music_box_project

Users use this music_box app to listen to music. The project's goal is to predict whether a user will churn.

unpack_and_clean_files

This is used to unzip log files, add file names and combine them into a whole log file.

Data Exploration

Have a look at what are included in the log data.

music_box_project

1. Use pyspark on google cloud platform to load and prepare data.
2. Fit logistic regression, xgboost and randorm forest models with grid search.