GitOps for machine learning

MLOPS-GIT.md

GitOps for machine learning aka MLOps

ENSAE: Introduction aux enjeux du MLOps

For open-source projects, GitHub could provide the workflow and runners for total automation.

wandb looks used widely for reporting.

https://github.blog/enterprise-software/ci-cd/streamlining-your-mlops-pipeline-with-github-actions-and-arm64-runners/

repos:

• Full GitHub CI/CD : https://github.com/BenedictusAryo/Simple-MLOps-Configuration/
• basic tf text classification : https://github.com/DamienBouvet/lab-continuous-ml-2
• Example basic sklearn + python prediction GitOps repo : https://github.com/kingabzpro/CICD-for-Machine-Learning
• transformers + translation training + eval : https://github.com/liangwen12year/nmstate-yamlsmith
• Trash classification with Jupyter : https://github.com/pradanaadn/trash-classification
• ONNX export examples https://github.com/bookbot-hive/sherpa-onnx/blob/a9dc16df153adffcbaa7268aee500b13b518653a/.github/workflows/export-wenet-to-onnx.yaml
• https://onnx.ai/sklearn-onnx/
• ModerBert : https://github.com/fkuhne/ModernBERT-cats-and-dogs

---

MLOps and GitHub Actions series : https://lo-victoria.com/series/mlops

First try with sklearn :

• repo : https://github.com/revolunet/mlops-classifier
• demo : https://revolunet.github.io/mlops-classifier/