Fix llm hp optimization error

[helenxie-bit]

What this PR does / why we need it:
This PR aims to fix errors when using Katib LLM hyperparameter optimization API—which depends on the Trainer SDK v1.9.0—for running the example in the user guide.

Which issue(s) this PR fixes (optional, in Fixes #<issue number>, #<issue number>, ... format, will close the issue(s) when PR gets merged):
Fixes #2575

Checklist:

• Docs included if any changes are user facing

[helenxie-bit]

Please review when you have time @andreyvelich @mahdikhashan . Thank you!

[helenxie-bit]

The E2E test for train API failed due to the following error TypeError: Object of type LoraRuntimeConfig is not JSON serializable. I'm working on fixing it.

Updated 2025-03-31:
I fixed the issue by updating the following line of code:

trainer/sdk/python/kubeflow/training/api/training_client.py

Lines 296 to 298 in 77bd5cd

	json.dumps(
	trainer_parameters.lora_config.__dict__, cls=utils.SetEncoder
	),

to:

json.dumps(trainer_parameters.lora_config.to_dict(), cls=utils.SetEncoder),

This change follows the official documentation, which recommends using LoraConfig.to_dict() for serialization.

@mahdikhashan Can you help test if this fix your issues? Since I remember you've met with the same issue.

[mahdikhashan]

/assign

[andreyvelich]

/rerun-all

[mahdikhashan]

what is the use-case for tensorboard in this changes?

[helenxie-bit]

I believe I've met an error which said the version of tensorboard is not correct, so I explicitly set its version here.

[mahdikhashan]

thank you

[mahdikhashan]

@helenxie-bit thank you for this pr and sorry for the delay. code changes seems good to me, let me check if i can run an example from notebook with this changes.

[helenxie-bit]

/rerun-all

[mahdikhashan]

/lgtm

thank you for your patience

[andreyvelich]

Thank you @helenxie-bit!
/lgtm
/approve

[google-oss-prow]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: andreyvelich

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

The pull request process is described here

Needs approval from an approver in each of these files:

• sdk/python/OWNERS [andreyvelich]

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