Kubeflow
This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these template messages)
|
File:Kubeflow-logo.png | |
Original author(s) | |
---|---|
Developer(s) | Kubeflow Contributors[1] - AWS, Bloomberg, Google, IBM, NVIDIA, Nutanix, Red Hat, Arrikto, and others |
Initial release | April 5, 2018[2] |
Stable release | 1.6[3]
/ September 7, 2022 |
Repository | github |
Written in | Go, Python |
Platform | Kubernetes |
Type | Machine Learning Platform |
License | Apache License 2.0 |
Website | kubeflow |
Kubeflow is a free and open-source platform for machine learning on Kubernetes. The Kubeflow project has multiple distinct software components which each address specific stages of the machine learning lifecycle, including model development (Kubeflow Notebooks[4]), model training (Kubeflow Pipelines,[5] Kubeflow Training Operator[6]), model serving (KServe[a][7]), and automated machine learning (Katib[8]).
Each component of Kubeflow can be deployed separately, and it is not a requirement to deploy every component.[9]
History
The Kubeflow project was first announced at KubeCon + CloudNativeCon North America 2017 by Google engineers David Aronchick, Jeremy Lewi, and Vishnu Kannan[10] to address a perceived lack of flexible options for building production-ready machine learning systems.[11] The project has also stated it began as a way for Google to open-source how they ran TensorFlow internally.[12]
The first release of Kubeflow (Kubeflow 0.1) was announced at KubeCon + CloudNativeCon Europe 2018[13] with claims of having already become among the top 2% of GitHub projects ever.[14] Kubeflow 1.0 was released in March 2020 via a public blog post announcing that many Kubeflow components were graduating to a "stable status", indicating they were now ready for production usage.[15]
Components
Kubeflow Notebooks
The Kubeflow Notebooks component provides a way to run web-based development environments inside a Kubernetes cluster, with native support for Jupyter Notebook, Visual Studio Code, and RStudio.[16]
Kubeflow Pipelines
The Kubeflow Pipelines component provides a platform for building and deploying portable, scalable machine learning workflows based on Docker containers.[17] Google Cloud Platform has adopted the Kubeflow Pipelines DSL within its Vertex AI Pipelines product.[18]
Kubeflow Training Operator
The Kubeflow Training Operator component provides Kubernetes custom resources that make it easy to run distributed or non-distributed TensorFlow, PyTorch, Apache MXNet, XGBoost, and MPI training jobs on Kubernetes.[6]
KServe
The KServe component (previously named KFServing[19]) provides Kubernetes custom resources for serving machine learning models on arbitrary frameworks including TensorFlow, XGBoost, scikit-learn, PyTorch, and ONNX.[20] KServe was developed collaboratively by Google, IBM, Bloomberg, NVIDIA, and Seldon.[19] Publicly disclosed adopters of KServe include Bloomberg,[21] Gojek,[22] and others[23]
Katib
The Katib component is described as a Kubernetes-native project for automated machine learning with support for hyperparameter tuning, early stopping, and neural architecture search.[24]
Release timeline
Notes
References
- ^ "Kubeflow Website - Working Groups".
- ^ a b "Kubeflow 0.1 - Release Tag".
- ^ a b "Kubeflow 1.6 - Release Information".
- ^ "Kubeflow Website - Kubeflow Notebooks".
- ^ "Kubeflow Website - Kubeflow Pipelines".
- ^ a b "Kubeflow GitHub - Kubeflow Training Operator".
- ^ "Kubeflow Website - KServe".
- ^ "Kubeflow Website - Katib".
- ^ "Kubeflow Website - Installing Kubeflow".
- ^ ""Hot Dogs or Not" - At Scale with Kubernetes [I] - Vish Kannan & David Aronchick, Google".
- ^ "Introducing Kubeflow - A Composable, Portable, Scalable ML Stack Built for Kubernetes".
- ^ "Kubeflow Website - History".
- ^ "Google-led Kubeflow, machine learning for Kubernetes, begins to take shape".
- ^ "Announcing Kubeflow 0.1".
- ^ "Kubeflow 1.0: Cloud-Native ML for Everyone".
- ^ "Kubeflow Website - Kubeflow Notebooks Overview".
- ^ "Kubeflow Website - Kubeflow Pipelines Introduction".
- ^ "Vertex AI - Building a pipeline".
- ^ a b c "KServe: The next generation of KFServing".
- ^ "KServe GitHub".
- ^ "The journey to build Bloomberg's ML Inference Platform Using KServe (formerly KFServing)".
- ^ "Merlin: Making ML Model Deployments Magical".
- ^ "KServe Website - Adopters of KServe".
- ^ "Kubeflow GitHub - Katib".
- ^ "Kubeflow 0.2 - Release Tag".
- ^ "Kubeflow 0.3 - Release Tag".
- ^ "Kubeflow 0.4 - Release Tag".
- ^ "Kubeflow 0.5 - Release Tag".
- ^ "Kubeflow 0.6 - Release Information".
- ^ "Kubeflow 0.7 - Release Information".
- ^ "Kubeflow 1.0 - Release Information".
- ^ "Kubeflow 1.1 - Release Information".
- ^ "Kubeflow 1.2 - Release Information".
- ^ "Kubeflow 1.3 - Release Information".
- ^ "Kubeflow 1.4 - Release Information".
- ^ "Kubeflow 1.5 - Release Information".
External links
- Articles with a promotional tone from September 2022
- All articles with a promotional tone
- Justapedia articles with possible conflicts of interest from September 2022
- Articles with invalid date parameter in template
- Articles with multiple maintenance issues
- Articles with missing files
- Official website missing URL
- 2018 software
- Cloud infrastructure
- Data mining and machine learning software
- Software using the Apache license
- Free software programmed in Python
- Free software programmed in Go