WG ML Experience proposal

[ederign]

This PR proposes a new WG to the Kubeflow community.

The ML Experience Working Group is responsible for developing and integrating IDE-based tools and extensions to provide a streamlined user experience to data scientists and machine learning practitioners on Kubeflow.

This PR closes #808.

[google-oss-prow]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign terrytangyuan for approval. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ederign]

/cc @andreyvelich @StefanoFioravanzo ^

[StefanoFioravanzo]

Awesome, thank you Eder!

[juliusvonkohout]

I integrated for example labelstudio some time ago. Would this be interesting?

[franciscojavierarceo]

I integrated for example labelstudio some time ago. Would this be interesting?

that would be awesome.

[shivay-couchbase]

I would love to join this

[varodrig]

I'd love to join this as well.

[franciscojavierarceo]

Awesome @shivay-couchbase.

Also are you working with @ejscribner?

[franciscojavierarceo]

FYI @ejscribner

[jesuino]

Great proposal, I would like to join!

[ederign]

@juliusvonkohout I would love to hear more about labelstudio? Is there a recording or doc from our example?

[ederign]

@shivay-couchbase @varodrig @jesuino awesome! Feel free to review this proposal! We would love your feedback!

[juliusvonkohout]

@juliusvonkohout I would love to hear more about labelstudio? Is there a recording or doc from our example?

https://labelstud.io/ we can offer it similarly to jupyterlabs.

[andreyvelich]

I also found this solution interesting: https://lightning.ai/docs/overview/studios/background-execution

[andreyvelich]

Appreciate your review for the WG ML Experience goals and charter..

/assign @bigsur0 @Zsailer @shravan-achar @akshaychitneni @lresende

[google-oss-prow]

@andreyvelich: GitHub didn't allow me to assign the following users: bigsur0, Zsailer, shravan-achar, akshaychitneni.

Note that only kubeflow members with read permissions, repo collaborators and people who have commented on this issue/PR can be assigned. Additionally, issues/PRs can only have 10 assignees at the same time.
For more information please see the contributor guide

In response to this:

Appreciate your review for the WG ML Experience goals and charter..

/assign @bigsur0 @Zsailer @shravan-achar @akshaychitneni @lresende

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[franciscojavierarceo]

I also found this solution interesting: https://lightning.ai/docs/overview/studios/background-execution

This is neat! It's similar in spirit to some work that's being discussed in Llama Stack on a Jobs API.

CC @booxter

[tarekabouzeid]

This is really interesting and i would love to join as well.

[StefanoFioravanzo]

Label studio is a very interesting project, and I can see how it would be valuable to have a notebook or a vs studio code image with label studio pre-installed. But since it also exposes a web UI and run on the backbone of a database, it seems more fit to propose is a "separate" dev environment. Maintain said image and the deployment options would probably be more of a responsibility of the Notebooks WG (Although maybe with Notebooks 2.0 it will become easier to add a new IDE tot he deployment options)

[juliusvonkohout]

Label studio is a very interesting project, and I can see how it would be valuable to have a notebook or a vs studio code image with label studio pre-installed. But since it also exposes a web UI and run on the backbone of a database, it seems more fit to propose is a "separate" dev environment. Maintain said image and the deployment options would probably be more of a responsibility of the Notebooks WG (Although maybe with Notebooks 2.0 it will become easier to add a new IDE tot he deployment options)

It worked quite well with notebooks V1 since it just uses the underlying pvc as storage and V2 notebooks will make the implementation more elegant. We can also add other tools that way as self-service. A central installation does not make sense, because the community edition has no multi-tenancy, but we get that via our kubeflow user namespaces and self-service per namespace.

[franciscojavierarceo]

Label studio is a very interesting project, and I can see how it would be valuable to have a notebook or a vs studio code image with label studio pre-installed. But since it also exposes a web UI and run on the backbone of a database, it seems more fit to propose is a "separate" dev environment. Maintain said image and the deployment options would probably be more of a responsibility of the Notebooks WG (Although maybe with Notebooks 2.0 it will become easier to add a new IDE tot he deployment options)

My two cents here: it's incredibly useful for the GenAI and tabular story. As it allows MLEs/Data Scientists to label their data.

FWIW, we used it at my last job alongside Kubeflow.