https://github.com/alexioannides/bodywork-mlops-demo

Demonstrating how Bodywork can be used to deploy a simulation of the lifecycle of a train-and-serve ML pipeline, responding to new data undergoing concept drift.
https://github.com/alexioannides/bodywork-mlops-demo

aws data-science docker kubernetes machine-learning mlops numpy python scikit-learn

Last synced: 2 months ago
JSON representation

Demonstrating how Bodywork can be used to deploy a simulation of the lifecycle of a train-and-serve ML pipeline, responding to new data undergoing concept drift.

• Host: GitHub
• URL: https://github.com/alexioannides/bodywork-mlops-demo
• Owner: AlexIoannides
• Created: 2021-01-06T06:45:35.000Z (almost 4 years ago)
• Default Branch: master
• Last Pushed: 2022-09-30T20:12:08.000Z (about 2 years ago)
• Last Synced: 2023-02-28T18:18:52.597Z (almost 2 years ago)
• Topics: aws, data-science, docker, kubernetes, machine-learning, mlops, numpy, python, scikit-learn
• Language: Jupyter Notebook
• Homepage: https://bodywork.readthedocs.io/en/latest/
• Size: 541 KB
• Stars: 9
• Watchers: 1
• Forks: 0
• Open Issues: 3
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Simulating the Lifecycle of a ML Pipeline on Kubernetes

![bodywork](https://bodywork-media.s3.eu-west-2.amazonaws.com/ml_system_testing.png)

This repository contains a Bodywork machine learning project that simulates the lifecycle of a train-and-deploy pipeline responding to new data undergoing concept drift. Each day a new tranche of synthetic data is simulated and used to test a model deployed as a model-scoring service. The new data is then combined with historical data and used to train a new model that will be used for the following day.