Book Image

Data Augmentation with Python

By : Duc Haba
Book Image

Data Augmentation with Python

By: Duc Haba

Overview of this book

Data is paramount in AI projects, especially for deep learning and generative AI, as forecasting accuracy relies on input datasets being robust. Acquiring additional data through traditional methods can be challenging, expensive, and impractical, and data augmentation offers an economical option to extend the dataset. The book teaches you over 20 geometric, photometric, and random erasing augmentation methods using seven real-world datasets for image classification and segmentation. You’ll also review eight image augmentation open source libraries, write object-oriented programming (OOP) wrapper functions in Python Notebooks, view color image augmentation effects, analyze safe levels and biases, as well as explore fun facts and take on fun challenges. As you advance, you’ll discover over 20 character and word techniques for text augmentation using two real-world datasets and excerpts from four classic books. The chapter on advanced text augmentation uses machine learning to extend the text dataset, such as Transformer, Word2vec, BERT, GPT-2, and others. While chapters on audio and tabular data have real-world data, open source libraries, amazing custom plots, and Python Notebook, along with fun facts and challenges. By the end of this book, you will be proficient in image, text, audio, and tabular data augmentation techniques.
Table of Contents (17 chapters)
Part 1: Data Augmentation
Part 2: Image Augmentation
Part 3: Text Augmentation
Part 4: Audio Data Augmentation
Part 5: Tabular Data Augmentation

Real-world audio datasets

By now, you should be familiar with downloading Pluto and real-world datasets from the Kaggle website. We chose to download Pluto from Chapter 2 because the image augmentation functions shown in Chapters 3 and 4, and the text augmentation techniques shown in Chapters 5 and 6, are not beneficial for audio augmentation.

The three real-world audio datasets we will use are as follows:

  • The Musical Emotions Classification (MEC) real-world audio dataset from Kaggle contains 2,126 songs separated into train and test folders. They are instrumental music, and the goal is to predict happy or sad music. Each piece is about 9 to 10 minutes in length and is in *.wav format. It was published in 2020 and is available to the public. Its license is Attribution-ShareAlike 4.0 International (CC BY-SA 4.0):
  • The Crowd Sourced Emotional Multimodal Actors Dataset (CREMA-D) real-world audio dataset from Kaggle contains...