From e4afb1082645ee26b83fe2da0bf137ee8f527d5a Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Sun, 28 Apr 2024 20:28:16 +0100 Subject: [PATCH 01/26] Update Kubeflow Installation with Standalone Components Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 97 +++++++++++-------- 1 file changed, 54 insertions(+), 43 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index ff35f26b73..ae696e7752 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,30 +5,65 @@ weight = 20 +++ -## What is Kubeflow? +Kubeflow is the ecosystem of various applications created to address each stage in the AI/ML lifecycle, +from exploration to training and serving. You can deploy Kubeflow components as standalone applications +or deploy Kubeflow as an end-to-end AI/ML platform. Anywhere you are running Kubernetes, +you should be able to run Kubeflow. -Kubeflow is an end-to-end Machine Learning (ML) platform for Kubernetes, it provides components for each stage in the ML lifecycle, from exploration through to training and deployment. -Operators can choose what is best for their users, there is no requirement to deploy every component. +There are a few ways to install Kubeflow: -Learn more about Kubeflow in the [Introduction](/docs/started/introduction/) and -[Architecture](/docs/started/architecture/) pages. +- [**Install Kubeflow Components Standalone**](#install-kubeflow-components-standalone) +- [**Install Kubeflow Platform from Raw Manifests**](#install-kubeflow-platform-from-raw-manifests) +- [**Install Kubeflow Platform from Packaged Distributions**](#install-kubeflow-platform-from-packages-distributions) -## How to install Kubeflow? +## Install Kubeflow Components Standalone -Anywhere you are running Kubernetes, you should be able to run Kubeflow. -There are two primary ways to install Kubeflow: +Kubeflow components can be deployed as standalone applications. You can integrate those components +to your existing AI/ML platform. For example, for Model Training you can install Training Operator +or for Model Serving you can install KServe. -1. [**Packaged Distributions**](#packaged-distributions-of-kubeflow) -1. [**Raw Manifests**](#raw-kubeflow-manifests) (advanced users) +Follow the appropriate guides to install required Kubeflow components in standalone mode: - - +- Install [Kubeflow Pipelines](/docs/components/pipelines/v2/installation/quickstart/) +- Install [Kubeflow Training Operator](/docs/components/training/installation/#installing-training-operator) +- Install [Kubeflow MPI Operator](/docs/components/training/user-guides/mpi/#installation) +- Install [Kubeflow Katib] (TODO: Add link after #3723) +- Install [KServe](https://kserve.github.io/website/0.10/admin/serverless/serverless/) +- Install Kubeflow Model Registry (TODO: Add link after #3698) -## Packaged Distributions of Kubeflow +## Install Kubeflow Platform from Raw Manifests + +The raw Kubeflow Manifests are aggregated by the +[Manifests Working Group](https://github.com/kubeflow/community/tree/master/wg-manifests) +and are intended to be used as the **base of packaged distributions**. + +Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow +applications which makes **Kubeflow Platform**. + +Users may choose to install the manifests for a specific Kubeflow version by following the +instructions in the `README` of the [`kubeflow/manifests`](https://github.com/kubeflow/manifests) repository. + +- [**Kubeflow 1.8:**](/docs/releases/kubeflow-1.8/) + - [`v1.8-branch`](https://github.com/kubeflow/manifests/tree/v1.8-branch#installation) (development branch) + - [`v1.8.0`](https://github.com/kubeflow/manifests/tree/v1.8.0#installation) +- [**Kubeflow 1.7:**](/docs/releases/kubeflow-1.7/) + - [`v1.7-branch`](https://github.com/kubeflow/manifests/tree/v1.7-branch#installation) (development branch) + - [`v1.7.0`](https://github.com/kubeflow/manifests/tree/v1.7.0#installation) + +{{% alert title="Warning" color="warning" %}} +Kubeflow is a complex system with many components and dependencies. +Using the raw manifests requires a deep understanding of Kubernetes, Istio, and Kubeflow itself. + +When using the raw manifests, the Kubeflow community is not able to provide support for environment-specific issues or custom configurations. +If you need support, please consider using a [packaged distribution](#packaged-distributions-of-kubeflow). +Nevertheless, we welcome contributions and bug reports very much. +{{% /alert %}} + +## Install Kubeflow Platform from Packages Distributions Packaged distributions are maintained by various organizations and typically aim to provide -a simplified installation and management experience for Kubeflow. Some distributions can be -deployed on [all certified Kubernetes distributions](https://kubernetes.io/partners/#conformance), +a simplified installation and management experience for **Kubeflow Platform**. Some distributions +can be deployed on [all certified Kubernetes distributions](https://kubernetes.io/partners/#conformance), while others target a specific platform (e.g. EKS or GKE). {{% alert title="Note" color="warning" %}} @@ -200,33 +235,9 @@ The following table lists distributions which are maintained by their r -## Raw Kubeflow Manifests - -The raw Kubeflow Manifests are aggregated by the [Manifests Working Group](https://github.com/kubeflow/community/tree/master/wg-manifests) -and are intended to be used as the **base of packaged distributions**. - -Advanced users may choose to install the manifests for a specific Kubeflow version by following the -instructions in the `README` of the [`kubeflow/manifests`](https://github.com/kubeflow/manifests) repository. - -- [**Kubeflow 1.8:**](/docs/releases/kubeflow-1.8/) - - [`v1.8-branch`](https://github.com/kubeflow/manifests/tree/v1.8-branch#installation) (development branch) - - [`v1.8.0`](https://github.com/kubeflow/manifests/tree/v1.8.0#installation) -- [**Kubeflow 1.7:**](/docs/releases/kubeflow-1.7/) - - [`v1.7-branch`](https://github.com/kubeflow/manifests/tree/v1.7-branch#installation) (development branch) - - [`v1.7.0`](https://github.com/kubeflow/manifests/tree/v1.7.0#installation) - -{{% alert title="Warning" color="warning" %}} -Kubeflow is a complex system with many components and dependencies. -Using the raw manifests requires a deep understanding of Kubernetes, Istio, and Kubeflow itself. - -When using the raw manifests, the Kubeflow community is not able to provide support for environment-specific issues or custom configurations. -If you need support, please consider using a [packaged distribution](#packaged-distributions-of-kubeflow). -Nevertheless, we welcome contributions and bug reports very much. -{{% /alert %}} - - - ## Next steps -- Review the Kubeflow component documentation -- Explore the Kubeflow Pipelines SDK +- Check [the Kubeflow introduction page](/docs/started/introduction/). +- Explore the [Kubeflow architecture](/docs/started/architecture) and how Kubeflow components fit + into the AI/ML lifecycle. +- Review [the Kubeflow components documentation](/docs/components/). From 72ef59397f91fd15bfbeb01ef0dfd4e02e6b7c22 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Mon, 29 Apr 2024 21:10:58 +0100 Subject: [PATCH 02/26] Add helpful message for manifests installation Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index ae696e7752..d676e82683 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,7 +5,7 @@ weight = 20 +++ -Kubeflow is the ecosystem of various applications created to address each stage in the AI/ML lifecycle, +Kubeflow is an ecosystem of various applications created to address each stage in the AI/ML lifecycle, from exploration to training and serving. You can deploy Kubeflow components as standalone applications or deploy Kubeflow as an end-to-end AI/ML platform. Anywhere you are running Kubernetes, you should be able to run Kubeflow. @@ -38,7 +38,8 @@ The raw Kubeflow Manifests are aggregated by the and are intended to be used as the **base of packaged distributions**. Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow -applications which makes **Kubeflow Platform**. +applications which makes **Kubeflow Platform**. This installation is helpful when you want to try +out the end-to-end Kubeflow Platform capabilities. Users may choose to install the manifests for a specific Kubeflow version by following the instructions in the `README` of the [`kubeflow/manifests`](https://github.com/kubeflow/manifests) repository. From 4110705d2690a816301b129b038e56355502a784 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 8 May 2024 12:57:17 +0100 Subject: [PATCH 03/26] Add explanation for Kubeflow Platform and Kubeflow Standalone Components Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 101 +++++++++++------- 1 file changed, 65 insertions(+), 36 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index d676e82683..4ab09b5e12 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,18 +5,47 @@ weight = 20 +++ -Kubeflow is an ecosystem of various applications created to address each stage in the AI/ML lifecycle, -from exploration to training and serving. You can deploy Kubeflow components as standalone applications -or deploy Kubeflow as an end-to-end AI/ML platform. Anywhere you are running Kubernetes, -you should be able to run Kubeflow. +## What is Kubeflow ? -There are a few ways to install Kubeflow: +Kubeflow is a community and ecosystem of open-source projects to address each stage in the +machine learning (ML) lifecycle. It makes ML on Kubernetes simple, portable, and scalable. +Its goal is to facilitate the orchestration of Kubernetes workloads for ML and to empower users to +deploy best-in-class open-source systems for ML to diverse cloud infrastructures. +Whether you’re a researcher, data scientist, ML engineer, or a team of developers, Kubeflow offers +modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to +deploying them to production for AI applications. + +## What are Kubeflow Standalone Components? + +Kubeflow is composed of multiple, independent open-source projects which address different aspects +of a ML lifecycle. These standalone components are designed to be usable both within the Kubeflow +platform and independently. These components can be installed on their own on a Kubernetes cluster, +providing flexibility to users who may not require the full capabilities of Kubeflow Platform but +wish to leverage specific functionalities in the ML lifecycle. + +## What is Kubeflow Platform ? + +The Kubeflow Platform refers to the full suite of Kubeflow components bundled together with +additional integration and management tools. Installing Kubeflow as a platform means deploying a +comprehensive ML toolkit that integrates these components into a cohesive system, optimized for +managing the end-to-end ML lifecycle. These includes not only the standalone components but also: + +- Central Dashboard for easy navigation and management. +- Multi-user capabilities and access management. +- Additional tooling and services for data management, visualization, and more. + +This integrated environment ensures that all the different pieces work together seamlessly, +providing a more robust and streamlined user experience. + +Kubeflow Platform can be installed via Raw Manifests or Package Distributions. + +## Installing Kubeflow - [**Install Kubeflow Components Standalone**](#install-kubeflow-components-standalone) -- [**Install Kubeflow Platform from Raw Manifests**](#install-kubeflow-platform-from-raw-manifests) - [**Install Kubeflow Platform from Packaged Distributions**](#install-kubeflow-platform-from-packages-distributions) +- [**Install Kubeflow Platform from Raw Manifests**](#install-kubeflow-platform-from-raw-manifests) -## Install Kubeflow Components Standalone +### Install Kubeflow Components Standalone Kubeflow components can be deployed as standalone applications. You can integrate those components to your existing AI/ML platform. For example, for Model Training you can install Training Operator @@ -31,35 +60,6 @@ Follow the appropriate guides to install required Kubeflow components in standal - Install [KServe](https://kserve.github.io/website/0.10/admin/serverless/serverless/) - Install Kubeflow Model Registry (TODO: Add link after #3698) -## Install Kubeflow Platform from Raw Manifests - -The raw Kubeflow Manifests are aggregated by the -[Manifests Working Group](https://github.com/kubeflow/community/tree/master/wg-manifests) -and are intended to be used as the **base of packaged distributions**. - -Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow -applications which makes **Kubeflow Platform**. This installation is helpful when you want to try -out the end-to-end Kubeflow Platform capabilities. - -Users may choose to install the manifests for a specific Kubeflow version by following the -instructions in the `README` of the [`kubeflow/manifests`](https://github.com/kubeflow/manifests) repository. - -- [**Kubeflow 1.8:**](/docs/releases/kubeflow-1.8/) - - [`v1.8-branch`](https://github.com/kubeflow/manifests/tree/v1.8-branch#installation) (development branch) - - [`v1.8.0`](https://github.com/kubeflow/manifests/tree/v1.8.0#installation) -- [**Kubeflow 1.7:**](/docs/releases/kubeflow-1.7/) - - [`v1.7-branch`](https://github.com/kubeflow/manifests/tree/v1.7-branch#installation) (development branch) - - [`v1.7.0`](https://github.com/kubeflow/manifests/tree/v1.7.0#installation) - -{{% alert title="Warning" color="warning" %}} -Kubeflow is a complex system with many components and dependencies. -Using the raw manifests requires a deep understanding of Kubernetes, Istio, and Kubeflow itself. - -When using the raw manifests, the Kubeflow community is not able to provide support for environment-specific issues or custom configurations. -If you need support, please consider using a [packaged distribution](#packaged-distributions-of-kubeflow). -Nevertheless, we welcome contributions and bug reports very much. -{{% /alert %}} - ## Install Kubeflow Platform from Packages Distributions Packaged distributions are maintained by various organizations and typically aim to provide @@ -236,6 +236,35 @@ The following table lists distributions which are maintained by their r +### Install Kubeflow Platform from Raw Manifests + +The raw Kubeflow Manifests are aggregated by the +[Manifests Working Group](https://github.com/kubeflow/community/tree/master/wg-manifests) +and are intended to be used as the **base of packaged distributions**. + +Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow +applications which makes **Kubeflow Platform**. This installation is helpful when you want to try +out the end-to-end Kubeflow Platform capabilities. + +Users may choose to install the manifests for a specific Kubeflow version by following the +instructions in the `README` of the [`kubeflow/manifests`](https://github.com/kubeflow/manifests) repository. + +- [**Kubeflow 1.8:**](/docs/releases/kubeflow-1.8/) + - [`v1.8-branch`](https://github.com/kubeflow/manifests/tree/v1.8-branch#installation) (development branch) + - [`v1.8.0`](https://github.com/kubeflow/manifests/tree/v1.8.0#installation) +- [**Kubeflow 1.7:**](/docs/releases/kubeflow-1.7/) + - [`v1.7-branch`](https://github.com/kubeflow/manifests/tree/v1.7-branch#installation) (development branch) + - [`v1.7.0`](https://github.com/kubeflow/manifests/tree/v1.7.0#installation) + +{{% alert title="Warning" color="warning" %}} +Kubeflow is a complex system with many components and dependencies. +Using the raw manifests requires a deep understanding of Kubernetes, Istio, and Kubeflow itself. + +When using the raw manifests, the Kubeflow community is not able to provide support for environment-specific issues or custom configurations. +If you need support, please consider using a [packaged distribution](#packaged-distributions-of-kubeflow). +Nevertheless, we welcome contributions and bug reports very much. +{{% /alert %}} + ## Next steps - Check [the Kubeflow introduction page](/docs/started/introduction/). From 7420e03384d8a881e5a24bb34c76e410a746cde3 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Fri, 10 May 2024 23:39:26 +0100 Subject: [PATCH 04/26] Move Kubeflow explanation to introduction guide Add table for Kubeflow components Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 164 +++++++++++++----- content/en/docs/started/introduction.md | 69 ++++---- 2 files changed, 158 insertions(+), 75 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 4ab09b5e12..4520c82f8c 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,62 +5,134 @@ weight = 20 +++ -## What is Kubeflow ? - -Kubeflow is a community and ecosystem of open-source projects to address each stage in the -machine learning (ML) lifecycle. It makes ML on Kubernetes simple, portable, and scalable. -Its goal is to facilitate the orchestration of Kubernetes workloads for ML and to empower users to -deploy best-in-class open-source systems for ML to diverse cloud infrastructures. -Whether you’re a researcher, data scientist, ML engineer, or a team of developers, Kubeflow offers -modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to -deploying them to production for AI applications. - -## What are Kubeflow Standalone Components? - -Kubeflow is composed of multiple, independent open-source projects which address different aspects -of a ML lifecycle. These standalone components are designed to be usable both within the Kubeflow -platform and independently. These components can be installed on their own on a Kubernetes cluster, -providing flexibility to users who may not require the full capabilities of Kubeflow Platform but -wish to leverage specific functionalities in the ML lifecycle. - -## What is Kubeflow Platform ? - -The Kubeflow Platform refers to the full suite of Kubeflow components bundled together with -additional integration and management tools. Installing Kubeflow as a platform means deploying a -comprehensive ML toolkit that integrates these components into a cohesive system, optimized for -managing the end-to-end ML lifecycle. These includes not only the standalone components but also: - -- Central Dashboard for easy navigation and management. -- Multi-user capabilities and access management. -- Additional tooling and services for data management, visualization, and more. - -This integrated environment ensures that all the different pieces work together seamlessly, -providing a more robust and streamlined user experience. - -Kubeflow Platform can be installed via Raw Manifests or Package Distributions. +This pages shows how to install Kubeflow standalone components or Kubeflow Platform using package +distributions or raw manifests. Check [the introduction guide](/docs/started/introduction) to +understand what are Kubeflow standalone components and what is Kubeflow Platform. ## Installing Kubeflow +You can install Kubeflow components using one of these methods: + - [**Install Kubeflow Components Standalone**](#install-kubeflow-components-standalone) -- [**Install Kubeflow Platform from Packaged Distributions**](#install-kubeflow-platform-from-packages-distributions) -- [**Install Kubeflow Platform from Raw Manifests**](#install-kubeflow-platform-from-raw-manifests) +- [**Install Kubeflow Platform from Packaged Distributions**](#install-kubeflow-platform-from-packaged-distributions) +- [**Install Kubeflow Platform from Raw Manifests**](#install-kubeflow-platform-from-raw-manifests) (advanced users) ### Install Kubeflow Components Standalone Kubeflow components can be deployed as standalone applications. You can integrate those components -to your existing AI/ML platform. For example, for Model Training you can install Training Operator -or for Model Serving you can install KServe. +to your existing AI/ML platform. This is the easiest method to get started with Kubeflow ecosystem +since those components don't require additional management used in Kubeflow Platform. -Follow the appropriate guides to install required Kubeflow components in standalone mode: +This table points to the installation guide for every Kubeflow component, the GitHub repository, and corresponding [stage of ML lifecycle](/docs/started/architecture/#kubeflow-components-in-the-ml-lifecycle) for every Kubeflow component. + +
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
ComponentSource CodeStage of ML Lifecycle
+ + Kubeflow Training Operator + + + + kubeflow/training-operator + + + Model Training and Fine-Tuning +
+ + Kubeflow Katib + + + + kubeflow/katib + + + Model Optimization and AutoML +
+ + Kubeflow MPI Operator + + + + kubeflow/mpi-operator + + + All-Reduce Model Training +
+ + Kubeflow Model Registry + + + + kubeflow/model-registry + + + Model Registry +
+ + KServe + + + + kserve/kserve + + + Model Serving +
+ + Kubeflow Pipelines + + + + kubeflow/pipelines + + + ML Workflows and Schedulers +
+
-- Install [Kubeflow Pipelines](/docs/components/pipelines/v2/installation/quickstart/) -- Install [Kubeflow Training Operator](/docs/components/training/installation/#installing-training-operator) -- Install [Kubeflow MPI Operator](/docs/components/training/user-guides/mpi/#installation) -- Install [Kubeflow Katib] (TODO: Add link after #3723) -- Install [KServe](https://kserve.github.io/website/0.10/admin/serverless/serverless/) -- Install Kubeflow Model Registry (TODO: Add link after #3698) +**Note**. Currently, Kubeflow Notebooks can't be deployed as a standalone application, but Notebooks +WG is working on that as part of [this issue](https://github.com/kubeflow/kubeflow/issues/7549). -## Install Kubeflow Platform from Packages Distributions +## Install Kubeflow Platform from Packaged Distributions Packaged distributions are maintained by various organizations and typically aim to provide a simplified installation and management experience for **Kubeflow Platform**. Some distributions @@ -269,5 +341,5 @@ Nevertheless, we welcome contributions and bug reports very much. - Check [the Kubeflow introduction page](/docs/started/introduction/). - Explore the [Kubeflow architecture](/docs/started/architecture) and how Kubeflow components fit - into the AI/ML lifecycle. + into the ML lifecycle. - Review [the Kubeflow components documentation](/docs/components/). diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index f8d0576fae..bc2389bb95 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -4,11 +4,41 @@ description = "An introduction to Kubeflow" weight = 1 +++ -The Kubeflow project is dedicated to making deployments of machine learning (ML) -workflows on Kubernetes simple, portable and scalable. Our goal is not to -recreate other services, but to provide a straightforward way to deploy -best-of-breed open-source systems for ML to diverse infrastructures. Anywhere -you are running Kubernetes, you should be able to run Kubeflow. +## What is Kubeflow ? + +Kubeflow is a community and ecosystem of open-source projects to address each stage in the +machine learning (ML) lifecycle. It makes ML on Kubernetes simple, portable, and scalable. +Its goal is to facilitate the orchestration of Kubernetes workloads for ML and to empower users to +deploy best-in-class open-source systems for ML to diverse cloud infrastructures. +Whether you’re a researcher, data scientist, ML engineer, or a team of developers, Kubeflow offers +modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to +deploying them to production for AI applications. + +## What are Kubeflow Standalone Components? + +Kubeflow is composed of multiple, independent open-source projects which address different aspects +of a ML lifecycle. These standalone components are designed to be usable both within the Kubeflow +Platform and independently. These components can be installed on their own on a Kubernetes cluster, +providing flexibility to users who may not require the full capabilities of Kubeflow Platform but +wish to leverage specific functionalities in the ML lifecycle. + +## What is Kubeflow Platform ? + +The Kubeflow Platform refers to the full suite of Kubeflow components bundled together with +additional integration and management tools. Installing Kubeflow as a platform means deploying a +comprehensive ML toolkit that integrates these components into a cohesive system, optimized for +managing the end-to-end ML lifecycle. These includes not only the standalone components but also: + +- Central Dashboard for easy navigation and management. +- Multi-user capabilities and access management. +- Additional tooling and services for data management, visualization, and more. + +This integrated environment ensures that all the different pieces work together seamlessly, +providing a more robust and streamlined user experience. + +Kubeflow Platform can be installed via Raw Manifests or Package Distributions. + +## Getting started with Kubeflow The following diagram shows the main Kubeflow components to cover each step of ML lifecycle on top of Kubernetes. @@ -17,8 +47,6 @@ on top of Kubernetes. alt="Kubeflow overview" class="mt-3 mb-3"> -## Getting started with Kubeflow - Read the [architecture overview](/docs/started/architecture/) for an introduction to the architecture of Kubeflow and to see how you can use Kubeflow to manage your ML workflow. @@ -30,28 +58,6 @@ Watch the following video which provides an introduction to Kubeflow. {{< youtube id="cTZArDgbIWw" title="Introduction to Kubeflow">}} -## What is Kubeflow? - -Kubeflow is _the machine learning toolkit for Kubernetes_. - -To use Kubeflow, the basic workflow is: - -- Download and run the Kubeflow deployment binary. -- Customize the resulting configuration files. -- Run the specified script to deploy your containers to your specific - environment. - -You can adapt the configuration to choose the platforms and services that you -want to use for each stage of the ML workflow: - -1. data preparation -2. model training, -3. prediction serving -4. service management - -You can choose to deploy your Kubernetes workloads locally, on-premises, or to -a cloud environment. - ## The Kubeflow mission Our goal is to make scaling machine learning (ML) models and deploying them to @@ -94,3 +100,8 @@ The following components also have roadmaps: There are many ways to contribute to Kubeflow, and we welcome contributions! Read the [contributor's guide](/docs/about/contributing/) to get started on the code, and learn about the community on the [community page](/docs/about/community/). + +## Next Steps + +- Follow [the installation guide](/docs/started/installing-kubeflow) to deploy Kubeflow standalone + components or Kubeflow Platform. From 26b262af336b486a2a69ee1b11fab43f3fa54262 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Fri, 10 May 2024 23:47:36 +0100 Subject: [PATCH 05/26] Add Spark Operator Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 25 +++++++++++++++---- 1 file changed, 20 insertions(+), 5 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 4520c82f8c..3f1e84f6bb 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -37,17 +37,17 @@ This table points to the installation guide for every Kubeflow component, the Gi - - Kubeflow Training Operator + + Kubeflow Spark Operator - - kubeflow/training-operator + + kubeflow/spark-operator - Model Training and Fine-Tuning + Data Preparation @@ -65,6 +65,21 @@ This table points to the installation guide for every Kubeflow component, the Gi Model Optimization and AutoML + + + + Kubeflow Training Operator + + + + + kubeflow/training-operator + + + + Model Training and Fine-Tuning + + From 6cf8f38edb7f2b3b13eeadd1ab3c51f33b6d2063 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Fri, 10 May 2024 23:53:35 +0100 Subject: [PATCH 06/26] Add links to the introduction Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 12 +++++++----- content/en/docs/started/introduction.md | 4 +++- 2 files changed, 10 insertions(+), 6 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 3f1e84f6bb..be3683de3b 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,7 +5,7 @@ weight = 20 +++ -This pages shows how to install Kubeflow standalone components or Kubeflow Platform using package +This pages describes how to install Kubeflow standalone components or Kubeflow Platform using package distributions or raw manifests. Check [the introduction guide](/docs/started/introduction) to understand what are Kubeflow standalone components and what is Kubeflow Platform. @@ -21,11 +21,13 @@ You can install Kubeflow components using one of these methods: Kubeflow components can be deployed as standalone applications. You can integrate those components to your existing AI/ML platform. This is the easiest method to get started with Kubeflow ecosystem -since those components don't require additional management used in Kubeflow Platform. +since those components usually don't require additional management tools used in Kubeflow Platform. -This table points to the installation guide for every Kubeflow component, the GitHub repository, and corresponding [stage of ML lifecycle](/docs/started/architecture/#kubeflow-components-in-the-ml-lifecycle) for every Kubeflow component. +This table points to the installation guide for every Kubeflow component, the GitHub repository, +and corresponding [stage of ML lifecycle](/docs/started/architecture/#kubeflow-components-in-the-ml-lifecycle) +for every Kubeflow component. -
+
@@ -147,7 +149,7 @@ This table points to the installation guide for every Kubeflow component, the Gi **Note**. Currently, Kubeflow Notebooks can't be deployed as a standalone application, but Notebooks WG is working on that as part of [this issue](https://github.com/kubeflow/kubeflow/issues/7549). -## Install Kubeflow Platform from Packaged Distributions +### Install Kubeflow Platform from Packaged Distributions Packaged distributions are maintained by various organizations and typically aim to provide a simplified installation and management experience for **Kubeflow Platform**. Some distributions diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index bc2389bb95..5c57979226 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -36,7 +36,9 @@ managing the end-to-end ML lifecycle. These includes not only the standalone com This integrated environment ensures that all the different pieces work together seamlessly, providing a more robust and streamlined user experience. -Kubeflow Platform can be installed via Raw Manifests or Package Distributions. +Kubeflow Platform can be installed via +[Packaged Distributions](/docs/started/installing-kubeflow/#install-kubeflow-platform-from-packaged-distributions) or +[Raw Manifests](/docs/started/installing-kubeflow/#install-kubeflow-platform-from-raw-manifests). ## Getting started with Kubeflow From 15ed4c7b00ab1dabe663a574cb9dbc5e6e7c218c Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Mon, 13 May 2024 14:38:30 +0100 Subject: [PATCH 07/26] Remove Manifests WG link Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index be3683de3b..a392fcd847 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -328,8 +328,7 @@ The following table lists distributions which are maintained by their r ### Install Kubeflow Platform from Raw Manifests The raw Kubeflow Manifests are aggregated by the -[Manifests Working Group](https://github.com/kubeflow/community/tree/master/wg-manifests) -and are intended to be used as the **base of packaged distributions**. +Manifests Working Group and are intended to be used as the **base of packaged distributions**. Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow applications which makes **Kubeflow Platform**. This installation is helpful when you want to try From a022b346de9bcedbfc87e9d75567636e0a2dc985 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Tue, 14 May 2024 12:06:00 +0100 Subject: [PATCH 08/26] Modify table column Fix description text Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 74 ++++++++++--------- 1 file changed, 39 insertions(+), 35 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index a392fcd847..d946253aee 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,13 +5,15 @@ weight = 20 +++ -This pages describes how to install Kubeflow standalone components or Kubeflow Platform using package -distributions or raw manifests. Check [the introduction guide](/docs/started/introduction) to +This page describes how to install Kubeflow standalone components or Kubeflow Platform using package +distributions or raw manifests. + +Read [the introduction guide](/docs/started/introduction) to understand what are Kubeflow standalone components and what is Kubeflow Platform. ## Installing Kubeflow -You can install Kubeflow components using one of these methods: +You can install Kubeflow using one of these methods: - [**Install Kubeflow Components Standalone**](#install-kubeflow-components-standalone) - [**Install Kubeflow Platform from Packaged Distributions**](#install-kubeflow-platform-from-packaged-distributions) @@ -19,21 +21,24 @@ You can install Kubeflow components using one of these methods: ### Install Kubeflow Components Standalone -Kubeflow components can be deployed as standalone applications. You can integrate those components -to your existing AI/ML platform. This is the easiest method to get started with Kubeflow ecosystem -since those components usually don't require additional management tools used in Kubeflow Platform. +Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptual-overview) may be +deployed as standalone services, without the need to install the full platform. You might integrate +these services as part of your existing AI/ML platform or use them independently. + +This is the easiest method to get started with Kubeflow ecosystem since those components usually +don't require additional management tools used in Kubeflow Platform. -This table points to the installation guide for every Kubeflow component, the GitHub repository, -and corresponding [stage of ML lifecycle](/docs/started/architecture/#kubeflow-components-in-the-ml-lifecycle) -for every Kubeflow component. +The following table lists Kubeflow components that may be deployed in a standalone mode. It also +lists their associated GitHub repository and +corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-components-in-the-ml-lifecycle).
+ - @@ -43,14 +48,14 @@ for every Kubeflow component. Kubeflow Spark Operator + - + - + - + - + - + - + -
ComponentML Lifecycle Stage Source CodeStage of ML Lifecycle
+ Data Preparation + kubeflow/spark-operator - Data Preparation -
@@ -58,14 +63,14 @@ for every Kubeflow component. Kubeflow Katib + Model Optimization and AutoML + kubeflow/katib - Model Optimization and AutoML -
@@ -73,14 +78,14 @@ for every Kubeflow component. Kubeflow Training Operator + Model Training and Fine-Tuning + kubeflow/training-operator - Model Training and Fine-Tuning -
@@ -88,14 +93,14 @@ for every Kubeflow component. Kubeflow MPI Operator + All-Reduce Model Training + kubeflow/mpi-operator - All-Reduce Model Training -
@@ -103,14 +108,14 @@ for every Kubeflow component. Kubeflow Model Registry + Model Registry + kubeflow/model-registry - Model Registry -
@@ -118,14 +123,14 @@ for every Kubeflow component. KServe + Model Serving + kserve/kserve - Model Serving -
@@ -133,14 +138,14 @@ for every Kubeflow component. Kubeflow Pipelines + ML Workflows and Schedules + kubeflow/pipelines - ML Workflows and Schedulers -
@@ -355,7 +360,6 @@ Nevertheless, we welcome contributions and bug reports very much. ## Next steps -- Check [the Kubeflow introduction page](/docs/started/introduction/). -- Explore the [Kubeflow architecture](/docs/started/architecture) and how Kubeflow components fit - into the ML lifecycle. -- Review [the Kubeflow components documentation](/docs/components/). +- Review our [Introduction to Kubeflow](/docs/started/introduction/). +- Explore the [Architecture of Kubeflow](/docs/started/architecture). +- Learn more about the [Components of Kubeflow](/docs/components/). From f9d5a522a1b9a711c3aa0dd75a8a60a9b64a7710 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Tue, 14 May 2024 23:28:07 +0100 Subject: [PATCH 09/26] Order components alphabeticaly Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 60 +++++++++---------- 1 file changed, 30 insertions(+), 30 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index d946253aee..578286aaaa 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -44,46 +44,46 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen - - Kubeflow Spark Operator + + Kubeflow Katib - Data Preparation + Model Optimization and AutoML - - kubeflow/spark-operator + + kubeflow/katib - - Kubeflow Katib + + KServe - Model Optimization and AutoML + Model Serving - - kubeflow/katib + + kserve/kserve - - Kubeflow Training Operator + + Kubeflow Model Registry - Model Training and Fine-Tuning + Model Registry - - kubeflow/training-operator + + kubeflow/model-registry @@ -104,46 +104,46 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen - - Kubeflow Model Registry + + Kubeflow Pipelines - Model Registry + ML Workflows and Schedules - - kubeflow/model-registry + + kubeflow/pipelines - - KServe + + Kubeflow Spark Operator - Model Serving + Data Preparation - - kserve/kserve + + kubeflow/spark-operator - - Kubeflow Pipelines + + Kubeflow Training Operator - ML Workflows and Schedules + Model Training and Fine-Tuning - - kubeflow/pipelines + + kubeflow/training-operator From 8fbad119bc84382c7ac22a7385e3846ef313b03a Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Tue, 14 May 2024 23:38:39 +0100 Subject: [PATCH 10/26] Update introduction Signed-off-by: Andrey Velichkevich --- content/en/docs/started/introduction.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 5c57979226..367e185189 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -8,8 +8,8 @@ weight = 1 Kubeflow is a community and ecosystem of open-source projects to address each stage in the machine learning (ML) lifecycle. It makes ML on Kubernetes simple, portable, and scalable. -Its goal is to facilitate the orchestration of Kubernetes workloads for ML and to empower users to -deploy best-in-class open-source systems for ML to diverse cloud infrastructures. +The Kubeflow goal is to facilitate the orchestration of Kubernetes ML workloads and to empower users +to deploy best-in-class open-source systems to any Cloud infrastructure. Whether you’re a researcher, data scientist, ML engineer, or a team of developers, Kubeflow offers modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to deploying them to production for AI applications. @@ -18,9 +18,9 @@ deploying them to production for AI applications. Kubeflow is composed of multiple, independent open-source projects which address different aspects of a ML lifecycle. These standalone components are designed to be usable both within the Kubeflow -Platform and independently. These components can be installed on their own on a Kubernetes cluster, +Platform and independently. These components can be installed independently on a Kubernetes cluster, providing flexibility to users who may not require the full capabilities of Kubeflow Platform but -wish to leverage specific functionalities in the ML lifecycle. +wish to leverage specific ML functionalities. ## What is Kubeflow Platform ? From 29391f869a28f34e729f8913589a806b4d59c3f4 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 15 May 2024 12:24:07 +0100 Subject: [PATCH 11/26] Update Kubeflow intro Signed-off-by: Andrey Velichkevich --- content/en/docs/started/introduction.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 367e185189..034e48c372 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -8,8 +8,8 @@ weight = 1 Kubeflow is a community and ecosystem of open-source projects to address each stage in the machine learning (ML) lifecycle. It makes ML on Kubernetes simple, portable, and scalable. -The Kubeflow goal is to facilitate the orchestration of Kubernetes ML workloads and to empower users -to deploy best-in-class open-source systems to any Cloud infrastructure. +The goal of Kubeflow is to facilitate the orchestration of Kubernetes ML workloads and to empower +users to deploy best-in-class open-source systems to any Cloud infrastructure. Whether you’re a researcher, data scientist, ML engineer, or a team of developers, Kubeflow offers modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to deploying them to production for AI applications. From db5c5423f6374bd3f4e07870fbf2ac0433e7a7b7 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Mon, 20 May 2024 13:44:51 +0100 Subject: [PATCH 12/26] Fix KFP install link Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 578286aaaa..0ba5a6dfd5 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -104,7 +104,7 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen - + Kubeflow Pipelines From 5c80be77d9f34af4796cbd93b3c0ac9519e15f6e Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Mon, 20 May 2024 13:57:27 +0100 Subject: [PATCH 13/26] Review comments Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 18 +++++++++--------- content/en/docs/started/introduction.md | 3 ++- 2 files changed, 11 insertions(+), 10 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 0ba5a6dfd5..98119a93c2 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,11 +5,11 @@ weight = 20 +++ -This page describes how to install Kubeflow standalone components or Kubeflow Platform using package +This guide describes how to install Kubeflow standalone components or Kubeflow Platform using package distributions or raw manifests. -Read [the introduction guide](/docs/started/introduction) to -understand what are Kubeflow standalone components and what is Kubeflow Platform. +Read [the introduction guide](/docs/started/introduction) to understand what are Kubeflow +standalone components and what is Kubeflow Platform. ## Installing Kubeflow @@ -25,8 +25,8 @@ Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptu deployed as standalone services, without the need to install the full platform. You might integrate these services as part of your existing AI/ML platform or use them independently. -This is the easiest method to get started with Kubeflow ecosystem since those components usually -don't require additional management tools used in Kubeflow Platform. +This is a easier method to get started with Kubeflow ecosystem since those components usually +don't require additional management tools used in a Kubeflow Platform. The following table lists Kubeflow components that may be deployed in a standalone mode. It also lists their associated GitHub repository and @@ -157,7 +157,7 @@ WG is working on that as part of [this issue](https://github.com/kubeflow/kubefl ### Install Kubeflow Platform from Packaged Distributions Packaged distributions are maintained by various organizations and typically aim to provide -a simplified installation and management experience for **Kubeflow Platform**. Some distributions +a simplified installation and management experience for your **Kubeflow Platform**. Some distributions can be deployed on [all certified Kubernetes distributions](https://kubernetes.io/partners/#conformance), while others target a specific platform (e.g. EKS or GKE). @@ -360,6 +360,6 @@ Nevertheless, we welcome contributions and bug reports very much. ## Next steps -- Review our [Introduction to Kubeflow](/docs/started/introduction/). -- Explore the [Architecture of Kubeflow](/docs/started/architecture). -- Learn more about the [Components of Kubeflow](/docs/components/). +- Review our [introduction to Kubeflow](/docs/started/introduction/). +- Explore the [architecture of Kubeflow](/docs/started/architecture). +- Learn more about the [components of Kubeflow](/docs/components/). diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 034e48c372..6b5df67ecf 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -27,7 +27,8 @@ wish to leverage specific ML functionalities. The Kubeflow Platform refers to the full suite of Kubeflow components bundled together with additional integration and management tools. Installing Kubeflow as a platform means deploying a comprehensive ML toolkit that integrates these components into a cohesive system, optimized for -managing the end-to-end ML lifecycle. These includes not only the standalone components but also: +managing the end-to-end ML lifecycle. This includes the standalone components coupled with these +integrations and management tools: - Central Dashboard for easy navigation and management. - Multi-user capabilities and access management. From 78f683ec7d6902e4979ceac3e973a76d1dbaca6b Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Tue, 21 May 2024 20:18:19 +0100 Subject: [PATCH 14/26] Add Kubeflow Platform Header Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 25 +++++++++++-------- 1 file changed, 15 insertions(+), 10 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 98119a93c2..ca3f8c8193 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -11,21 +11,21 @@ distributions or raw manifests. Read [the introduction guide](/docs/started/introduction) to understand what are Kubeflow standalone components and what is Kubeflow Platform. -## Installing Kubeflow - You can install Kubeflow using one of these methods: - [**Install Kubeflow Components Standalone**](#install-kubeflow-components-standalone) -- [**Install Kubeflow Platform from Packaged Distributions**](#install-kubeflow-platform-from-packaged-distributions) -- [**Install Kubeflow Platform from Raw Manifests**](#install-kubeflow-platform-from-raw-manifests) (advanced users) -### Install Kubeflow Components Standalone +- Install Kubeflow Platform + - [**From Packaged Distributions**](#from-packaged-distributions) + - [**From Raw Manifests**](#from-raw-manifests) (advanced users) + +## Install Kubeflow Components Standalone Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptual-overview) may be deployed as standalone services, without the need to install the full platform. You might integrate these services as part of your existing AI/ML platform or use them independently. -This is a easier method to get started with Kubeflow ecosystem since those components usually +This is a quick and easier method to get started with Kubeflow ecosystem since those components usually don't require additional management tools used in a Kubeflow Platform. The following table lists Kubeflow components that may be deployed in a standalone mode. It also @@ -154,7 +154,12 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen **Note**. Currently, Kubeflow Notebooks can't be deployed as a standalone application, but Notebooks WG is working on that as part of [this issue](https://github.com/kubeflow/kubeflow/issues/7549). -### Install Kubeflow Platform from Packaged Distributions +## Install Kubeflow Platform + +You can use one of the following methods to install Kubeflow Platform to get full suite of Kubeflow +components bundled together with additional integration and management tools. + +### From Packaged Distributions Packaged distributions are maintained by various organizations and typically aim to provide a simplified installation and management experience for your **Kubeflow Platform**. Some distributions @@ -330,10 +335,10 @@ The following table lists distributions which are maintained by their r
-### Install Kubeflow Platform from Raw Manifests +### From Raw Manifests -The raw Kubeflow Manifests are aggregated by the -Manifests Working Group and are intended to be used as the **base of packaged distributions**. +The raw Kubeflow Manifests are aggregated by the Manifests Working Group and are intended to be +used as the **base of packaged distributions**. Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow applications which makes **Kubeflow Platform**. This installation is helpful when you want to try From 35ea61535afb3f3dca7d061441aa26b1827f2554 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 22 May 2024 00:01:46 +0100 Subject: [PATCH 15/26] Modify headers and text Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 15 ++++---- content/en/docs/started/introduction.md | 38 +++++++++---------- 2 files changed, 25 insertions(+), 28 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index ca3f8c8193..449a1504aa 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -8,16 +8,15 @@ weight = 20 This guide describes how to install Kubeflow standalone components or Kubeflow Platform using package distributions or raw manifests. -Read [the introduction guide](/docs/started/introduction) to understand what are Kubeflow -standalone components and what is Kubeflow Platform. +Read [the introduction guide](/docs/started/introduction) to learn more about Kubeflow, Kubeflow +standalone components and Kubeflow Platform. + +## Installing Kubeflow You can install Kubeflow using one of these methods: - [**Install Kubeflow Components Standalone**](#install-kubeflow-components-standalone) - -- Install Kubeflow Platform - - [**From Packaged Distributions**](#from-packaged-distributions) - - [**From Raw Manifests**](#from-raw-manifests) (advanced users) +- [**Install Kubeflow Platform**](#install-kubeflow-platform) ## Install Kubeflow Components Standalone @@ -159,7 +158,7 @@ WG is working on that as part of [this issue](https://github.com/kubeflow/kubefl You can use one of the following methods to install Kubeflow Platform to get full suite of Kubeflow components bundled together with additional integration and management tools. -### From Packaged Distributions +### Packaged Distributions Packaged distributions are maintained by various organizations and typically aim to provide a simplified installation and management experience for your **Kubeflow Platform**. Some distributions @@ -335,7 +334,7 @@ The following table lists distributions which are maintained by their r
-### From Raw Manifests +### Raw Manifests The raw Kubeflow Manifests are aggregated by the Manifests Working Group and are intended to be used as the **base of packaged distributions**. diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 6b5df67ecf..9b818541b7 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -4,42 +4,40 @@ description = "An introduction to Kubeflow" weight = 1 +++ -## What is Kubeflow ? +## What is Kubeflow Kubeflow is a community and ecosystem of open-source projects to address each stage in the machine learning (ML) lifecycle. It makes ML on Kubernetes simple, portable, and scalable. The goal of Kubeflow is to facilitate the orchestration of Kubernetes ML workloads and to empower -users to deploy best-in-class open-source systems to any Cloud infrastructure. +users to deploy best-in-class open-source tools on any Cloud infrastructure. Whether you’re a researcher, data scientist, ML engineer, or a team of developers, Kubeflow offers modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to deploying them to production for AI applications. -## What are Kubeflow Standalone Components? +## What are Kubeflow Standalone Components -Kubeflow is composed of multiple, independent open-source projects which address different aspects -of a ML lifecycle. These standalone components are designed to be usable both within the Kubeflow -Platform and independently. These components can be installed independently on a Kubernetes cluster, -providing flexibility to users who may not require the full capabilities of Kubeflow Platform but +Kubeflow is composed of multiple open-source projects which address different aspects +of the ML lifecycle. These standalone components are designed to be usable both within the Kubeflow +Platform and independently. These components can be installed standalone on a Kubernetes cluster, +providing flexibility to users who may not require the full Kubeflow Platform capabilities but wish to leverage specific ML functionalities. -## What is Kubeflow Platform ? +## What is Kubeflow Platform The Kubeflow Platform refers to the full suite of Kubeflow components bundled together with -additional integration and management tools. Installing Kubeflow as a platform means deploying a -comprehensive ML toolkit that integrates these components into a cohesive system, optimized for -managing the end-to-end ML lifecycle. This includes the standalone components coupled with these -integrations and management tools: +additional integration and management tools. Using Kubeflow as a platform means deploying a +comprehensive ML toolkit fot the entire ML lifecycle. This includes the standalone components +coupled with these integrations and management tools: -- Central Dashboard for easy navigation and management. -- Multi-user capabilities and access management. -- Additional tooling and services for data management, visualization, and more. +- [**Central Dashboard**](/docs/components/central-dash/overview/) for easy navigation and management. -This integrated environment ensures that all the different pieces work together seamlessly, -providing a more robust and streamlined user experience. +- Multi-user capabilities and access management with [Kubeflow Profiles](/docs/components/central-dash/profiles/). -Kubeflow Platform can be installed via -[Packaged Distributions](/docs/started/installing-kubeflow/#install-kubeflow-platform-from-packaged-distributions) or -[Raw Manifests](/docs/started/installing-kubeflow/#install-kubeflow-platform-from-raw-manifests). +- Additional tooling for data management (PVC Viewer), visualization (TensorBoards), and more. + +The Kubeflow Platform can be installed via +[Packaged Distributions](/docs/started/installing-kubeflow/#packaged-distributions) or +[Raw Manifests](/docs/started/installing-kubeflow/#raw-manifests). ## Getting started with Kubeflow From a438b5a4ba7355b63384632d41bb564abb399515 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 22 May 2024 14:34:57 +0100 Subject: [PATCH 16/26] Add Kubeflow Notebooks to Kubeflow Platform tools Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 3 --- content/en/docs/started/introduction.md | 8 ++++---- 2 files changed, 4 insertions(+), 7 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 449a1504aa..35d49f0d6f 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -150,9 +150,6 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen -**Note**. Currently, Kubeflow Notebooks can't be deployed as a standalone application, but Notebooks -WG is working on that as part of [this issue](https://github.com/kubeflow/kubeflow/issues/7549). - ## Install Kubeflow Platform You can use one of the following methods to install Kubeflow Platform to get full suite of Kubeflow diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 9b818541b7..9ccd2fbcf4 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -29,10 +29,10 @@ additional integration and management tools. Using Kubeflow as a platform means comprehensive ML toolkit fot the entire ML lifecycle. This includes the standalone components coupled with these integrations and management tools: -- [**Central Dashboard**](/docs/components/central-dash/overview/) for easy navigation and management. - -- Multi-user capabilities and access management with [Kubeflow Profiles](/docs/components/central-dash/profiles/). - +- [Kubeflow Notebooks](/docs/components/notebooks/overview) for interactive data exploration and + model development. +- [Central Dashboard](/docs/components/central-dash/overview/) for easy navigation and management + with [Kubeflow Profiles](/docs/components/central-dash/profiles/) for access control. - Additional tooling for data management (PVC Viewer), visualization (TensorBoards), and more. The Kubeflow Platform can be installed via From 2495604acf036fa8dd6969c3a01f23f324dd1926 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 22 May 2024 14:41:16 +0100 Subject: [PATCH 17/26] Modify the install headers Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 35d49f0d6f..d8016c2308 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -11,14 +11,14 @@ distributions or raw manifests. Read [the introduction guide](/docs/started/introduction) to learn more about Kubeflow, Kubeflow standalone components and Kubeflow Platform. -## Installing Kubeflow +## Installing Methods You can install Kubeflow using one of these methods: -- [**Install Kubeflow Components Standalone**](#install-kubeflow-components-standalone) -- [**Install Kubeflow Platform**](#install-kubeflow-platform) +- [**Kubeflow Components Standalone**](#kubeflow-components-standalone) +- [**Kubeflow Platform**](#kubeflow-platform) -## Install Kubeflow Components Standalone +### Kubeflow Components Standalone Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptual-overview) may be deployed as standalone services, without the need to install the full platform. You might integrate @@ -150,12 +150,12 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen -## Install Kubeflow Platform +### Kubeflow Platform You can use one of the following methods to install Kubeflow Platform to get full suite of Kubeflow components bundled together with additional integration and management tools. -### Packaged Distributions +#### Packaged Distributions Packaged distributions are maintained by various organizations and typically aim to provide a simplified installation and management experience for your **Kubeflow Platform**. Some distributions @@ -331,7 +331,7 @@ The following table lists distributions which are maintained by their r -### Raw Manifests +#### Raw Manifests The raw Kubeflow Manifests are aggregated by the Manifests Working Group and are intended to be used as the **base of packaged distributions**. From cf7c50be55e8af005d5ee91fa818f244c697560e Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 22 May 2024 14:56:06 +0100 Subject: [PATCH 18/26] Update what are Kubeflow Standalone Components Signed-off-by: Andrey Velichkevich --- content/en/docs/started/introduction.md | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 9ccd2fbcf4..1b3bd69b2f 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -10,17 +10,20 @@ Kubeflow is a community and ecosystem of open-source projects to address each st machine learning (ML) lifecycle. It makes ML on Kubernetes simple, portable, and scalable. The goal of Kubeflow is to facilitate the orchestration of Kubernetes ML workloads and to empower users to deploy best-in-class open-source tools on any Cloud infrastructure. + Whether you’re a researcher, data scientist, ML engineer, or a team of developers, Kubeflow offers modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to deploying them to production for AI applications. ## What are Kubeflow Standalone Components -Kubeflow is composed of multiple open-source projects which address different aspects -of the ML lifecycle. These standalone components are designed to be usable both within the Kubeflow -Platform and independently. These components can be installed standalone on a Kubernetes cluster, -providing flexibility to users who may not require the full Kubeflow Platform capabilities but -wish to leverage specific ML functionalities. +Kubeflow ecosystem is composed of multiple open-source projects that address different aspects of +the ML lifecycle. Many of these projects are designed to be usable both within the +Kubeflow Platform and independently. + +These Kubeflow components can be installed standalone on a Kubernetes cluster. It provides +flexibility to users who may not require the full Kubeflow Platform capabilities but wish to +leverage specific ML functionalities such as model training or model serving. ## What is Kubeflow Platform From 224a638b6386f4392be92947b7070def57d23633 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 22 May 2024 22:57:18 +0100 Subject: [PATCH 19/26] Change to Standalone Kubeflow Components Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 6 +++--- content/en/docs/started/introduction.md | 11 ++++++----- 2 files changed, 9 insertions(+), 8 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index d8016c2308..be0d821254 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -11,14 +11,14 @@ distributions or raw manifests. Read [the introduction guide](/docs/started/introduction) to learn more about Kubeflow, Kubeflow standalone components and Kubeflow Platform. -## Installing Methods +## Installation Methods You can install Kubeflow using one of these methods: -- [**Kubeflow Components Standalone**](#kubeflow-components-standalone) +- [**Standalone Kubeflow Components**](#standalone-kubeflow-components) - [**Kubeflow Platform**](#kubeflow-platform) -### Kubeflow Components Standalone +### Standalone Kubeflow Components Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptual-overview) may be deployed as standalone services, without the need to install the full platform. You might integrate diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 1b3bd69b2f..10dbc4df91 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -15,10 +15,10 @@ Whether you’re a researcher, data scientist, ML engineer, or a team of develop modular and scalable tools that cater to all aspects of the ML lifecycle: from building ML models to deploying them to production for AI applications. -## What are Kubeflow Standalone Components +## What are Standalone Kubeflow Components -Kubeflow ecosystem is composed of multiple open-source projects that address different aspects of -the ML lifecycle. Many of these projects are designed to be usable both within the +The Kubeflow ecosystem is composed of multiple open-source projects that address different aspects +of the ML lifecycle. Many of these projects are designed to be usable both within the Kubeflow Platform and independently. These Kubeflow components can be installed standalone on a Kubernetes cluster. It provides @@ -29,8 +29,9 @@ leverage specific ML functionalities such as model training or model serving. The Kubeflow Platform refers to the full suite of Kubeflow components bundled together with additional integration and management tools. Using Kubeflow as a platform means deploying a -comprehensive ML toolkit fot the entire ML lifecycle. This includes the standalone components -coupled with these integrations and management tools: +comprehensive ML toolkit for the entire ML lifecycle. + +In addition to the standalone Kubeflow components, the Kubeflow Platform includes - [Kubeflow Notebooks](/docs/components/notebooks/overview) for interactive data exploration and model development. From fb14fb16a254456b34341cb928410459fa5b2ddc Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Wed, 22 May 2024 23:14:21 +0100 Subject: [PATCH 20/26] Add link to Kubeflow Platform Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 14 +++++++------- content/en/docs/started/introduction.md | 8 ++++---- 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index be0d821254..3e141ef9de 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -5,11 +5,11 @@ weight = 20 +++ -This guide describes how to install Kubeflow standalone components or Kubeflow Platform using package +This guide describes how to install standalone Kubeflow components or Kubeflow Platform using package distributions or raw manifests. -Read [the introduction guide](/docs/started/introduction) to learn more about Kubeflow, Kubeflow -standalone components and Kubeflow Platform. +Read [the introduction guide](/docs/started/introduction) to learn more about Kubeflow, standalone +Kubeflow components and Kubeflow Platform. ## Installation Methods @@ -21,8 +21,8 @@ You can install Kubeflow using one of these methods: ### Standalone Kubeflow Components Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptual-overview) may be -deployed as standalone services, without the need to install the full platform. You might integrate -these services as part of your existing AI/ML platform or use them independently. +deployed as standalone services, without the need to install the full Kubeflow platform. You might +integrate these services as part of your existing AI/ML platform or use them independently. This is a quick and easier method to get started with Kubeflow ecosystem since those components usually don't require additional management tools used in a Kubeflow Platform. @@ -152,8 +152,8 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen ### Kubeflow Platform -You can use one of the following methods to install Kubeflow Platform to get full suite of Kubeflow -components bundled together with additional integration and management tools. +You can use one of the following methods to install [Kubeflow Platform](/docs/started/introduction/#what-is-kubeflow-platform) +to get full suite of standalone Kubeflow components bundled together with additional tools. #### Packaged Distributions diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index 10dbc4df91..cf5e0ea4c7 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -96,9 +96,9 @@ To see what's coming up in future versions of Kubeflow, refer to the [Kubeflow r The following components also have roadmaps: - [Kubeflow Pipelines](https://github.com/kubeflow/pipelines/blob/master/ROADMAP.md) -- [KF Serving](https://github.com/kubeflow/kfserving/blob/master/ROADMAP.md) +- [KServe](https://github.com/kserve/kserve/blob/master/ROADMAP.md) - [Katib](https://github.com/kubeflow/katib/blob/master/ROADMAP.md) -- [Training Operator](https://github.com/kubeflow/common/blob/master/ROADMAP.md) +- [Training Operator](https://github.com/kubeflow/training-operator/blob/master/docs/roadmap.md) ## Getting involved @@ -108,5 +108,5 @@ Read the [contributor's guide](/docs/about/contributing/) to get started on the ## Next Steps -- Follow [the installation guide](/docs/started/installing-kubeflow) to deploy Kubeflow standalone - components or Kubeflow Platform. +- Follow [the installation guide](/docs/started/installing-kubeflow) to deploy standalone + Kubeflow components or Kubeflow Platform. From e18ac0854360e7829af3c2392c554b0a45fb278b Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Thu, 23 May 2024 12:05:03 +0100 Subject: [PATCH 21/26] Rename Raw Manifests to Kubeflow Manifests Signed-off-by: Andrey Velichkevich --- .../en/docs/started/installing-kubeflow.md | 30 +++++++++---------- content/en/docs/started/introduction.md | 2 +- 2 files changed, 16 insertions(+), 16 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 3e141ef9de..fd3c80aaa7 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -6,7 +6,7 @@ weight = 20 +++ This guide describes how to install standalone Kubeflow components or Kubeflow Platform using package -distributions or raw manifests. +distributions or Kubeflow manifests. Read [the introduction guide](/docs/started/introduction) to learn more about Kubeflow, standalone Kubeflow components and Kubeflow Platform. @@ -43,31 +43,31 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen - - Kubeflow Katib + + KServe - Model Optimization and AutoML + Model Serving - - kubeflow/katib + + kserve/kserve - - KServe + + Kubeflow Katib - Model Serving + Model Optimization and AutoML - - kserve/kserve + + kubeflow/katib @@ -331,9 +331,9 @@ The following table lists distributions which are maintained by their r -#### Raw Manifests +#### Kubeflow Manifests -The raw Kubeflow Manifests are aggregated by the Manifests Working Group and are intended to be +The Kubeflow Manifests are aggregated by the Manifests Working Group and are intended to be used as the **base of packaged distributions**. Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow @@ -352,9 +352,9 @@ instructions in the `README` of the [`kubeflow/manifests`](https://github.com/ku {{% alert title="Warning" color="warning" %}} Kubeflow is a complex system with many components and dependencies. -Using the raw manifests requires a deep understanding of Kubernetes, Istio, and Kubeflow itself. +Using the Kubeflow manifests requires a deep understanding of Kubernetes, Istio, and Kubeflow itself. -When using the raw manifests, the Kubeflow community is not able to provide support for environment-specific issues or custom configurations. +When using the Kubeflow manifests, the community is not able to provide support for environment-specific issues or custom configurations. If you need support, please consider using a [packaged distribution](#packaged-distributions-of-kubeflow). Nevertheless, we welcome contributions and bug reports very much. {{% /alert %}} diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index cf5e0ea4c7..a098004d3f 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -41,7 +41,7 @@ In addition to the standalone Kubeflow components, the Kubeflow Platform include The Kubeflow Platform can be installed via [Packaged Distributions](/docs/started/installing-kubeflow/#packaged-distributions) or -[Raw Manifests](/docs/started/installing-kubeflow/#raw-manifests). +[Kubeflow Manifests](/docs/started/installing-kubeflow/#kubeflow-manifests). ## Getting started with Kubeflow From 7df289cc9536f7b66d5316543a50c4cc19d74ce2 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Fri, 24 May 2024 01:01:12 +0100 Subject: [PATCH 22/26] Change H3 Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index fd3c80aaa7..c3fa4fcd59 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -18,7 +18,7 @@ You can install Kubeflow using one of these methods: - [**Standalone Kubeflow Components**](#standalone-kubeflow-components) - [**Kubeflow Platform**](#kubeflow-platform) -### Standalone Kubeflow Components +## Standalone Kubeflow Components Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptual-overview) may be deployed as standalone services, without the need to install the full Kubeflow platform. You might @@ -150,12 +150,12 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen -### Kubeflow Platform +## Kubeflow Platform You can use one of the following methods to install [Kubeflow Platform](/docs/started/introduction/#what-is-kubeflow-platform) to get full suite of standalone Kubeflow components bundled together with additional tools. -#### Packaged Distributions +### Packaged Distributions Packaged distributions are maintained by various organizations and typically aim to provide a simplified installation and management experience for your **Kubeflow Platform**. Some distributions @@ -331,7 +331,7 @@ The following table lists distributions which are maintained by their r -#### Kubeflow Manifests +### Kubeflow Manifests The Kubeflow Manifests are aggregated by the Manifests Working Group and are intended to be used as the **base of packaged distributions**. From 90df53dea03b99799df2b0c12701b1f0fd4f8315 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Sat, 25 May 2024 22:32:10 +0100 Subject: [PATCH 23/26] Rename to quick and easy Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index c3fa4fcd59..59053facaa 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -24,7 +24,7 @@ Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptu deployed as standalone services, without the need to install the full Kubeflow platform. You might integrate these services as part of your existing AI/ML platform or use them independently. -This is a quick and easier method to get started with Kubeflow ecosystem since those components usually +This is a quick and easy method to get started with Kubeflow ecosystem since those components usually don't require additional management tools used in a Kubeflow Platform. The following table lists Kubeflow components that may be deployed in a standalone mode. It also From 63c3a61cc26cae9a7f9e64915ab0d613d663fb2f Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Sat, 25 May 2024 22:34:57 +0100 Subject: [PATCH 24/26] Remove new line in Introduction Signed-off-by: Andrey Velichkevich --- content/en/docs/started/introduction.md | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/content/en/docs/started/introduction.md b/content/en/docs/started/introduction.md index a098004d3f..b404c161cc 100644 --- a/content/en/docs/started/introduction.md +++ b/content/en/docs/started/introduction.md @@ -19,11 +19,9 @@ deploying them to production for AI applications. The Kubeflow ecosystem is composed of multiple open-source projects that address different aspects of the ML lifecycle. Many of these projects are designed to be usable both within the -Kubeflow Platform and independently. - -These Kubeflow components can be installed standalone on a Kubernetes cluster. It provides -flexibility to users who may not require the full Kubeflow Platform capabilities but wish to -leverage specific ML functionalities such as model training or model serving. +Kubeflow Platform and independently. These Kubeflow components can be installed standalone on a +Kubernetes cluster. It provides flexibility to users who may not require the full Kubeflow Platform +capabilities but wish to leverage specific ML functionalities such as model training or model serving. ## What is Kubeflow Platform From b0de894a2bdb19b40bf52d4ecf6811367fc286c4 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Mon, 27 May 2024 17:57:23 +0100 Subject: [PATCH 25/26] Modify the install standalone section Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index 59053facaa..fc83c7099c 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -24,8 +24,8 @@ Some components in the [Kubeflow ecosystem](/docs/started/architecture/#conceptu deployed as standalone services, without the need to install the full Kubeflow platform. You might integrate these services as part of your existing AI/ML platform or use them independently. -This is a quick and easy method to get started with Kubeflow ecosystem since those components usually -don't require additional management tools used in a Kubeflow Platform. +These components are a quick and easy method to get started with the Kubeflow ecosystem. They +provide flexibility to users who may not require the capabilities of a full Kubeflow Platform. The following table lists Kubeflow components that may be deployed in a standalone mode. It also lists their associated GitHub repository and From 7e7b91c5d08ea9fb0f6dc9f4dbd5f6085bd44e56 Mon Sep 17 00:00:00 2001 From: Andrey Velichkevich Date: Tue, 28 May 2024 14:02:53 +0100 Subject: [PATCH 26/26] Fix install guide Signed-off-by: Andrey Velichkevich --- content/en/docs/started/installing-kubeflow.md | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/content/en/docs/started/installing-kubeflow.md b/content/en/docs/started/installing-kubeflow.md index fc83c7099c..502b2bd3b1 100644 --- a/content/en/docs/started/installing-kubeflow.md +++ b/content/en/docs/started/installing-kubeflow.md @@ -152,8 +152,8 @@ corresponding [ML lifecycle stage](/docs/started/architecture/#kubeflow-componen ## Kubeflow Platform -You can use one of the following methods to install [Kubeflow Platform](/docs/started/introduction/#what-is-kubeflow-platform) -to get full suite of standalone Kubeflow components bundled together with additional tools. +You can use one of the following methods to install the [Kubeflow Platform](/docs/started/introduction/#what-is-kubeflow-platform) +and get the full suite of Kubeflow components bundled together with additional tools. ### Packaged Distributions @@ -337,8 +337,8 @@ The Kubeflow Manifests are aggregated by the Manifests Working Group and are int used as the **base of packaged distributions**. Kubeflow Manifests contain all Kubeflow Components, Kubeflow Central Dashboard, and other Kubeflow -applications which makes **Kubeflow Platform**. This installation is helpful when you want to try -out the end-to-end Kubeflow Platform capabilities. +applications that comprise the **Kubeflow Platform**. This installation is helpful when you want to +try out the end-to-end Kubeflow Platform capabilities. Users may choose to install the manifests for a specific Kubeflow version by following the instructions in the `README` of the [`kubeflow/manifests`](https://github.com/kubeflow/manifests) repository. @@ -355,7 +355,7 @@ Kubeflow is a complex system with many components and dependencies. Using the Kubeflow manifests requires a deep understanding of Kubernetes, Istio, and Kubeflow itself. When using the Kubeflow manifests, the community is not able to provide support for environment-specific issues or custom configurations. -If you need support, please consider using a [packaged distribution](#packaged-distributions-of-kubeflow). +If you need support, please consider using a [packaged distribution](#packaged-distributions). Nevertheless, we welcome contributions and bug reports very much. {{% /alert %}}