Udacity Data Scientist Nanodegree Capstone Project
Sparkify is a fictional music streaming app created by Udacity. For this project we are given application data of sizes mini, medium and large. To make this project happen, we need machine learning solutions that are not only accurate, but also powerful enough to process large data fast, so I have used mini scale data (128 mb) that I have processed with Spark on IBM Cloud Pak for Data.
Please refer to this Medium post for details about this model: HERE
- Python 3.*
- Numpy
- Pandas
- Seaborn
- Matplotlib
- PySpark 2.4
- Sparkify_Model.ipynb : Main Notebook for both Analysis and Modeling Sparkify Data
- mini_sparkify_event_data.json.zip : Sparkify App Data provided by Udacity
- Panjapol Ampornratana