This repository demonstrates a few techniques for using argo and docker in a machine learning workflow.
The /code
directory contains some source code in python to
train and serve a scikit-learn model.
The /workflows
directory contains argo workflows and templates
to demonstrate a kubernetes based ML pipeline (although admittely over simplified).
In order to run, you'll need access to a kubernetes cluster. You can start a local cluster using minikube or docker-for-mac. You could also use a cloud managed kubernetes service such as Google's GKE.
To run this workflow, you'll need to install Argo.
In the templates with s3 configured artifacts, you will need to change the s3 paths to your own bucket. Finally, you need to configure secrets in the kubernetes cluster to access the bucket.