Introduction to Time Series Forecasting with Python: How to Prepare Data and Develop Models to Predict the Future (version 1.9)

id: 646133 VIP Open in MaM

By Jason Brownlee

Language: English

Category: Ebook - Computer/Internet

Categories: Computer/Internet

Size: 6.4 MiB pdf Files: 1

Uploader: SaltGirl

Uploaded At: 2020-08-04 07:41:03

Machine Learning Master, 2020; Developer focused; YOU MUST KNOW PYTHON; no maths or scikit prior knowledge needed though helpful

Description

Time series forecasting is different from other machine learning problems.

The key difference is the fixed sequence of observations and the constraints and additional structure this provides.

In this mega Ebook written in the friendly Machine Learning Mastery style that you’re used to, finally cut through the math and specialized methods for time series forecasting.

Using clear explanations, standard Python libraries and step-by-step tutorials you will discover how to load and prepare data, evaluate model skill, and implement forecasting models for time series data.



Book Details
Time Series Problems are Important
Time series forecasting is an important area of machine learning that is often neglected.

It is important because there are so many prediction problems that involve a time component. These problems are neglected because it is this time component that makes time series problems more difficult to handle.

You can’t just fire a machine learning algorithm at a time series dataset.

Time series data must be transformed into a supervised learning problem.
Time series data has temporal structure like trends and seasonality that must be handled.
Time series data has a forecast horizon.
There are a few conceptual steps you must make before you can start developing forecasting models.

There are also specialized terminology and algorithms to consider and use when working with time series data.

It can feel overwhelming for a beginner and standard machine learning libraries like scikit-learn do not make it easy to get started.

Introducing: “Time Series Forecasting With Python“
This is the book I wish I had when I was getting started with univariate time series forecasting.

It is designed for the practical and hands-on way you prefer to learn.

The goal of this book is to:

Show you how to get results on univariate time series forecasting problems using the Python ecosystem.

It is a cookbook designed for immediate use.

This book was developed using five principles.

They are:

Application: The focus is on the application of forecasting rather than the theory.
Lessons: The book is broken down into short lessons, each focused on a specific topic.
Value: Lessons focus on the most used and most useful aspects of a forecasting project.
Results: Each lesson provides a path to a usable and reproducible result.
Speed: Each lesson is designed to provide the shortest path to a result.
These principles shape the structure and organization of the book.

What You Will Know and Be Able to Do (Reading Outcomes)
If you choose to work through all of the lessons and projects of this book, you can set some reasonable expectations on your new found capabilities.

They are:

Time Series Foundations: You will be able to identify time series forecasting problems as distinct from other predictive modeling problems and how time series can be framed as supervised learning.
Transform Data For Modeling: You will be able to transform, rescale, smooth and engineer features from time series data in order to best expose the underlying inherent properties of the problem (the signal) to learning algorithms for forecasting.
Harness Temporal Structure: You will be able to analyze time series data and understand the temporal structure inherent in it such as trends and seasonality and how these structures may be addressed, removed and harnessed when forecasting.
Evaluate Models: You will be able to devise a model test harness for a univariate forecasting problem and estimate the baseline skill and expected model performance on unseen data with various performance measures.
Apply Classical Methods: You will be able to select, apply and interpret the results from classical linear methods such as Autoregression, Moving Average and ARIMA models on univariate time series forecasting problems.
You will be a capable predictive modeler for univariate time series forecasting problems using the Python ecosystem.

‘Time Series Forecasting With Python‘ is for Python Developers…
This book makes some assumptions about you.

They are:

You’re a Developer: This is a book for developers. You are a developer of some sort. You know how to read and write code. You know how to develop and debug a program.
You know Python: This is a book for Python people. You know the Python program- ming language, or you’re a skilled enough developer that you can pick it up as you go along.
You know some Machine Learning: This is a book for novice machine learning practitioners. You know some basic practical machine learning, or you can figure it out quickly.
No mathematical prerequisites are needed.
No scikit-learn prerequisites are needed.

This is a playbook, a cookbook, a field guide, not a textbook for academics.

Time Series Forecasting for Beginners
It is an introductory book for time series forecasting.

As such, it focuses on univariate (one variable) data, rather than more complex multivariate problems. It also focuses on using powerful linear methods like ARIMA, rather than more exotic methods.

Other Torrents