This project aims to build a end to end system to predict hard disk failures in a datacenter based on the disk parameters collected daily. The data comes from the backlaze hard disk statistics published here
https://www.backblaze.com/b2/hard-drive-test-data.html
The project aims to better the scores obtained in http://www.kdd.org/kdd2016/papers/files/adf0849-botezatuA.pdf