All Tools
K
Dev ToolsFreeOpen Source
KUBEFLOW
Machine learning toolkit for Kubernetes
Apache-2.0
ABOUT
Deploying and managing machine learning workflows in production requires orchestrating notebooks, distributed training jobs, hyperparameter tuning, and model serving across a cluster. Doing this manually with raw Kubernetes is complex and error-prone. Kubeflow solves this by providing a dedicated, integrated platform that packages best practices for ML on Kubernetes into reusable components. Data scientists can work in Jupyter notebooks, launch distributed training with TFJob or PyTorchJob, run automated hyperparameter tuning with Katib, and deploy models with KServe — all within a single Kubernetes-native ecosystem that scales from experimentation to production.
INSTALL
kubectl apply -k "github.com/kubeflow/pipelines/manifests/kustomize/cluster-scoped-resources?ref=2.0.3"
kubectl apply -k "github.com/kubeflow/pipelines/manifests/kustomize/env/platform-agnostic-pns?ref=2.0.3"
INTEGRATION GUIDE
1. Building end-to-end ML pipelines on Kubernetes from notebooks to model serving
2. Running distributed TensorFlow or PyTorch training jobs with automatic cluster scheduling
3. Automating hyperparameter optimization and model deployment in a shared Kubernetes environment
TAGS
kubernetesml-pipelinestrainingservingopen-source