Human-Activity-Prediction

This document aims to predict how accurately “Unilateral Dumbbell Biceps Curl” is performed by the user. Wearable sensors comprised of Inertial measurement units (IMU) provides the data required. In this document machine learning algorithms rpart, gradient boosting and random forests were used to predict/classify the activity. This document also discusses the model built, cross-validation and out of sample errors.