https://github.com/risdorn/predicting-drive-failure

Predicting Hard Drive Failures attempts to tackle this problem via the use of different Machine Learning models. In this repo I have used various models for classifcation as well as regression, this is easily done via the use of PyCaret.
https://github.com/risdorn/predicting-drive-failure

hard-drive-protection machine-learning

Last synced: 8 days ago
JSON representation

Predicting Hard Drive Failures attempts to tackle this problem via the use of different Machine Learning models. In this repo I have used various models for classifcation as well as regression, this is easily done via the use of PyCaret.

• Host: GitHub
• URL: https://github.com/risdorn/predicting-drive-failure
• Owner: Risdorn
• License: mit
• Created: 2024-04-02T05:49:26.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-04-29T09:41:06.000Z (8 months ago)
• Last Synced: 2024-11-08T07:14:48.556Z (2 months ago)
• Topics: hard-drive-protection, machine-learning
• Language: Jupyter Notebook
• Homepage:
• Size: 3.64 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        

# Predicting Drive Failure

This code goes through 3 different methodologies for predicting drive failures.

## Dataset Features used


capacity_bytes = Capacity of the Hard Drive

smart_5_normalized and smart_5_raw = Reallocated Sectors Count 

smart_187_normalized and smart_187_raw = Reported Uncorrectable Errors 

smart_188_normalized and smart_188_raw = Command Timeout 	

smart_197_normalized and smart_197_raw = Current Pending Sectors Count	

smart_198_normalized and smart_198_raw = Offline Uncorrectable Sectors Count	

date_diff = Number of days left till failure

failure = Whether hard drive has failed or not



## How to Run

The code can be run by opening the .ipynb file in colab and running it.

For running it locally, the file path will have to be changed accordingly.

## Data

Data has been taken from BackBlaze. We are using data from the year 2016.

https://www.backblaze.com/cloud-storage/resources/hard-drive-test-data

# References

https://www.backblaze.com/blog/using-machine-learning-to-predict-hard-drive-failures/

https://en.wikipedia.org/wiki/Self-Monitoring,_Analysis_and_Reporting_Technology#ATA_S.M.A.R.T._attributes

https://medium.com/geekculture/a-complete-solution-to-the-backbaze-com-kaggle-problem-cf1fab1af529