sync: main to stable#764
Conversation
* feat(catalog): Add performance artifacts endpoint with Pareto filtering
Implements new GET /api/model_catalog/v1alpha1/sources/{source_id}/models/{model_name}/artifacts/performance
endpoint that returns performance metrics artifacts, optionally filtered for Pareto-optimal configurations.
Key changes:
- Add getAllModelPerformanceArtifacts OpenAPI operation and handler
- Support targetRPS and recommendations query parameters for filtering
- Calculate replicas and total_requests_per_second based on targetRPS
- Add database models and service layer for performance artifacts
Signed-off-by: Paul Boyd <paul@pboyd.io>
* fix(catalog): performance metrics edge case
If two performance metrics artifacts have the same latency and cost,
prefer the one with the lower hardware count.
Signed-off-by: Paul Boyd <paul@pboyd.io>
---------
Signed-off-by: Paul Boyd <paul@pboyd.io>
…1959) * Created Vards wrap between sections in Model overview page Signed-off-by: Yulia Krimerman <juliapiterova@hotmail.com> * addressed nit Signed-off-by: Yulia Krimerman <juliapiterova@hotmail.com> --------- Signed-off-by: Yulia Krimerman <juliapiterova@hotmail.com>
…operties (kubeflow#1961) * feat: load size, tensor type and variant group ID from metadata as customProperties Signed-off-by: lugi0 <lgiorgi@redhat.com> * fix: two tests shoudl expect error Signed-off-by: lugi0 <lgiorgi@redhat.com> --------- Signed-off-by: lugi0 <lgiorgi@redhat.com>
…beflow#1955) * chore(catalog): Don't deploy demo data by default Move the demo data to an overlay, leaving the base empty. Tilt still deploys the demo data by default. Also splitting the performance data into ndjson files to help with testing the performance metrics loading code. The istio manifests were taking the "overlay" name, so I'm moving them out of the way (into `options/istio`). Signed-off-by: Paul Boyd <paul@pboyd.io> * chore(catalog): support local manifests in tilt If someone defines "local" overlay manifests for the catalog, use it. Signed-off-by: Paul Boyd <paul@pboyd.io> --------- Signed-off-by: Paul Boyd <paul@pboyd.io>
Always return items from the sources endpoint, even if it's an empty list. Signed-off-by: Paul Boyd <paul@pboyd.io>
* feat: add preview endpoint in model catalog Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * feat: improve openapi spec on preview endpoint Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * fix: make builds work with mock function Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * feat: implement preview logic for yaml sources Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * fix: remove unused fields from the openapi schema Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * fix: sync gitattributes file Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * feat: models can be passed to the preview endpoint Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * chore: improve doc and errors Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * fix: panic when using pageSize < 1 Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * feat: add support to preview hugging face sources Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * feat: add hf maxModels safeguard Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * fix: ignore url properties in sources to guard on SSRF attacks Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * fix: let users use a full source in preview Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * feat: log when an user tries to use a custom url for hf Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> * fix: catalog unit tests Signed-off-by: Alessio Pragliola <seth.pro@gmail.com> --------- Signed-off-by: Alessio Pragliola <seth.pro@gmail.com>
[pull] main from kubeflow:main
|
Important Review skippedBot user detected. To trigger a single review, invoke the You can disable this status message by setting the Comment |
|
Hi @github-actions[bot]. Thanks for your PR. I'm waiting for a github.com member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. DetailsInstructions 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-sigs/prow repository. |
…eflow#1975) Signed-off-by: manaswinidas <dasmanaswini10@gmail.com>
…ubeflow#1976) Implements - Sources endpoint should return all sources regardless of the enabled field's value 🤖 Generated with [Claude Code](https://claude.ai/code) Signed-off-by: Chris Hambridge <chambrid@redhat.com> Co-authored-by: Claude <noreply@anthropic.com>
Signed-off-by: Adysen Rothman <85646824+adysenrothman@users.noreply.github.com>
[pull] main from kubeflow:main
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## stable #764 +/- ##
==========================================
+ Coverage 39.56% 40.95% +1.39%
==========================================
Files 150 152 +2
Lines 18185 19013 +828
Branches 162 162
==========================================
+ Hits 7195 7787 +592
- Misses 10424 10590 +166
- Partials 566 636 +70 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: adysenrothman, github-actions[bot] The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
…io#764) Signed-off-by: Alessio Pragliola <seth.pro@gmail.com>
sync-branches: New code has just landed in main, so let's bring stable up to speed!