Course project writeup for JHU's Practical Machine Learning Coursera course. After downloading, you can find a pre-compiled document named 'CourseProject.html' in the root directory of the master branch.
Alternatively, you can directly take a look at the same document online: http://pederpansen.github.io/coursera-practical_machine_learning/
Please note that I have submitted this update after the deadline, so technically it should not be graded... anyway:
I realized that I had not included a single word about cross-validation in my report, which is carried out by default in caret's "train" function. I added a paragraph clarifying that I actually did use cross-validation for fitting my model and for predicting the submitted answers, but that I did not use it when compiling the report, as cross validation significantly increases the overall runtime, so it was commented out in the report. This update was added to clarify this fact.