spobendonis/itu_predictive_maintenance

Industrial Software Engineering 2023 group 3: camelCase crew

Requirments

Docker and docker-compose: https://docs.docker.com/engine/install/ubuntu/

GIT LFS: https://git-lfs.com/

Description

This repository contains three components.

A postgres database

The database is used as a backend for storing data for the data generator

A data generator REST service

The REST services provides log messages upon make a regquest.

A REST endpoint for prediction

The REST endpoint for prediction takes a json as input and predicts the probability that the harddrive is failing. Optinally a threshold for anomalies can be provided.

Running the code

Ensure that GIT LFS is installed and that the CSV file is downloaded propperly. If the CSV only contains a SHA pointer, run the following commands:

git lfs fetch
git lfs checkout

From the root of the project (same folder as the readme), run

docker-compose -f docker/docker-compose.yaml up   

Alternatively you can use the make command (requires make)

make compose_up

It can take several minutes for the services to be ready. Go grab a cup of coffe!

Acessing the data generator

By default the data generator listen on localhost:8000
Swagger ui is provided on localhost:8000/docs

Acessing the prediction service

The prediction rest services listens on localhost:8001
Swagger ui is provided on localhost:8001/docs