https://github.com/double-em/maintrend

Maintenance date predictor developed as a school project for my Computer Science test
https://github.com/double-em/maintrend

Last synced: 2 months ago
JSON representation

Maintenance date predictor developed as a school project for my Computer Science test

• Host: GitHub
• URL: https://github.com/double-em/maintrend
• Owner: double-em
• Created: 2020-04-22T18:17:42.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2023-12-15T11:41:27.000Z (over 1 year ago)
• Last Synced: 2025-01-19T11:14:19.948Z (4 months ago)
• Language: Python
• Homepage: https://double-em.github.io/
• Size: 35.8 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Maintrend

![Travis (.com)](https://img.shields.io/travis/com/double-em/Maintrend?label=Travis%20CI%20Tests)


![Predictor build](https://github.com/Zxited/TrendlogDataApp/workflows/Predictor%20build/badge.svg?branch=master)


![Serving API Build and Publish to GKE](https://github.com/Zxited/Maintrend/workflows/Serving%20API%20Build%20and%20Publish%20to%20GKE/badge.svg?branch=master)


![Uptime Robot ratio (30 days)](https://img.shields.io/uptimerobot/ratio/m784993822-a76d37ac3e6259c2a679aebb?label=Serving%20API)

![Uptime Robot ratio (30 days)](https://img.shields.io/uptimerobot/ratio/m784993836-849c5728c21c6df110e3e605?label=Predictor%20Service)

Maintrend is a maintenance date predictor developed as a school project for my Computer Science semester exam.

The predictor uses a RNN architecture with LSTM cells. The mean absolute loss for never before seen data is around 0.3 days with a maximum absolute loss at around 1 - 4 days.