Get hands-on experience in NLP with challenging applications.

Mastering Natural Language Processing

In this module, we’ll learn advanced deep learning concepts and practice building some advanced deep learning and natural language processing (NLP) projects. By the end, we’ll be able to utilize deep learning algorithms used at large in the industry. This is a project-based module with 12 projects in total. This will get us used to building real-world applications that are being used in a wide range of industries. We’ll be exposed to the most common tools used for machine learning projects, including NumPy, Matplotlib, scikit-learn, Tensorflow, and more.
Once we’re finished, we’ll have the experience to start building our amazing projects and some great new additions to our portfolio.

Building Advanced Deep Learning and NLP Projects

## What are LSTMs?
Long Short Term Memory Networks (LSTMs) in short, are modifications in the standard Recurrent Neural Network. Look at the image of a standard RNN architecture below.

In the above diagram:
* $X_{t-1}$ represents the input at time step `t-1`.
* $h_{t-1}$ represents the output of the RNN cell at time step `t-1`.

This is the architecture that has the vanishing gradient problem, which we discussed in our previous lessons. We’ve also maintained that, as a rule of thumb, if the recurring weight (denoted by $W_{rec}$) is small, the gradient is vanishing, and if the recurring weight ($W_{rec}$) is large, the gradient is exploding. We can say that we have a vanishing gradient if $W_{rec}$ < 1 or an exploding gradient if $W_{rec}$ > 1. With that being said, what’s the first thing that comes to your mind to solve this problem?

The easiest and fastest solution will probably be to make $W_{rec}$ = 1. That’s exactly what was done in LSTMs. Have a look at the architecture of an LSTM cell. 

# What are LSTMs?
Long Short Term Memory Networks (LSTMs) in short, are modifications in the standard Recurrent Neural Network. Look at the image of a standard RNN architecture below.

In this lesson, we will discuss a variation of RNN and its workings.

Project: Build a COVID-19 Detection System Using X-Rays

Project: Building a Pokemon Classifier Using Transfer Learning

Project: Text Generation Using Markov Chains

Word Embedding: Two Mini Projects

Project: IMDB Reviews Sentiment Analysis

Project: Deciphering Text Using Character-Level RNNs

Project: Emoji Predictor Using Transfer Learning in NLP

Final Exam

Where to Go Next?

Conslusion

Long Short Term Memory Networks (LSTM)

What are LSTMs?