Machine Learning for Time Series Forecasting with Python

Оглавление

Francesca Lazzeri. Machine Learning for Time Series Forecasting with Python

Table of Contents

List of Tables

List of Illustrations

Guide

Pages

Machine Learning for Time Series Forecasting with Python®

Introduction

What Does This Book Cover?

Reader Support for This Book

Companion Download Files

How to Contact the Publisher

How to Contact the Author

CHAPTER 1 Overview of Time Series Forecasting

Flavors of Machine Learning for Time Series Forecasting

Supervised Learning for Time Series Forecasting

Python for Time Series Forecasting

Experimental Setup for Time Series Forecasting

Conclusion

CHAPTER 2 How to Design an End-to-End Time Series Forecasting Solution on the Cloud

Time Series Forecasting Template

Business Understanding and Performance Metrics

Data Ingestion

Data Exploration and Understanding

Data Pre-processing and Feature Engineering

Modeling Building and Selection

An Overview of Demand Forecasting Modeling Techniques

Model Evaluation

Model Deployment

Forecasting Solution Acceptance

Use Case: Demand Forecasting

Conclusion

CHAPTER 3 Time Series Data Preparation

Python for Time Series Data

Common Data Preparation Operations for Time Series

Time stamps vs. Periods

Converting to Time stamps

Providing a Format Argument

Indexing

Time/Date Components

Frequency Conversion

Time Series Exploration and Understanding

How to Get Started with Time Series Data Analysis

Data Cleaning of Missing Values in the Time Series

Time Series Data Normalization and Standardization

Time Series Feature Engineering

Date Time Features

Lag Features and Window Features

Rolling Window Statistics

Expanding Window Statistics

Conclusion

CHAPTER 4 Introduction to Autoregressive and Automated Methods for Time Series Forecasting

Autoregression

Moving Average

Autoregressive Moving Average

Autoregressive Integrated Moving Average

Automated Machine Learning

Conclusion

CHAPTER 5 Introduction to Neural Networks for Time Series Forecasting

Reasons to Add Deep Learning to Your Time Series Toolkit

Deep Learning Neural Networks Are Capable of Automatically Learning and Extracting Features from Raw and Imperfect Data

Deep Learning Supports Multiple Inputs and Outputs

Recurrent Neural Networks Are Good at Extracting Patterns from Input Data

Recurrent Neural Networks for Time Series Forecasting

Recurrent Neural Networks

Long Short-Term Memory

Gated Recurrent Unit

How to Prepare Time Series Data for LSTMs and GRUs

How to Develop GRUs and LSTMs for Time Series Forecasting

Keras

TensorFlow

Univariate Models

Multivariate Models

Conclusion

CHAPTER 6 Model Deployment for Time Series Forecasting

Experimental Set Up and Introduction to Azure Machine Learning SDK for Python

Workspace

Experiment

Run

Model

Compute Target, RunConfiguration, and ScriptRunConfig

Image and Webservice

Machine Learning Model Deployment

How to Select the Right Tools to Succeed with Model Deployment

Solution Architecture for Time Series Forecasting with Deployment Examples

Train and Deploy an ARIMA Model

Configure the Workspace

Create an Experiment

Create or Attach a Compute Cluster

Upload the Data to Azure

Create an Estimator

Submit the Job to the Remote Cluster

Register the Model

Deployment

Define Your Entry Script and Dependencies

Automatic Schema Generation

Conclusion

References

Index

About the Author

About the Technical Editor

Acknowledgments

WILEY END USER LICENSE AGREEMENT

Отрывок из книги

Francesca Lazzeri, PhD

Currently, most of the resources and tutorials for machine learning model-based time series forecasting generally fall into two categories: code demonstration repo for certain specific forecasting scenarios, without conceptual details, and academic-style explanations of the theory behind forecasting and mathematical formula. Both of these approaches are very helpful for learning purposes, and I highly recommend using those resources if you are interested in understanding the math behind theoretical hypotheses.

.....

In the next section of this chapter, we will discuss how to shape time series as a supervised learning problem and, as a consequence, get access to a large portfolio of linear and nonlinear machine learning algorithms.

Machine learning is a subset of artificial intelligence that uses techniques (such as deep learning) that enable machines to use experience to improve at tasks (aka.ms/deeplearningVSmachinelearning). The learning process is based on the following steps:

.....