Practical Machine Learning Project
Contents
README.md
index.Rmd
- R Markdown script that does the following: downloads the data directly from website; cleans data by removing missing values and unnecessary variables; creates a training and testing set to run prediction algorithms; compares prediction algorithms and selects best model based on accuracy rate; predicts classes from smaller out-of-sample test set.index.html
- compiled HTML file.figure
- folder that contains 4 figures:baseline-model-1.png
illustrates the decision tree algorithm.plot-accuracy-rf-1.png
andplot-error-rf-1.png
show the cross-validation accuracy rate and the cross-validation error rate, respectively.varimp-rf-1.png
lists the top 10 most important predictors (variable importance) given by the random forest model.data
- folder with the csv files used in the analysis.
Getting Started
- Run the
index.Rmd
script to generate the predictions. - For more information on the data and study, go to: http://groupware.les.inf.puc-rio.br/har