Python for NLP: Neural Machine Translation with Seq2Seq in Keras
This is the 22nd article in my series of articles on Python for NLP. In one of my previous articles on solving sequence problems with Keras, I explained how to solve many to many sequence problems where both inputs and outputs are divided over multiple time-steps. The seq2seq architecture is a type of many-to-many sequence modeling, and is commonly used for a variety of tasks such as Text-Summarization, chatbot development, conversational modeling, and neural machine translation, etc.
In this article, we will see how to create a language translation model which is also a very famous application of neural machine translation. We will use seq2seq architecture to create our language translation model using Python’s Keras library.
It is assumed that you have good knowledge of recurrent neural networks, particularly LSTM. The code in this article is written in Python with the Keras library. Therefore, it is assumed that you have good knowledge of the Python language, as well as the Keras library. So, without any further ado, let’s begin.
Libraries and Configuration Settings
As a first step, we will import the required libraries and