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

Mel-spectrogram and Chroma STFT plots

Pluto spends additional time plotting various Spectrograms because the augmentation technique is the same as in the Waveform graph in Chapter 7. Pluto will write fewer new wrapper functions. He will reuse the methods from the previous chapter, but before that, let’s draw more Spectrograms.

The subjective unit of pitch, also known as the mel scale, is a pitch unit with equal distance between pitches. S. S. Stevens, John Volkmann, and E. B. Newmann proposed the mel scale in the scholarly paper titled, A scale for the measurement of the psychological magnitude of pitch, in 1937.

The math calculation for the mel scale is complex. Thus, Pluto relies on the melspectrogram() method from the Librosa library to perform the computation. The Pluto draw_melspectrogram() wrapper method uses the Librosa melspectrogram() function, and the code snippet is as follows:

# code snippeet for the melspectrogram
mel = librosa.feature.melspectrogram(y...