music-rec | Music Recommendation Service | Machine Learning library

music-rec is a Python library typically used in Telecommunications, Media, Media, Entertainment, Artificial Intelligence, Machine Learning, Deep Learning, Tensorflow, Keras, Docker applications. music-rec has no bugs, it has no vulnerabilities, it has build file available and it has low support. You can download it from GitHub.

This project is a music recommendation service. It draws from the Million Song Dataset, and attempts to recommend songs based on a list of liked tracks from the user. The recommendation engine is a deep learning artificial neural network implemented with Keras and TensorFlow. The model is run using Flask to host middleware APIs, and the frontend is based on Skeleton. To get the dataset, you can either run setup.sh or the following commands. Alternatively, visit the Million Song Dataset's site to download the file and unzip it manually. Training the neural network may take some trial and error based on the hardware available. Running ./lib/main.py will attempt to train the network using the full dataset and default hyperparameters. Once the network is trained and a workable model is created, you can run the web app that recommends songs. First, move your best working model (and all the associated files) into ./model/working/. Then, to standup web app, first run. This will serve the static site html pages. In another terminal window, run. This will stand up the middleware api endpoints. Then visit the static site at and try submitting some songs for recommendation.