Time Series Analysis on AWS: Learn how to build forecasting models and detect anomalies in your time series data
- Length: 458 pages
- Edition: 1
- Language: English
- Publisher: Packt Publishing
- Publication Date: 2022-03-04
- ISBN-10: 1801816840
- ISBN-13: 9781801816847
- Sales Rank: #3509600 (See Top 100 Books)
Leverage AWS AI/ML managed services to generate value from your time series data
Key Features
- Solve modern time series analysis problems such as forecasting and anomaly detection
- Gain a solid understanding of AWS AI/ML managed services and apply them to your business problems
- Explore different algorithms to build applications that leverage time series data
Book Description
Being a business analyst and data scientist, you have to use many algorithms and approaches to prepare, process, and build ML-based applications by leveraging time series data, but you face common problems, such as not knowing which algorithm to choose or how to combine and interpret them. Amazon Web Services (AWS) provides numerous services to help you build applications fueled by artificial intelligence (AI) capabilities. This book helps you get to grips with three AWS AI/ML-managed services to enable you to deliver your desired business outcomes.
The book begins with Amazon Forecast, where you’ll discover how to use time series forecasting, leveraging sophisticated statistical and machine learning algorithms to deliver business outcomes accurately. You’ll then learn to use Amazon Lookout for Equipment to build multivariate time series anomaly detection models geared toward industrial equipment and understand how it provides valuable insights to reinforce teams focused on predictive maintenance and predictive quality use cases. In the last 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 understood how to use the three AWS AI services effectively to perform time series analysis.
What you will learn
- Understand how time series data differs from other types of data
- Explore the key challenges that can be solved using time series data
- Forecast future values of business metrics using Amazon Forecast
- Detect anomalies and deliver forewarnings using Lookout for Equipment
- Detect anomalies in business metrics using Amazon Lookout for Metrics
- Visualize your predictions to reduce the time to extract insights
Who this book is for
If you’re a data analyst, business analyst, or data scientist looking to analyze time series data effectively for solving business problems, this is the book for you. Basic statistics knowledge is assumed, but no machine learning knowledge is necessary. Prior experience with time series data and how it relates to various business problems will help you get the most out of this book. This guide will also help machine learning practitioners find new ways to leverage their skills to build effective time series-based applications.
Time Series Analysis on AWS Contributors About the author About the reviewers Preface Who this book is for What this book covers To get the most out of this book Download the example code files Download the color images Conventions used Get in touch Share Your Thoughts Section 1: Analyzing Time Series and Delivering Highly Accurate Forecasts with Amazon Forecast Chapter 1: An Overview of Time Series Analysis Technical requirements What is a time series dataset? Recognizing the different families of time series Univariate time series data Continuous multivariate data Event-based multivariate data Multiple time series data Adding context to time series data Labels Related time series Metadata Learning about common time series challenges Technical challenges Data quality Visualization challenges Behavioral challenges Missing insights and context Selecting an analysis approach Using raw time series data Summarizing time series into tabular datasets Using imaging techniques Symbolic transformations Typical time series use cases Virtual sensors Activity detection Predictive quality Setpoint optimization Summary Chapter 2: An Overview of Amazon Forecast Technical requirements What kinds of problems can we solve with forecasting? Framing your forecasting problem What is Amazon Forecast? How does Amazon Forecast work? Amazon Forecast workflow overview Pricing Choosing the right applications Latency requirements Dataset requirements Use-case requirements Summary Chapter 3: Creating a Project and Ingesting Your Data Technical requirements Understanding the components of a dataset group Target time series Related time series Item metadata Preparing a dataset for forecasting purposes Preparing the raw dataset (optional) Uploading your data to Amazon S3 for storage Authorizing Amazon Forecast to access your S3 bucket (optional) Creating an Amazon Forecast dataset group Ingesting data in Amazon Forecast Ingesting your target time series dataset Ingesting related time series Ingesting metadata What's happening behind the scenes? Summary Chapter 4: Training a Predictor with AutoML Technical requirements Using your datasets to train a predictor How Amazon Forecast leverages automated machine learning Understanding the predictor evaluation dashboard Predictor overview Predictor metrics Exporting and visualizing your predictor backtest results What is backtesting? Exporting backtest results Backtest predictions overview Backtest accuracy overview Summary Chapter 5: Customizing Your Predictor Training Technical requirements Choosing an algorithm and configuring the training parameters ETS ARIMA NPTS Prophet DeepAR+ CNN-QR When should you select an algorithm? Leveraging HPO What is HPO? Training a CNN-QR predictor with HPO Introducing tunable hyperparameters with HPO Reinforcing your backtesting strategy Including holiday and weather data Enabling the Holidays feature Enabling the Weather index Implementing featurization techniques Configuring featurization parameters Introducing featurization parameter values Customizing quantiles to suit your business needs Configuring your forecast types Choosing forecast types Summary Chapter 6: Generating New Forecasts Technical requirements Generating a forecast Creating your first forecast Generating new subsequent forecasts Using lookup to get your items forecast Exporting and visualizing your forecasts Exporting the predictions Visualizing your forecast's results Performing error analysis Generating explainability for your forecasts Generating forecast explainability Visualizing forecast explainability Summary Chapter 7: Improving and Scaling Your Forecast Strategy Technical requirements Deep diving into forecasting model metrics Weighted absolute percentage error (WAPE) Mean absolute percentage error (MAPE) Mean absolute scaled error (MASE) Root mean square error (RMSE) Weighted quantile loss (wQL) Understanding your model accuracy Model monitoring and drift detection Serverless architecture orchestration Solutions overview Solutions deployment Configuring the solution Using the solution Cleanup Summary Section 2: Detecting Abnormal Behavior in Multivariate Time Series with Amazon Lookout for Equipment Chapter 8: An Overview of Amazon Lookout for Equipment Technical requirements What is Amazon Lookout for Equipment? What are the different approaches to tackle anomaly detection? What is an anomaly? Model-based approaches Other anomaly detection methods Using univariate methods with a multivariate dataset The challenges encountered with multivariate time series data How does Amazon Lookout for Equipment work? Defining the key concepts Amazon Lookout for Equipment workflow overview Pricing How do you choose the right applications? Latency requirements Dataset requirements Use case requirements Summary Chapter 9: Creating a Dataset and Ingesting Your Data Technical requirements Preparing a dataset for anomaly detection purposes Preparing the dataset Uploading your data to Amazon S3 for storage Creating an Amazon Lookout for Equipment dataset Generating a JSON schema Dataset schema structure Using CloudShell to generate a schema Creating a data ingestion job Understanding common ingestion errors and workarounds Wrong S3 location Component not found Missing values for a given time series Summary Chapter 10: Training and Evaluating a Model Technical requirements Using your dataset to train a model Training an anomaly detection model How is the historical event file used? Deep dive into the off-time detection feature Model organization best practices Choosing a good data split between training and evaluation Evaluating a trained model Model evaluation dashboard overview Interpreting the model performance dashboard's overview Using the events diagnostics dashboard Summary Chapter 11: Scheduling Regular Inferences Technical requirements Using a trained model Configuring a scheduler Preparing your Amazon S3 bucket Configuring your scheduler Preparing a dataset for inference Understanding the scheduled inference process Preparing the inference data Extracting the inference results Summary Chapter 12: Reducing Time to Insights for Anomaly Detections Technical requirements Improving your model's accuracy Reducing the number of signals Using the off-time conditions Selecting the best signals Processing the model diagnostics Deploying a CloudWatch-based dashboard Using the Lookout for Equipment dashboards Post-processing detected events results in building deeper insights Monitoring your models Orchestrating each step of the process with a serverless architecture Assembling and configuring the AWS components Summary Section 3: Detecting Anomalies in Business Metrics with Amazon Lookout for Metrics Chapter 13: An Overview of Amazon Lookout for Metrics Technical requirements Recognizing different types of anomalies What is Amazon Lookout for Metrics? How does Amazon Lookout for Metrics work? Key concept definitions Amazon Lookout for Metrics workflow overview Pricing Identifying suitable metrics for monitoring Dataset requirements Use case requirements Choosing between Lookout for Equipment and Lookout for Metrics Summary Chapter 14: Creating and Activating a Detector Technical requirements Preparing a dataset for anomaly detection purposes Collecting the dataset Uploading your data to Amazon S3 for storage Giving access to your data to Amazon Lookout for Metrics Creating a detector Adding a dataset and connecting a data source Understanding the backtesting mode Configuring alerts Summary Chapter 15: Viewing Anomalies and Providing Feedback Technical requirements Training a continuous detector Configuring a detector in continuous mode Preparing data to feed a continuous detector Reviewing anomalies from a trained detector Detector details dashboard Anomalies dashboard Interacting with a detector Delivering readable alerts Providing feedback to improve a detector Summary Why subscribe? Other Books You May Enjoy Packt is searching for authors like you Share Your Thoughts
Donate to keep this site alive
How to download source code?
1. Go to: https://github.com/PacktPublishing
2. In the Find a repository… box, search the book title: Time Series Analysis on AWS: Learn how to build forecasting models and detect anomalies in your time series data
, sometime you may not get the results, please search the main title.
3. Click the book title in the search results.
3. Click Code to download.
1. Disable the AdBlock plugin. Otherwise, you may not get any links.
2. Solve the CAPTCHA.
3. Click download link.
4. Lead to download server to download.