Kubeflow is an open-source Kubernetes native workflow engine created for deploying complex machine learning workflows. It was developed from the Argo Workflow engine. This tool is one among many in the open-source ecosystem simply the use of machine learning models.
Kubeflow began life within Google, designed to help run TensorFlow jobs but grew to support multi-cloud and multi-architecture frameworks. Today, companies like IBM, Red Hat, Cisco, and other large tech companies contribute to this project. And it supports a wide array of projects like TensorFlow, PyTorch, Chainer, MXNet, Ambassador, XGBoost, Istio, Nuclio, and more.
- Tool: Kubeflow
- Author: Google
- Initial Release: March 2018
- Type: Workflow Engine
- License: Apache-2.0
- Supports: Various frameworks and libraries
- Github: kubeflow
- Runs On: Anywhere
- Hardware: Supports CPUs, GPUs, and TPUs
- Twitter: kubeflow
- Supports various architectural use cases
- Scales workloads without much effort
- Works for training and inference
- Built on top of Kubernetes