Machine learning project to predict the genre of a movie using the movie's soundtrack. Stemmed from an interest in the added storytelling power of music and the scoring of movies.
- loads audio buffers from audio files
- serializes audio buffers by movie
- serializes features
- serializes final dataframe
- generates the production model to deploy using dataframe created by process_data script
- train and test predictions of songs' genres
- train and test using an ensemble approach for the predictions of movies' genres
- gets the top 200 movies' names and meta data from each movie genre from IMDB