Kubeflow — what does it provide?

medium

Answer

$Kubernetes-native ML platform.$
$Components: (1) Kubeflow Pipelines (KFP): DAG orchestration on K8s.$
$(2) Katib: hyperparameter tuning + NAS.$
$(3) KServe (formerly KFServing): model serving with autoscaling.$
$(4) Notebooks: managed Jupyter on K8s.$
$(5) Distributed training operators (PyTorch, TF, MPI, XGBoost).$
$Modular: use pieces separately.$
$Standard for K8s-first ML organizations.$
$Alternatives: MLRun, ZenML, Metaflow.$

Check yourself — multiple choice

Random
K8s-native ML: KFP pipelines + Katib HPO/NAS + KServe serving + Notebooks + distributed training operators (PyTorch/TF/MPI/XGB); modular; K8s-first orgs; alt MLRun/ZenML/Metaflow
Just Jupyter
Not real

Kubeflow: K8s-native ML; KFP + Katib + KServe + notebooks + training ops.

#mlops#infrastructure

Practise MLOps & Data Quality

215 interview questions in this topic.

Read the lesson Flashcards Take the quiz

Related questions