AI & ML

Song Popularity Prediction

An end-to-end ML pipeline joining 328K Billboard chart entries with 170K Spotify tracks, benchmarking KNN, SVM, Random Forest, and a PyTorch network — shipped as a CLI that rates any FLAC/MP3.

Pythonpandasscikit-learnPyTorchSpotify API
~9,500 songs modeled
Flight recording

See it run

CMPT353_ALLIN — four models, one FLAC
 

Spotify has retired the audio-features endpoint Predict.py calls, so the song’s features come from the response cached in the repo (same track). Training, evaluation and the saved neural net are the project’s own code.

Log photos

Frames from the run