NLP from Scratch

In these three-part series you will build and train a basic character-level Recurrent Neural Network (RNN) to classify words.

You will learn:

• How to construct Recurrent Neural Networks from scratch

• Essential data handling techniques for NLP

• How to train an RNN to identify the language origin of words.

Before you begin, we recommend that you review the following:

• PyTorch Learn the Basics series

• How to install PyTorch

NLP From Scratch - Part 1: Classifying Names with a Character-Level RNN

Learn how to use an RNN to classify names into their language of origin.

Code

https://pytorch.org/tutorials/intermediate/char_rnn_classification_tutorial.html

NLP From Scratch - Part 2: Generating Names with a Character-Level RNN

Expand the RNN we created in Part 1 to generate names from languages.

Code

https://pytorch.org/tutorials/intermediate/char_rnn_generation_tutorial.html

NLP From Scratch - Part 3: Translation with a Sequence to Sequence Network and Attention

Create a sequence-to-sequence model that can translate your text from French to English.

Code

https://pytorch.org/tutorials/intermediate/seq2seq_translation_tutorial.html