Kubeflow: TensorFlow on Kubernetes

on YouTube ↗29m

" summary

Kubeflow v0.3 is deployed on Kubernetes using ksonnet to generate manifests and Ambassador plus Istio for routing, providing a JupyterHub notebook environment. Distributed TensorFlow runs via TFJob CustomResources that launch GPU-backed Docker containers with master/worker/ps topology, persisting models to PVC/blob storage and serving them with TF Serving; hyperparameter sweeps spawn multiple TFJobs (e.g., a 3×3 matrix creating nine jobs).

AI-written summary · about

" chapters (17)

1. 0:00Introduction
2. 1:55ML lifecycle challenges
3. 3:20What is Kubeflow
4. 5:46Kubeflow architecture
5. 7:09Deploy Kubeflow with KSonnet
6. 8:37Kubeflow components in action
7. 10:03Jupyter Notebooks
8. 11:49Training models with TFJobs
9. 13:33Model management with volumes
10. 14:35Distributed model training
11. 16:19Distributed jobs with TFConfig
12. 18:27Hyperparameter optimization
13. 20:31Deploy and test models
14. 21:55TensorFlow Serving
15. 24:01Model operations
16. 26:09Monitoring tools
17. 27:15Conclusion

" related talks

• 36mKubeflow: Tensorflow on KubernetesPar Laurent Grangeau, Par Sylvain Lequeux · DataXDay 2019
• 35mKubeflow: Data Scientist ready?Sylvain Lequeux · DevFest Paris 2019
• 46mKubeflow, TensorFlow on KubernetesIsrael Olalla · Commit Conf 2018
• 21mIntroducing Kubeflow (w. Special Guests Tensorflow and Apache Spark)Holden Karau, Trevor Grant · Open Source Summit + Embedded Linux Conference North America 2019
• 14mKubeflow: Where ML Meets KubernetesSuresh Peiris · KCD Sri Lanka 2023
• 22mThe Containerization of Machine Learning: TensorFlow: Kubernetes, and KubeflowSyed Ahmed · Montréal Cloud Native Day 2019
• 31mKubeflow - Data Science on SteroidsSascha Grunert · ContainerDay 2019
• 31mKubeflow Intro – Michał Jastrzębski & Ala Raddaoui, IntelMichał Jastrzębski, Ala Raddaoui · KubeCon + CloudNativeCon Europe 2018