This repository contains the code used to investigate supervised learning algorithms for activity identification from sensor data. The data was downloaded from the UC Irvine Machine Learning Repository: http://archive.ics.uci.edu/ml/datasets/PAMAP2+Physical+Activity+Monitoring
The data_processing.py
should be run first. This script outputs the training data in the form of CSV's. In order to run this script, the PAMAP2_Dataset folder should be placed in the same directory as the data_processing.py
script (or the script should be edited with the path to this folder).
The activity_identification_ML.ipynb
file is an iPython Notebook file containing the machine learning analysis. It requires the X_train.csv
and Y_train.csv
files as inputs.