fastText Quick Start Guide · Get Started With Facebook's Library for Text Representation and Classification by Bhattacharjee, Joydeep -- Read -- Imperial Library of Trantor

Index

Title Page Copyright and Credits

fastText Quick Start Guide

Dedication Packt Upsell

Why subscribe? PacktPub.com

Contributors

About the author About the reviewer Packt is searching for authors like you

Preface

Who this book is for What this book covers To get the most out of this book

Download the example code files Conventions used

Get in touch

Reviews

First Steps Introducing FastText

Introducing fastText Installing fastText

Prerequisites

Windows Linux

Installing dependencies on RHEL machines supporting the yum package manager Installing dependencies on Debian-based machines such as Ubuntu Installing dependencies on Arch Linux using pacman

Installing dependencies on Mac systems Installing Python dependencies

Installing fastText on Windows Installing fastText in Linux and macOS

Using a Docker image for fastText Summary

Creating Models Using FastText Command Line

Text classification using fastText

Text preprocessing English text and text using other Roman alphabets Downloading the data Preprocessing the Yelp data Text normalization

Removing stop words Normalizing Shuffling all the data

Dividing into training and validation Model building

Model training Model testing and evaluation

Precision and recall Confusion matrix

Hyperparameters

Epoch Learning rate N-grams Start with pretrained word vectors Finding the best fastText hyperparameters

Model quantization Understanding the model

FastText word vectors

Creating word vectors

Downloading from Wikipedia Text normalization Create word vectors Model evaluation

Nearest neighbors Word analogies

Other parameters when training Out of vocabulary words

Facebook word vectors Using pretrained word vectors

Machine translation

Summary

The FastText Model Word Representations in FastText

Word-to-vector representations

Types of word representations Getting vector representations from text

One-hot encoding Bag of words

TF-IDF N-grams

Model architecture in fastText

The unsupervised model

Skipgram Subword information skipgram Implementing skipgram  CBOW CBOW implementation Comparison between skipgram and CBOW

Loss functions and optimization

Softmax Hierarchical softmax Negative sampling Subsampling of frequent words

Context definitions

Summary

Sentence Classification in FastText

Sentence classification fastText supervised learning

Architecture

Hierarchical softmax architecture The n-gram features and the hashing trick

The FNV hash

Word embeddings and their use in sentence classification

fastText model quantization

Compression techniques

Quantization Vector quantization

Finding the codebook for high-dimensional spaces

Product quantization Additional steps

Summary

Using FastText in Your Own Models FastText in Python

FastText official bindings

PyBind Preprocessing data Unsupervised learning

Training in fastText Evaluating the model

Word vectors Nearest neighbor queries Word similarity Model performance Model visualization

Supervised learning

Data preprocessing and normalization Training the model Prediction Testing the model Confusion matrix

Gensim

Training a fastText model

Hyperparameters Model saving and loading Word vectors Model Evaluation

Word Mover's Distance

Getting more out of the training process

Machine translation using Gensim

Summary

Machine Learning and Deep Learning Models

Scikit-learn and fastText

Custom classifiers for fastText Bringing the whole thing together

Embeddings Keras

Embedding layer in Keras Convolutional neural networks

TensorFlow

Word embeddings in TensorFlow RNN architectures

PyTorch

The torchtext library

Data classes in torchtext Using the iterators

Bringing it all together

Summary

Deploying Models to Web and Mobile

Deploying to the web

Flask

The fastText functions The flask endpoints

Deploying to smaller devices

Prerequisites – Completing the Google tutorial App considerations Adding the fastText model FastText in Java Adding the library dependencies to Android Using library dependencies in Android Finally the app

Summary

Notes for the Readers

Windows and Linux Python 2 and Python 3 The fastText command line

The fastText supervised The fastText skipgram  The fastText cbow

Gensim fastText parameters

References

Chapter 3 Chapter 4 Chapter 5 Chapter 6 Chapter 7

Other Books You May Enjoy

Leave a review - let other readers know what you think

← Prev
Back
Next →

← Prev
Back
Next →