Preface
As a business analyst or data scientist, you will have to use many algorithms and approaches to prepare, process, and build machine learning (ML)-based applications by leveraging time series data, but you will likely face problems such as not knowing which algorithm to choose or how to combine and interpret these algorithms. Amazon Web Services (AWS) provides numerous services to help you build applications fueled by artificial intelligence (AI) capabilities. This book will see you get to grips with three AWS AI/ML managed services and enable you to deliver your desired business outcomes.
The book begins with Amazon Forecast, where you'll discover how to use time series forecasting and sophisticated statistical and ML algorithms to deliver business outcomes accurately. You'll then learn how to use Amazon Lookout for Equipment to build multivariate time series anomaly detection models geared toward industrial equipment and discover how it provides valuable insights to support teams with predictive maintenance and predictive quality use cases. In the final chapters, you'll explore Amazon Lookout for Metrics and automatically detect and diagnose outliers in your business and operational data.
By the end of this AWS book, you'll have learned how to use the three AWS AI services effectively to perform time series analysis.
Who this book is for
This book is for data analysts, business analysts, and data scientists looking to analyze time series data effectively to solve business problems. Basic statistics knowledge would be useful, but no ML knowledge is needed to follow the content in this book. Prior experience with time series data and how they relate to various business problems will help you get the most out of this book. This book will help ML practitioners find new ways to leverage their skills to build effective time series-based applications with the technologies described.
What this book covers
Chapter 1, An Overview of Time Series Analysis, will establish how time series data is different from regular tabular data and will hint at why we need different approaches to process them. You will also learn about the different types of time series you may encounter and gain an understanding of the kind of predictive power your time series data contains.
Chapter 2, An Overview of Amazon Forecast, will teach you about what Amazon Forecast is designed for, how it works, and the kinds of situations it is suited for. By the end of the chapter, you will also have a good command of the underlying concepts of Amazon Forecast, such as dataset groups, datasets, predictors, and forecasts.
Chapter 3, Creating a Project and Ingesting Your Data, will describe how to create and organize multiple datasets within the dataset group construct, how to configure datasets, and how to ingest CSV data. You will also gain a high-level understanding of all the heavy lifting Amazon Forecast performs on your behalf to save as much data preparation effort as possible.
Chapter 4, Training a Predictor with AutoML, will use the datasets prepared and ingested previously to train a forecasting model. You will learn how to configure training and discover what impact each feature can have on the training duration and outputs. The evaluation dashboard will be described in depth.
Chapter 5, Customizing Your Predictor Training, will go deeper into the different possible configurations that Amazon Forecast has to offer, after you have trained your first predictor using the automated features provided by AWS. From choosing the right algorithm for a given problem to leveraging supplementary features such as weather data, you will learn how you can increase the accuracy of your forecasts while optimizing your training time.
Chapter 6, Generating New Forecasts, will help you generate new forecasts and get new insights to support your daily business decisions, by leveraging the predictors you trained previously. This chapter will help you actually generate forecasts, download the results, and visualize them using your favorite spreadsheet software.
Chapter 7, Improving and Scaling Your Forecast Strategy, will help you get the most from Amazon Forecast. This chapter will point you in the right direction to monitor your models and compare predictions to real-life data, a crucial task to detect any drift in performance, which could trigger retraining. Last but not least, you will also leverage a sample from the AWS Solutions Library to automate your predictor training, forecast generation, and dashboard visualization.
Chapter 8, An Overview of Amazon Lookout for Equipment, will describe what Amazon Lookout for Equipment can do, how it works, and the kind of applications it's suited for. You will understand at a high level how to prepare your dataset and how you can integrate service results into your business processes.
Chapter 9, Creating a Dataset and Ingesting Your Data, will teach you how to create and organize multiple datasets and how to perform dataset ingestion. You will also gain a high-level understanding of all the heavy lifting AWS performs on your behalf to save as much data preparation effort as possible (in terms of imputation, time series alignment, resampling, and so on).
Chapter 10, Training and Evaluating a Model, will have you use the datasets prepared and ingested previously to train a multivariate anomaly detection model. You will learn about how to configure training and what impact each feature can have on the training output and the training duration. The evaluation and diagnostics dashboard will be described in depth to help you get a good view of the quality of the output.
Chapter 11, Scheduling Regular Inferences, will show you how to configure and run an inference scheduler that will run your data against your trained model. In this chapter, you will learn how to manage such schedulers and how to use the predictions obtained.
Chapter 12, Reducing Time to Insights for Anomaly Detections, will help you improve your model performance and go further in results post-processing. This chapter will also point you in the right direction when it comes to monitoring your models and detecting any drift, which would trigger either retraining or further investigation.
Chapter 13, An Overview of Amazon Lookout for Metrics, will explain what Amazon Lookout for Metrics is designed for, how it works, and the kind of situations it is suited for. By the end of this chapter, you will also have a good command of the underlying concepts of Amazon Lookout for Metrics (datasources, datasets, detectors, alerts, and anomalies).
Chapter 14, Creating and Activating a Detector, will describe the process of creating and activating a detector. You will also learn about the different integration paths that are available to connect Amazon Lookout for Metrics to various data sources and alerts.
Chapter 15, Viewing Anomalies and Providing Feedback, starts with a trained detector and shows you how to dive into detected anomalies and review them, as well as covering other key concepts, such as severity thresholds, how to leverage the impact analysis dashboard to perform root cause analysis, and how to provide feedback to the service.
To get the most out of this book
To get the most out of this book, you must have some basic knowledge about what time series data is and how it can relate to various business problems (sales or demand forecasting, predictive maintenance, and so on). You must also have an understanding of what an anomaly is when dealing with time series data.
More generally, having a high-level understanding of what cloud computing is and how you can leverage cloud services to solve your daily business problems is a good starting point.
To access the cloud services described in this book, you will need an AWS account. Account creation is covered in the Technical requirements section of Chapter 2, An Overview of Amazon Forecast.
If you are using the digital version of this book, we advise you to type the code yourself or access the code from the book's GitHub repository (a link is available in the next section). Doing so will help you avoid any potential errors related to the copying and pasting of code.
Download the example code files
You can download the example code files for this book from GitHub at https://github.com/PacktPublishing/Time-series-Analysis-on-AWS. If there's an update to the code, it will be updated in the GitHub repository.
We also have other code bundles from our rich catalog of books and videos available at https://github.com/PacktPublishing/. Check them out!
Download the color images
We also provide a PDF file that has color images of the screenshots and diagrams used in this book. You can download it here: https://static.packt-cdn.com/downloads/9781801816847_ColorImages.pdf.
Conventions used
There are a number of text conventions used throughout this book.
Code in text: Indicates code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles. Here is an example: "Mount the downloaded WebStorm-10*.dmg disk image file as another disk in your system."
A block of code is set as follows:
START = '2013-06-01'
END = '2013-07-31'
DATASET = 'household_energy_consumption'
FORECAST_PREFIX = 'export_energy_consumption_XXXX'
Any command-line input or output is written as follows:
python3 -m pip install --quiet s3fs pandas
wget https://raw.githubusercontent.com/PacktPublishing/
Time series-analysis-on-AWS-/main/Chapter09/create_
schema.py
Bold: Indicates a new term, an important word, or words that you see onscreen. For instance, words in menus or dialog boxes appear in bold. Here is an example: "Click on Amazon Forecast to go to the service home page."
Tips or Important Notes
Note that successive import jobs are not aggregated: only the most recent import will be considered when you use your datasets to train a predictor.
Get in touch
Feedback from our readers is always welcome.
General feedback: If you have questions about any aspect of this book, email us at [email protected] and mention the book title in the subject of your message.
Errata: Although we have taken every care to ensure the accuracy of our content, mistakes do happen. If you have found a mistake in this book, we would be grateful if you would report this to us. Please visit www.packtpub.com/support/errata and fill in the form.
Piracy: If you come across any illegal copies of our works in any form on the internet, we would be grateful if you would provide us with the location address or website name. Please contact us at [email protected] with a link to the material.
If you are interested in becoming an author: If there is a topic that you have expertise in and you are interested in either writing or contributing to a book, please visit authors.packtpub.com.
Share Your Thoughts
Once you've read Time series analysis on AWS, we'd love to hear your thoughts! Please click here to go straight to the Amazon review page for this book and share your feedback.
Your review is important to us and the tech community and will help us make sure we're delivering excellent quality content.