Book Image

Cloud Analytics with Google Cloud Platform

By : Sanket Thodge
Book Image

Cloud Analytics with Google Cloud Platform

By: Sanket Thodge

Overview of this book

With the ongoing data explosion, more and more organizations all over the world are slowly migrating their infrastructure to the cloud. These cloud platforms also provide their distinct analytics services to help you get faster insights from your data. This book will give you an introduction to the concept of analytics on the cloud, and the different cloud services popularly used for processing and analyzing data. If you’re planning to adopt the cloud analytics model for your business, this book will help you understand the design and business considerations to be kept in mind, and choose the best tools and alternatives for analytics, based on your requirements. The chapters in this book will take you through the 70+ services available in Google Cloud Platform and their implementation for practical purposes. From ingestion to processing your data, this book contains best practices on building an end-to-end analytics pipeline on the cloud by leveraging popular concepts such as machine learning and deep learning. By the end of this book, you will have a better understanding of cloud analytics as a concept as well as a practical know-how of its implementation
Table of Contents (16 chapters)
Title Page
Packt Upsell
Foreword
Contributors
Preface
Index

Google Cloud Machine Learning Engine


This is an API that creates a model in machine learning, and can work on any size and any type of data. A major use of this ML is to train a model and predict from it. The ML engine can use any model to perform large-scale analysis on a cluster for managing online and batch programming. It can support a few thousand users and performs on terabytes of data. This service can easily be combined with other services such as Dataflow, Storage, BigQuery, and so on provided by GCP. The ML Engine lets users build models using DataLab and can also build portable models that work on various devices.

The purpose of Cloud ML Engine is to train a new ML model at scale using the TensorFlow application, and the model is hosted to get predictions on a new set of data.

The ML Engine Workflow can be formatted into the following steps:

  1. Evaluating the problem
  2. Data exploration and preparation
  3. Model development and training
  4. Model testing and deployment
  5. Operational development and...