From 8504553e61adb72e65bd48b0c9093d31acad300c Mon Sep 17 00:00:00 2001 From: OpenAI Codex Date: Fri, 15 May 2026 05:09:45 +0000 Subject: [PATCH] docs: clarify KFP endpoint setup Signed-off-by: OpenAI Codex --- content/en/docs/components/pipelines/getting-started.md | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/content/en/docs/components/pipelines/getting-started.md b/content/en/docs/components/pipelines/getting-started.md index d49922842e..35d84757eb 100644 --- a/content/en/docs/components/pipelines/getting-started.md +++ b/content/en/docs/components/pipelines/getting-started.md @@ -44,7 +44,7 @@ You can submit the YAML file to a KFP-conformant backend for execution. If you h ```python from kfp.client import Client -client = Client(host='') +client = Client(host='http://localhost:3000') run = client.create_run_from_pipeline_package( 'pipeline.yaml', arguments={ @@ -53,6 +53,12 @@ run = client.create_run_from_pipeline_package( ) ``` +Replace `http://localhost:3000` with the endpoint for your Kubeflow Pipelines deployment. For a local standalone deployment, port-forward the `ml-pipeline-ui` service first: + +```sh +kubectl port-forward --namespace kubeflow svc/ml-pipeline-ui 3000:80 +``` + The client will print a link to view the pipeline execution graph and logs in the UI. In this case, the pipeline has one task that prints and returns `'Hello, World!'`. ## Next steps