[Doc] Develop Ray Serve Python script on KubeRay #1250
Conversation
There was a problem hiding this comment.
Looks great, just some minor comments.
Near the beginning, it would be good to have some sort of summary like
"In this tutorial, you will learn how to easily debug your RayService by manually running a Ray Serve script against a RayCluster. This is a useful strategy for debugging a RayService because (... you can either locally using kind or on a remote cluster, or other reasons...)"
I'm guessing that compared to RayService, the iteration speed is faster with this approach? Or it's easier to see the underlying error message?
docs/guidance/rayserve-dev-doc.md
Outdated
|
|
||
| Many RayService issues are related to the Ray Serve Python scripts. | ||
| We strongly recommend that users develop and test their Ray Serve scripts in a RayCluster using the same Ray image before deploying them to a RayService. | ||
| In this tutorial, we will show you how to develop [ray-service.mobilenet.yaml](https://github.com/ray-project/kuberay/blob/master/ray-operator/config/samples/ray-service.mobilenet.yaml). |
There was a problem hiding this comment.
Would be better to add a 1 sentence description of what mobilenet is and why we're using it in the tutorial
Co-authored-by: Archit Kulkarni <[email protected]> Signed-off-by: Kai-Hsun Chen <[email protected]>
Updated: f656a58 |
Develop Ray Serve Python script on KubeRay
Develop Ray Serve Python script on KubeRay
Develop Ray Serve Python script on KubeRay
Why are these changes needed?
Related issue number
Closes #1176
Checks