The COMP project for Allegheny College Computer Science department class of 2020. This project implemented a system to recommend music tracks based on user-submitted texts. The text can be the form of diary or video transcribes, which could contain users’ current emotions. The recommended music can be got by matching the text with the song lyrics. By using the system, people can fill their emotional needs whenever they want. The website of the system is music.xingbangliu.io.
This application was built on Linux system, the Python version was 3.7.6. Pyenv is recommended for Python version management. Here is a good tutorial on how to install Pyenv on your system.
Pipenv was used to manage the Python dependencies, here is a good tutorial on how to install and use Pipenv.
To install the dependencies, use pipenv install to install from Pipfile. If you
want to install by importing the requirement.txt, use
pipenv install -r path/to/requirements. txt to install.
The complete dependency list is as follows, please also refer to Pipfile and requirement.txt.
lyricsgenius = "==1.8.2"
spacy = "==2.2.4"
networkx = "==2.4"
graphviz = "==0.13.2"
pytextrank = "==2.0.1"
numpy = "==1.18.2"
pip = "*"
six = "==1.14.0"
py-stringmatching = "==0.4.1"
spacy-langdetect = "==0.1.2"
attrs = "==19.3.0"
bcrypt = "==3.1.7"
beautifulsoup4 = "==4.6.0"
blis = "==0.4.1"
catalogue = "==1.0.0"
certifi = "==2019.11.28"
cffi = "==1.14.0"
chardet = "==3.0.4"
click = "==7.1.1"
coverage = "==5.0.4"
cymem = "==2.0.3"
decorator = "==4.4.2"
idna = "==2.9"
importlib-metadata = "==1.5.0"
itsdangerous = "==1.1.0"
langdetect = "==1.0.7"
more-itertools = "==8.2.0"
murmurhash = "==1.0.2"
plac = "==1.1.3"
pluggy = "==0.13.1"
preshed = "==3.0.2"
py = "==1.8.1"
pycparser = "==2.20"
pyparsing = "==2.4.6"
pytest = "==5.4.1"
requests = "==2.23.0"
srsly = "==1.0.2"
thinc = "==7.4.0"
tqdm = "==4.43.0"
urllib3 = "==1.25.8"
wasabi = "==0.6.0"
wcwidth = "==0.1.8"
zipp = "==3.1.0"
Flask = "==1.1.1"
Flask-WTF = "==0.14.3"
Flask-SQLAlchemy = "==2.4.1"
Flask-Bcrypt = "==0.7.1"
Flask-Login = "==0.5.0"
Jinja2 = "==3.0.0a1"
MarkupSafe = "==1.1.1"
SQLAlchemy = "==1.3.15"
Werkzeug = "==1.0.0"
WTForms = "==2.2.1"
gunicorn = "*"
pytest-cov = "*"
Use pipenv run test to test the application. The test files are under music_sug/test/
directory. The bash script is music_sug/scripts/test.sh.
To deploy the project, first install the dependencies, then use pipenv run setup
to install the language model, generate the database, and commit the lyrics model.
The options for the web server could be Nginx and gunicorn.
Install nginx by using apt install nginx.
A simple Nginx proxy configuration under /etc/nginx/sites-enabled could be:
server {
listen 80;
server_name Your_IP_adress;
location / {
proxy_pass http://localhost:8000;
include /etc/nginx/proxy_params;
proxy_redirect off;
}
}
After setting up the configuration file, use systemctl restart nginx to restart
the service.
Gunicorn is responsible for handling the Python code, it should be installed with
Pipfile already.
The app can be run with gunicorn -w number_of_workers name_of_main_program:app.
The number of worker are defined as (2 * number_of_CPU_cores) + 1. The name of
main program in this project is app, so use app:app in the command.
Supervisor was used to monitor and manage the processes. To install supervisor,
use apt install supervisor.
Under /etc/supervisor/, create a configuration file, a simple sample can be:
[program:name_of_project]
directory=path/to/name_of_project
command=path/to/project/virtual/environment/bin/gunicorn -w number_of_workers name_of_main_program:app
user=system_user
autostart=true
autorestart=true
stopasgroup=true
killasgroup=true
stderr_logfile=/var/log/music_sug/music_sug.err.log
stdout_logfile=/var/log/music_sug/music_sug.out.log
Under /var/log/ create a directory for your project logs. In the example above,
the files music_sug.err.log and music_sug.out.log should be created.
Reload the configuration by using supervisorctl reload, now the application should
be up and running.
This project is licensed under the MIT License - see the LICENSE.md file for details.