kubeflow
Add kube-nodered
Which issue is resolved by this Pull Request: Resolves #
Description of your changes: Feature Area
It takes many laborious procedures to build a practical ML pipelines including pipeline docker file compilation and edit. Laborious manual operations on kubeflow package are required to observe trainning results.
What feature would you like to see?
Kube-node-red is aiming to integrate Kubeflow/Kubebeters with node-red, leveraging node-red's low-code modules, and using Kubeflow resources (e.g. Kubeflow pipeline, Kserve) to enhance its AI/ML ability.
What is the use case or pain point?
Although kubeflow pipeline is a convenient tool for major ML operations such training and testting, it takes many laborious procedures to build a practical ML pipelines, not to mention switching between various ML algorithms and data sets. A graphical UI can provide a rapid deployment medium for ML developer, professional/non professional alike, to exercise an ML operation without starting from scratch. The following site provide some completed integration suites and examples for illustration of this op. https://github.com/NightLightTw/kube-nodered
Is there a workaround currently?
No workaround as far as I know.
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).
View this failed invocation of the CLA check for more information.
For the most up to date status, view the checks section at the bottom of the pull request.
![google-cla[bot] avatar](https://avatars.githubusercontent.com/in/42202?v=4 "Published by a pub.dev verified publisher"){kind=small}
Ok, I have to do a thorough review. First, did you look at bentoml, ray etc. and the requirements mentioned there? E.g. runasnonroot and namespace isolation?
Dear Reviewer/Sir:
Thank you for commenting on our PR. This PR features low-code drag and drop deployment of a selection of frequently used AI, such as random forest, decision tree, and logistic regression on Node-red frontend with kubeflow pipeline running behind. Different from bentoml, our design aims at minimizing the labor in setting up dependencies, and coding efforts in model build. Although the users only need to operate required ML through a Node-red GUI, it actually exercises the complete ML training op.s with kubeflow pipeline running behind the scene. Thus all characters and requirements of bentoml, ray such as namespace isolation and runasnonroot are preserved. Moreover our implementation hides all the coding complexity for novice developers and professions eager to see fast outcome of their AI applications. Please feel free to ask questions of our PR, we will reply ASAP.
For example i would like to understand the differences to Elyra https://github.com/elyra-ai/elyra. Would you rather like to collaborate with elyra?
Dear Reviewer/Sir:
Elyra is a fantastic project. Even though we also aim to implement a graphical drag-and-drop programming feature, our project is somewhat different. Kube-nodered is committed to running on lightweight devices such as the Raspberry Pi. We hope to use Node-RED as the front end. In Node-RED, there are many IoT-related nodes that can be combined, and we aim to leverage the powerful container-based development capabilities of Kubeflow as the back end. Kube-nodered is designed to develop and implement a new AIOT way.
I took a first glimse and added some comments. So far it looks like you run a website locally that acts as a client to Kubeflow. But i really need a detailed architectural view similar to https://docs.google.com/document/d/1SP-COL6O-ETMQny7zXbrbEL4zZBG3REErWjfcvFZVEw/edit?usp=sharing&resourcekey=0-IaUGVemd2XX6MGTgifDbpg and understand how you want to really integrate. For example adding your service as a namespace isolated component and making it available via the sidebar similar to KFP Kserve etc. We will also need an owners file. Please also check the other basic requirements from https://github.com/kubeflow/manifests/blob/master/proposals/20220926-contrib-component-guidelines.md#component-requirements.
By the way i just write straightforward, so please do not feel offended if something sounds too harsh for you. And if i do not respond for a week fell free to ping me here. I am also available on the Kubeflow Slack.
I am always excited to see cool applications of KF, and this looks very interesting for IOT people.
I think the best place for this project would be its own GitHub org and repo owned by the creators, rather than Kubeflow itself (at least initially).
We can then help highlight this project by linking it on the Kubeflow website, possibly under: https://www.kubeflow.org/docs/external-add-ons/
@thesuperzapper Thank you for your helpful suggestion of linking our PR on the Kubeflow website. At current stage, the functionality of our PR is quite preliminary. It certainly needs more modification to be more general and flexible for the general public. Yeah, please help us link to the Kubeflow add-on websit. We are currently actively joininng the WG meeting of manifest and kubwflow, trying to seek a suitable place in the kubeflow community, maybe with a proper tuning to our current work.
@thesuperzapper I just posted an issue within the manifest group asking for help with the add-on merge of our PR. The link is as follows: https://github.com/NightLightTw/kube-nodered Please help to put it within the kubeflow add-on settings. Also please tell us how to proceed with the rest of the steps to merge.
I just wanted to link my response to the other issue with what I think your next steps should be:
- https://github.com/kubeflow/manifests/issues/2499#issuecomment-1699863245
Dear Sir,
I wish to thank you for help reviewing this project. As a matter of fact, we were continue working on boosting this project and advertizing its use on many media publicly. Recently, we introduced Node-red-kubeflow in a national event of DevOps with onsite audience of 300 more and online attendance of several thousands. I wish to reopen the issue so it can merge into your community.
Julius von Kohout @.***> 於 2024年2月23日 週五 上午1:04寫道：
i think this is stale. please reopen if necessary.
— Reply to this email directly, view it on GitHub https://github.com/kubeflow/manifests/pull/2470#issuecomment-1959882783, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHKUIPT67MCDYUNFEEEG7XLYU53CRAVCNFSM6AAAAAAZLZT2NWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNJZHA4DENZYGM . You are receiving this because you commented.Message ID: @.***>
-- 石志雄 Chihhsiong Shih
東海大學 資工系 Department of Computer Science, Tunghai University
Dear Sir:
Please let me know how to reopen it.
Julius von Kohout @.***> 於 2024年2月23日 週五 上午1:04寫道：
i think this is stale. please reopen if necessary.
— Reply to this email directly, view it on GitHub https://github.com/kubeflow/manifests/pull/2470#issuecomment-1959882783, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHKUIPT67MCDYUNFEEEG7XLYU53CRAVCNFSM6AAAAAAZLZT2NWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNJZHA4DENZYGM . You are receiving this because you commented.Message ID: @.***>
-- 石志雄 Chihhsiong Shih
東海大學 資工系 Department of Computer Science, Tunghai University
[APPROVALNOTIFIER] This PR is NOT APPROVED
This pull-request has been approved by: NightLightTw Once this PR has been reviewed and has the lgtm label, please assign juliusvonkohout for approval. For more information see the Kubernetes Code Review Process.
The full list of commands accepted by this bot can be found here.
Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment
![google-oss-prow[bot] avatar](https://avatars.githubusercontent.com/in/143327?v=4 "Published by a pub.dev verified publisher"){kind=small}
Thank you all.
google-oss-prow[bot] @.***> 於 2024年2月27日 週二 上午1:37寫道：
[APPROVALNOTIFIER] This PR is NOT APPROVED
This pull-request has been approved by: NightLightTw https://github.com/kubeflow/manifests/pull/2470# Once this PR has been reviewed and has the lgtm label, please assign juliusvonkohout https://github.com/juliusvonkohout for approval. For more information see the Kubernetes Code Review Process https://git.k8s.io/community/contributors/guide/owners.md#the-code-review-process .
The full list of commands accepted by this bot can be found here https://go.k8s.io/bot-commands?repo=kubeflow%2Fmanifests. Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing /approve in a comment Approvers can cancel approval by writing /approve cancel in a comment
— Reply to this email directly, view it on GitHub https://github.com/kubeflow/manifests/pull/2470#issuecomment-1964725526, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHKUIPSMRXYR3SCGLDS2CSLYVTB37AVCNFSM6AAAAAAZLZT2NWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNRUG4ZDKNJSGY . You are receiving this because you commented.Message ID: @.***>
-- 石志雄 Chihhsiong Shih
東海大學 資工系 Department of Computer Science, Tunghai University
May I ask if the 3/21 WGM is skiped?
https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail 乾淨無病毒。www.avast.com https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
cheesepuff287 shihc @.***> 於 2024年2月27日 週二 上午7:41寫道：
Thank you all.
google-oss-prow[bot] @.***> 於 2024年2月27日 週二 上午1:37寫道：
[APPROVALNOTIFIER] This PR is NOT APPROVED
This pull-request has been approved by: NightLightTw https://github.com/kubeflow/manifests/pull/2470# Once this PR has been reviewed and has the lgtm label, please assign juliusvonkohout https://github.com/juliusvonkohout for approval. For more information see the Kubernetes Code Review Process https://git.k8s.io/community/contributors/guide/owners.md#the-code-review-process .
The full list of commands accepted by this bot can be found here https://go.k8s.io/bot-commands?repo=kubeflow%2Fmanifests. Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing /approve in a comment Approvers can cancel approval by writing /approve cancel in a comment
— Reply to this email directly, view it on GitHub https://github.com/kubeflow/manifests/pull/2470#issuecomment-1964725526, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHKUIPSMRXYR3SCGLDS2CSLYVTB37AVCNFSM6AAAAAAZLZT2NWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNRUG4ZDKNJSGY . You are receiving this because you commented.Message ID: @.***>
-- 石志雄 Chihhsiong Shih
東海大學 資工系 Department of Computer Science, Tunghai University
-- 石志雄 Chihhsiong Shih
東海大學 資工系 Department of Computer Science, Tunghai University
OK! When will the next WGM be held?
Julius von Kohout @.***> 於 2024年3月28日 週四 下午7:14寫道：
@482170765 https://github.com/482170765 yes because of the Kubecon
— Reply to this email directly, view it on GitHub https://github.com/kubeflow/manifests/pull/2470#issuecomment-2024938491, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHKUIPWCGZGD7PUKPQPELFLY2PUR5AVCNFSM6AAAAAAZLZT2NWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMRUHEZTQNBZGE . You are receiving this because you were mentioned.Message ID: @.***>
-- 石志雄 Chihhsiong Shih
東海大學 資工系 Department of Computer Science, Tunghai University
Here is our calendar https://www.kubeflow.org/docs/about/community/