge and a certificate for the series!

And to get started sign up or login to IBM Cloud at

Workshops included in the series

You can access all the workshop resources here:

Check out our partners for the series!

Discussion (0)