Book Image

Natural Language Processing with Flair

By : Tadej Magajna
Book Image

Natural Language Processing with Flair

By: Tadej Magajna

Overview of this book

Flair is an easy-to-understand natural language processing (NLP) framework designed to facilitate training and distribution of state-of-the-art NLP models for named entity recognition, part-of-speech tagging, and text classification. Flair is also a text embedding library for combining different types of embeddings, such as document embeddings, Transformer embeddings, and the proposed Flair embeddings. Natural Language Processing with Flair takes a hands-on approach to explaining and solving real-world NLP problems. You'll begin by installing Flair and learning about the basic NLP concepts and terminology. You will explore Flair's extensive features, such as sequence tagging, text classification, and word embeddings, through practical exercises. As you advance, you will train your own sequence labeling and text classification models and learn how to use hyperparameter tuning in order to choose the right training parameters. You will learn about the idea behind one-shot and few-shot learning through a novel text classification technique TARS. Finally, you will solve several real-world NLP problems through hands-on exercises, as well as learn how to deploy Flair models to production. By the end of this Flair book, you'll have developed a thorough understanding of typical NLP problems and you’ll be able to solve them with Flair.
Table of Contents (15 chapters)
1
Part 1: Understanding and Solving NLP with Flair
6
Part 2: Deep Dive into Flair – Training Custom Models
11
Part 3: Real-World Applications with Flair

Stacked embeddings

Stacked embeddings are a type of meta embeddings that allow us to form new embeddings by combining two or more embeddings together. These meta embeddings, as the name suggests, simply stack multiple embeddings on top of each other. They are ideal for combining the forward and backward versions of Flair embeddings. They can also be used for adding other embedding types to the mix, for example, to mix contextual string embeddings with the classic word embeddings. In fact, this is the type of a combination suggested by Flair and often the combination yielding state-of-the-art results.

To use stacked embeddings, simply instantiate a StackedEmbeddings class, passing in a list of embedding objects. You may then use this meta embedding in the same exact way you would use any other embedding in Flair.

Let's try this out on the set of embeddings recommended by Flair for best performance. We will stack glove word embeddings combined with news-forward and news-backward...