Proposal: Add a tutorial on model evaluation with ranking metrics in Kubeflow Pipelines

[Kashvi05agarwal]

Hi maintainers,

I’m interested in contributing a small, focused tutorial under samples/tutorials that demonstrates model evaluation in Kubeflow Pipelines, specifically using ranking metrics such as DCG/NDCG with a simple, reproducible pipeline.
While there are good examples around building and composing pipelines, I noticed there are fewer end-to-end examples that focus on how to evaluate models and interpret evaluation results, especially for ranking or recommendation-style problems.
The idea would be to add a minimal tutorial that includes:
a simple pipeline with an evaluation step
a small toy dataset
clear documentation explaining what the metrics mean and how to interpret the output
Before starting any implementation, I wanted to check whether this would be a useful addition and whether this scope sounds reasonable to the maintainers.

Thanks for your time and feedback.

[Kashvi05agarwal]

Hi, just following up in case this discussion was missed.
I wanted to check whether this proposal seems reasonable before starting implementation.
Happy to adjust the scope based on maintainer feedback. Thanks!