What this book covers
Chapter 1, Exploring Data for Machine Learning, provides an overview of data analysis and visualization methods using various Python libraries. Additionally, it deep dives into unlocking data insights with natural language using OpenAI LLMs.
Chapter 2, Labeling Data for Classification, covers the process of labeling tabular data for training classification models. Various methods, such as Snorkel Python functions, semi-supervised learning, and clustering data using K-means, are explored.
Chapter 3, Labeling Data for Regression, addresses the labeling of tabular data for training regression models. Techniques include leveraging summary statistics, creating pseudo labels, employing data augmentation methods, and utilizing K-means clustering.
Chapter 4, Exploring Image Data, covers the analysis and visualization of image data and feature extraction from images using various Python libraries.
Chapter 5, Labeling Image Data Using Rules, discusses labeling images based on heuristics and image properties such as aspect ratio, and also covers image classification using pre-trained classifiers such as YOLO.
Chapter 6, Labeling Image Data Using Data Augmentation, explores methods of image data augmentation for training support vector machines and Convolutional Neural Networks (CNNs), as well as addressing image data labeling.
Chapter 7, Labeling Text Data, covers generative AI and various methods for labeling text data. This includes Azure OpenAI with real-world use cases, text classification, and sentiment analysis using Snorkel and K-means clustering.
Chapter 8, Exploring Video Data, focuses on loading video data, extracting features, visualizing video data, and clustering video data using K-means clustering.
Chapter 9, Labeling Video Data, delves into labeling video data using CNNs, segmenting video data with the watershed algorithm, and capturing important features using autoencoders, accompanied by real-world examples.
Chapter 10, Exploring Audio Data, provides the fundamentals of audio data, loading and visualizing audio data, extracting features, and real-life applications.
Chapter 11, Labeling Audio Data, covers transcribing audio data using OpenAI’s Whisper model, labeling the transcription, creating spectrograms for audio data classification, augmenting audio data, and using Azure Cognitive Services for speech.
Chapter 12, Hands-On Exploring Data Labeling Tools, covers various data labeling tools, including open source tools such as Label Studio, CVAT, pyOpenAnnotate, and Azure Machine Learning. It also includes a comparison of various data labeling tools for image, text, audio, and video data.
