Closed LaVLaS closed 9 months ago
Given that the PoC currently works both on ODH and RHODS, is it really important to remove that RHODS reference?
The intent was not to hide RHODS. As we take the knowledge gained from the Proof of Concept and use that to develop the feature in ODH, none of the RHODS releases planned for this year will have any interations of the work that is being developed.
but at this point, there is nothing ODH would be needed for, except for training the model.
In the context of the PoC it was optional but we are working to integrate these external workflows within ODH so the trained model can be packaged in a container, tested and pushed from within ODH
I don't mind removing the ODH 1.X reference but it's weird to do so when 2.X is still designated alpha in
ODH 1.x is planned for deprecation with the 1.11 release (See opendatahub-operator #648). The alpha
reference in opendatahub.io is just an oversight as it was used as a preview for the downstream release that started in July. We accelerated the development of v2.x and the opendatahub.io documentation for the ODH install just needs a refresh
If I had a vote, this would be a nack.
Noted. Your input is still appreciated.
The intent was not to hide RHODS. As we take the knowledge gained from the Proof of Concept and use that to develop the feature in ODH, none of the RHODS releases planned for this year will have any interations of the work that is being developed.
Which in turn means that if you want to see the approach on RHODS, for the time being this is the latest greatest you can get.
In the context of the PoC it was optional but we are working to integrate these external workflows within ODH so the trained model can be packaged in a container, tested and pushed from within ODH
OK. So is the plan to iterate in this repository on those integrations?
ODH 1.x is planned for deprecation with the 1.11 release (See opendatahub-operator #648). The
alpha
reference in opendatahub.io is just an oversight as it was used as a preview for the downstream release that started in July. We accelerated the development of v2.x and the opendatahub.io documentation for the ODH install just needs a refresh
Thanks for that https://github.com/opendatahub-io/opendatahub-operator/issues/648 link, that sums it nicely.
@LaVLaS: The following test failed, say /retest
to rerun all failed tests or /retest-required
to rerun all mandatory failed tests:
Test name | Commit | Details | Required | Rerun command |
---|---|---|---|---|
ci/prow/test-ai-edge | c84329318feb2d06ec40d27f69d223f55374db57 | link | true | /test test-ai-edge |
Full PR test history. Your PR dashboard.
PR needs rebase.
Closing this to follow-up on with another PR to update the documentation across the board
Description
Updates the README to show the ODH version and expectations for feature development. Removes references to RHODS as the features implemented will not be available in the downstream product until they are incubated in ODH until
How Has This Been Tested?
N/A
Merge criteria:
Information is clear and provides a quick high level summary of ODH integration