Skip to content

A comparison of various ensemble machine learning algorithms (XGboost, random forest, ranger) to predict accelerometers

Notifications You must be signed in to change notification settings

moritzkoerber/ensemble_machine_learning_comparison

Repository files navigation

Files for this blog post: https://moritzkoerber.github.io/analysis/r/2019/07/24/blogpost/

Background

Going to the gym, whether to boost your health, to lose weight, or simply because it is fun, is certainly a worthwile activity. FiveThirtyEight recently reported that according to the latest Physical Activity Guidelines for Americans, every form of activity counts. However, if you have eager goals, not only quantity but also quality and being efficient matters. In this project, I predict whether a certain exercise, a barbell lift, was well or sloppily executed on the basis of data obtained from accelerometers on the belt, forearm, arm, and dumbell of six participants. The participants performed the exercises correctly and incorrectly in five different ways.

The idea for this analysis stems from the Coursera course Practical Machine Learning by Johns Hopkins University. The data for this project come from http://groupware.les.inf.puc-rio.br/har.

About

A comparison of various ensemble machine learning algorithms (XGboost, random forest, ranger) to predict accelerometers

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages