LYRA by Arighi Andrea and Riccardo Tosin SETUP INSTRUCTIONS after cloning this repository in a local folder: 1) Edit file config.py to fit your computer's configuration. 2) Download the dataset * Either clone the repository https://github.com/salvioner/lyra-dataset.git inside the "dataset" folder * Or regenerate the dataset from scratch (last tested: February 2016) - navigate to folder "dataset" - run "python fetchAlbums.py" and wait - run "python fetchSongs.py" and wait NOTE: this will take quite a long time, since the script will attempt to download artist names, album names and song lyrics for more than 10.000 songs. A working Internet connection is required for both scripts to work. 3) Download additional packages (I suggest the use of Anaconda distribution of Python, which includes almost all of the required packages: https://www.anaconda.com/what-is-anaconda/) * Package requirements for alg.py: vaderSentiment numpy glob json * Package requirements for dboperations.py: numpy matplotlib.pyplot sklearn.cluster glob * Package requirements for dataset scripts: BeautifulSoup4 (AKA: bs4) urllib2 codecs 4) For help, run: python start.py -h from the root folder (where this file is saved).