Mood based music recommendation

Written in python, with SQLite3, Flask and Tensorflow

Install

git clone https://github.com/porridgewithraisins/mood-music-recommendation.git
cd mood-music-recommendation
unzip csv_files.zip -d flaskApp/csv_files

Then, start a virtual environment (recommended as machine learning packages are thicc)

python3 -m venv venv
source venv/bin/activate

Then install dependencies,

pip install -r requirements.txt

Then start the app on a local server,

export FLASK_APP=flaskr
export FLASK_ENV=development
flask run

Works like:
Upload image of your face.
It analyses your mood from it, and gives you a playlist of songs matching your mood.
The face-to-mood bit is done with FER
The song-to-mood bit was done with tensorflow. I used this dataset at Kaggle

Demo:

You can also perform a fuzzy search of the entire database