How do we specify which function objects are being returned as Artifacts using the v2 @component decorator?

[tmbluth]

In the code example here (https://www.kubeflow.org/docs/components/pipelines/sdk-v2/v2-component-io/#use-v2-component-decorator-for-python-function-based-components) the output that is specified is Output[Model] and OutputPath. Its unclear what objects are going to be returned to these 2 outputs given there are no objects being returned. How do we tell this code what those output objects are? Are they inferred?

If so, we'd run into an issue if a component trains several models but only want one of those models as output (or it could be several evaluations, datasets, paths, etc). Perhaps the answer is documented elsewhere, but I've been unable to find it. I had assumed the v2 component I/O page might have that detail.

Could someone help me understand this or point me in the right direction? It would be greatly appreciated.

[varodrig]

@tmbluth documentation has changed please review https://www.kubeflow.org/docs/components/pipelines/user-guides/data-handling/artifacts/ and https://www.kubeflow.org/docs/components/pipelines/user-guides/core-functions/

let us know if you are still facing issues otherwise this issue will be closed soon.
are you still facing issues? Issue will be closed if there's no activity. Due date is Jan 31th.

[varodrig]

/close

[google-oss-prow]

@varodrig: Closing this issue.

In response to this:

/close

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.

[varodrig]

@tmbluth we are closing this issue due to the inactivity and it seems the issue was resolved. Please let us know if you are still facing the issue.