https://github.com/wjayesh/mlflow-tracking-server

This repository hosts the code to make it easier to deploy a customizable and flexible MLflow tracking server solution to your Kubernetes cluster.
https://github.com/wjayesh/mlflow-tracking-server

kubernetes mlflow mlflow-tracking-server mlops

Last synced: 23 days ago
JSON representation

This repository hosts the code to make it easier to deploy a customizable and flexible MLflow tracking server solution to your Kubernetes cluster.

• Host: GitHub
• URL: https://github.com/wjayesh/mlflow-tracking-server
• Owner: wjayesh
• License: apache-2.0
• Created: 2022-06-16T11:08:55.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2022-07-01T05:47:33.000Z (over 2 years ago)
• Last Synced: 2024-04-16T12:44:32.024Z (9 months ago)
• Topics: kubernetes, mlflow, mlflow-tracking-server, mlops
• Language: Shell
• Homepage: https://dev.to/wjayesh/not-just-another-mlflow-on-kubernetes-article-2522
• Size: 11.7 KB
• Stars: 1
• Watchers: 1
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# MLflow Tracking Server on Kubernetes 🚀:  A customizable and flexible solution ✨

This repository hosts the code to make it easier to deploy an MLflow tracking server to your Kubernetes cluster. 

## Project Structure 🗺️

* `scripts` directory has the entrypoint that is executed when the image defined by the Dockerfile is run.

* The Dockerfile specifies the recipe for building the image. 

## Contributions Welcome! 👋🥰

The script inside the `scripts` folder currently has options for adding the metadata store and the artifact store.

It can be expanded to include the flags

- `--serve-artifacts` ([learn more](https://www.mlflow.org/docs/latest/tracking.html#scenario-5-mlflow-tracking-server-enabled-with-proxied-artifact-storage-access))

- `--artifacts-only` ([learn more](https://www.mlflow.org/docs/latest/tracking.html#scenario-6-mlflow-tracking-server-used-exclusively-as-proxied-access-host-for-artifact-storage-access))

## Blog ✍️

A detailed idea behind this repository and the steps to execute it in a cloud environment can be obtained from this blog post that I wrote on my experience building a solution like this.

Read it here 🧑‍💻: [Not just another MLflow on Kubernetes article](https://dev.to/wjayesh/not-just-another-mlflow-on-kubernetes-article-2522)