-
Notifications
You must be signed in to change notification settings - Fork 8.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Align inference pipeline card/option styling #172952
Conversation
...ions/enterprise_search_content/components/search_index/pipelines/inference_pipeline_card.tsx
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice work!
...ions/enterprise_search_content/components/search_index/pipelines/inference_pipeline_card.tsx
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, few non-blocking comments.
...enterprise_search_content/components/search_index/pipelines/inference_pipeline_card.test.tsx
Show resolved
Hide resolved
const actionButton = ( | ||
<EuiButtonEmpty |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I see some discrepancies with the model-related subcomponents here vs in the model selection component within the pipeline configuration flyout. Let's make sure we address these in the unified ML design:
- The status badge is part of the "..." button here. In the model selection list it's not.
- We display a wrench button to navigate to the Trained Models page if the model is not deployed vs in the model selection list we don't.
- The model status badge here works with the reduced set of states (implemented in
TrainedModelHealth
), for example the downloading/downloaded states all map to "Not started". If we want to display more fine-grained states, we'll need to use model details fromMlModel
in the pipeline card. - The model type is rendered as a badge here (
text_embedding
) vs a user-friendly title in the model selector ("Dense Vector Text Embedding"). (We may not have the space the display the full title here.)
I'll link to this comment in the unified UX Jira.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good callouts! I could address the first point in this PR by shrinking the button to not include the status badge, but IMO it's nice to have a larger button area, especially since there's no other click action to associate with the status badge. WDYT?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
To me it doesn't come intuitively that a badge is clickable. I'd either restrict clickability to the "..." button only, or turn the status badge into a (non-empty) button. But this is a good question to the UX folks, maybe they have some suggestions.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok, since this an open question for the UX folks, I will keep this as-is for now and we can adjust in a follow-up PR.
...ions/enterprise_search_content/components/search_index/pipelines/inference_pipeline_card.tsx
Show resolved
Hide resolved
...ions/enterprise_search_content/components/search_index/pipelines/inference_pipeline_card.tsx
Outdated
Show resolved
Hide resolved
LGTM |
… act as links" This reverts commit 95a0506.
💚 Build Succeeded
Metrics [docs]Async chunks
History
To update your PR or re-run it, just comment with: |
## Summary More closely align the styling used for inference pipeline cards on the indices page and existing inference pipeline options on the "Add an inference pipeline" flyout. Also used this as an opportunity to refactor `inference_pipeline_card.tsx` to improve readability and change the model/pipeline management popover to make the "Fix issue in Trained Models" button fit better with the other buttons. (cherry picked from commit 1abc0c2)
💚 All backports created successfully
Note: Successful backport PRs will be merged automatically after passing CI. Questions ?Please refer to the Backport tool documentation |
# Backport This will backport the following commits from `main` to `8.12`: - [Align inference pipeline card/option styling (#172952)](#172952) <!--- Backport version: 8.9.7 --> ### Questions ? Please refer to the [Backport tool documentation](https://github.com/sqren/backport) <!--BACKPORT [{"author":{"name":"Mike Pellegrini","email":"[email protected]"},"sourceCommit":{"committedDate":"2023-12-11T15:21:32Z","message":"Align inference pipeline card/option styling (#172952)\n\n## Summary\r\n\r\nMore closely align the styling used for inference pipeline cards on the\r\nindices page and existing inference pipeline options on the \"Add an\r\ninference pipeline\" flyout.\r\n\r\nAlso used this as an opportunity to refactor\r\n`inference_pipeline_card.tsx` to improve readability and change the\r\nmodel/pipeline management popover to make the \"Fix issue in Trained\r\nModels\" button fit better with the other buttons.","sha":"1abc0c2df75110512bd1069fe1cbd208b0895052","branchLabelMapping":{"^v8.13.0$":"main","^v(\\d+).(\\d+).\\d+$":"$1.$2"}},"sourcePullRequest":{"labels":["release_note:skip","Team:EnterpriseSearch","v8.12.0","v8.13.0"],"number":172952,"url":"https://github.com/elastic/kibana/pull/172952","mergeCommit":{"message":"Align inference pipeline card/option styling (#172952)\n\n## Summary\r\n\r\nMore closely align the styling used for inference pipeline cards on the\r\nindices page and existing inference pipeline options on the \"Add an\r\ninference pipeline\" flyout.\r\n\r\nAlso used this as an opportunity to refactor\r\n`inference_pipeline_card.tsx` to improve readability and change the\r\nmodel/pipeline management popover to make the \"Fix issue in Trained\r\nModels\" button fit better with the other buttons.","sha":"1abc0c2df75110512bd1069fe1cbd208b0895052"}},"sourceBranch":"main","suggestedTargetBranches":["8.12"],"targetPullRequestStates":[{"branch":"8.12","label":"v8.12.0","labelRegex":"^v(\\d+).(\\d+).\\d+$","isSourceBranch":false,"state":"NOT_CREATED"},{"branch":"main","label":"v8.13.0","labelRegex":"^v8.13.0$","isSourceBranch":true,"state":"MERGED","url":"https://github.com/elastic/kibana/pull/172952","number":172952,"mergeCommit":{"message":"Align inference pipeline card/option styling (#172952)\n\n## Summary\r\n\r\nMore closely align the styling used for inference pipeline cards on the\r\nindices page and existing inference pipeline options on the \"Add an\r\ninference pipeline\" flyout.\r\n\r\nAlso used this as an opportunity to refactor\r\n`inference_pipeline_card.tsx` to improve readability and change the\r\nmodel/pipeline management popover to make the \"Fix issue in Trained\r\nModels\" button fit better with the other buttons.","sha":"1abc0c2df75110512bd1069fe1cbd208b0895052"}}]}] BACKPORT--> Co-authored-by: Mike Pellegrini <[email protected]>
Summary
More closely align the styling used for inference pipeline cards on the indices page and existing inference pipeline options on the "Add an inference pipeline" flyout.
Also used this as an opportunity to refactor
inference_pipeline_card.tsx
to improve readability and change the model/pipeline management popover to make the "Fix issue in Trained Models" button fit better with the other buttons.Old:
Inference pipeline cards:
Desktop:
Mobile:
Popover (w/o "Fix issue in Trained Models"):
Popover (w/ "Fix issue in Trained Models"):
Existing inference pipeline options:
Desktop:
Mobile:
New:
Inference pipeline cards:
Desktop:
Mobile:
Popover (w/o "Fix issue in Trained Models"):
Popover (w/ "Fix issue in Trained Models"):
Existing inference pipeline options:
Desktop:
Mobile:
Checklist