You're reading from Codeless Time Series Analysis with KNIME A practical guide to implementing forecasting models for time series analysis applications

Product type Paperback

Published in Aug 2022

Publisher Packt

ISBN-13 9781803232065

Length 392 pages

Edition 1st Edition

Languages

Java

Tools

Knime

Concepts

Data Analysis

Authors (4):

Daniele Tonini

Maarit Widmann

Corey Weisinger

KNIME AG

View More author details

Table of Contents (20) Chapters

Preface

1. Part 1: Time Series Basics and KNIME Analytics Platform

2. Chapter 1: Introducing Time Series Analysis FREE CHAPTER

3. Chapter 2: Introduction to KNIME Analytics Platform

4. Chapter 3: Preparing Data for Time Series Analysis

5. Chapter 4: Time Series Visualization

6. Chapter 5: Time Series Components and Statistical Properties

7. Part 2: Building and Deploying a Forecasting Model

8. Chapter 6: Humidity Forecasting with Classical Methods

9. Chapter 7: Forecasting the Temperature with ARIMA and SARIMA Models

10. Chapter 8: Audio Signal Classification with an FFT and a Gradient-Boosted Forest

11. Chapter 9: Training and Deploying a Neural Network to Predict Glucose Levels

12. Chapter 10: Predicting Energy Demand with an LSTM Model

13. Chapter 11: Anomaly Detection – Predicting Failure with No Failure Examples

14. Part 3: Forecasting on Mixed Platforms

15. Chapter 12: Predicting Taxi Demand on the Spark Platform

16. Chapter 13: GPU Accelerated Model for Multivariate Forecasting

17. Chapter 14: Combining KNIME and H2O to Predict Stock Prices

18. Answers

19. Other Books You May Enjoy

Training and deployment

Training and deployment in KNIME Analytics Platform come in a variety of complexities; for this chapter, we’ll look at some of the simpler options. No matter how you plan to train your model, it’s good to first partition your data. To partition data in KNIME, we use the Partitioning node:

Figure 6.10 – The Partitioning node and configuration dialog

In Figure 6.10, we see the Partitioning node and its configuration dialog. You’ll notice the node has one input port to the left and two output ports to the right. The input port is our full dataset, and the two outputs are the two splits based on our configuration choices. Note that the top port aligns with the options in the configuration dialog.

There are two things to configure for this node: the size of the first partition and how it is created. A 70/30 split with 70% being for the training set is common practice, but this can vary by use case. For...

The rest of the chapter is locked

You're reading from Codeless Time Series Analysis with KNIME A practical guide to implementing forecasting models for time series analysis applications

Table of Contents (20) Chapters

Training and deployment

Unlock this book and the full library FREE for 7 days

Authors (4)

Personalised recommendations for you