kubeflow
katib: Update LLM HP tuning guide to clarify tunable fields and fix resource section
…es config, remove unsupported params
Checklist:
- [x] You have signed off your commits
- [x] Ensure you follow best practices from our guide. Contributing.
- [x] You have included screenshots when changing the website style or adding a new page.
Description of your changes:
As per andrey's recommendation I added the changes mentioned in the following screenshots attached. A minimal work.
Issue
Closes: #2522
Labels
/area katib
/area website
Hi @SanthoshToorpu. Thanks for your PR.
I'm waiting for a kubeflow member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.
Once the patch is verified, the new status will be reflected by the ok-to-test label.
I understand the commands that are listed here.
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.
![google-oss-prow[bot] avatar](https://avatars.githubusercontent.com/in/143327?v=4 "Published by a pub.dev verified publisher"){kind=small}
@andreyvelich as per @helenxie-bit instructions and after reading the thread i fixed 2/3 issues except for the params you told to remove irrelevant params.
I downloaded the docker image rn but can you be more specific on what to be removed and what not?
@andreyvelich as per @helenxie-bit instructions and after reading the thread i fixed 2/3 issues except for the params you told to remove irrelevant params.
I downloaded the docker image rn but can you be more specific on what to be removed and what not?
I think we need to remove unused parameters in this part: https://www.kubeflow.org/docs/components/katib/user-guides/llm-hp-optimization/#key-parameters-for-llm-hyperparameter-tuning. Can you remove parameters objective, base_image, and parameters? Since they will not be used when optimizing hyperparameters for LLMs.
@andreyvelich as per @helenxie-bit instructions and after reading the thread i fixed 2/3 issues except for the params you told to remove irrelevant params.
I downloaded the docker image rn but can you be more specific on what to be removed and what not?
I think we need to remove unused parameters in this part: https://www.kubeflow.org/docs/components/katib/user-guides/llm-hp-optimization/#key-parameters-for-llm-hyperparameter-tuning. Can you remove parameters objective, base_image, and parameters? Since they will not be used when optimizing hyperparameters for LLMs.
Okay so only those threee params?
@andreyvelich as per @helenxie-bit instructions and after reading the thread i fixed 2/3 issues except for the params you told to remove irrelevant params.
I think we need to remove unused parameters in this part: https://www.kubeflow.org/docs/components/katib/user-guides/llm-hp-optimization/#key-parameters-for-llm-hyperparameter-tuning. Can you remove parameters objective, base_image, and parameters? Since they will not be used when optimizing hyperparameters for LLMs.
Okay so only those threee params?
Yes, all other parameters may be used when optimizing hyperparameters for LLMs.
@helenxie-bit :Great! Maybe here it's better to use a different title since we include S3DatasetParams in this part too:
how about
Dataset and Model Parameter Classes
in legacy trainer docs
@helenxie-bit :Great! Maybe here it's better to use a different title since we include S3DatasetParams in this part too:
how about
Dataset and Model Parameter Classes
in legacy trainer docs
That sounds good to me.
Thanks for the contribution! LGTM! Please have a review when you have time @andreyvelich @mahdikhashan
@andreyvelich @Electronic-Waste is this ready to be merged ?
I think It was pending for my review, I had a quick look and seems good to me to be merged. as i remember it to be a link to old doc and further improvment.
[APPROVALNOTIFIER] This PR is APPROVED
This pull-request has been approved by: juliusvonkohout
The full list of commands accepted by this bot can be found here.
The pull request process is described here
- ~~content/en/docs/components/katib/OWNERS~~ [juliusvonkohout]
- ~~content/en/docs/components/trainer/OWNERS~~ [juliusvonkohout]
Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment