Book Image

Machine Learning for Time-Series with Python

By : Ben Auffarth
Book Image

Machine Learning for Time-Series with Python

By: Ben Auffarth

Overview of this book

The Python time-series ecosystem is huge and often quite hard to get a good grasp on, especially for time-series since there are so many new libraries and new models. This book aims to deepen your understanding of time series by providing a comprehensive overview of popular Python time-series packages and help you build better predictive systems. Machine Learning for Time-Series with Python starts by re-introducing the basics of time series and then builds your understanding of traditional autoregressive models as well as modern non-parametric models. By observing practical examples and the theory behind them, you will become confident with loading time-series datasets from any source, deep learning models like recurrent neural networks and causal convolutional network models, and gradient boosting with feature engineering. This book will also guide you in matching the right model to the right problem by explaining the theory behind several useful models. You’ll also have a look at real-world case studies covering weather, traffic, biking, and stock market data. By the end of this book, you should feel at home with effectively analyzing and applying machine learning methods to time-series.
Table of Contents (15 chapters)
Other Books You May Enjoy

Feature Transforms

Many models or training processes depend on the assumption that the data is distributed according to the normal distribution. Even the most widely used descriptors, the arithmetic mean and standard deviation, are largely useless if your dataset has a skew or several peaks (multi-modal). Unfortunately, observed data often doesn't fall within the normal distribution, so that traditional algorithms can yield invalid results.

When data is non-normal, transformations of data are applied to make the data as normal-like as possible and, thus, increase the validity of the associated statistical analyses.

Often it can be easier to eschew traditional machine learning algorithms of dealing with time-series data and, instead, use newer, so-called non-linear methods that are not dependent on the distribution of the data.

As a final remark, while all the following transformations and scaling methods can be applied to features directly, an interesting spin with...