Defining ML Workflows including:
From the Notebook menu, create a new notebook server.
The central user interface (UI) in Kubeflow
Using Jupyter notebooks in Kubeflow
A powerful platform for building end to end ML workflows. Pipelines allow you to build a set of steps to handle everything from collecting data to serving your model.
Kubeflow model deployment and serving toolkit
Katib is a Kubernetes-native project for automated machine learning (AutoML). Katib supports hyperparameter tuning, early stopping and neural architecture search (NAS).
Training of ML models in Kubeflow through operators such as Tensorflow and Pytorch.
Multi-user isolation and identity access management (IAM)
Defacto standard for data scientists for performing rapid data analysis.
A pipeline component is one step in the workflow that performs one specific task.
Takes inputs and produces outputs.
Reference | URL |
---|---|
Kubeflow 101 Videos | https://www.youtube.com/playlist?list=PLIivdWyY5sqLS4lN75RPDEyBgTro_YX7x |