Add dedicated UX for query template building (ML search req processors) #407
+313
−2
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
This PR improves the UX for query configuration/overriding the query via a
query_template
field exposed in the ML search request processors. This is a common pattern for vector search use cases, to generate a valid knn query, injecting the returned vector embedding directly into the new query. Before, this was nested under 'Advanced settings', under 'Query template'. Now, we add a dedicated modal similar to prompt-building for this particular scenario.More details
Demo video, showing the new modal, and the various options shown under "Model outputs" based on any output map configuration.
screen-capture.7.webm
Issues Resolved
Resolves #404
Resolves #407
Check List
--signoff
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.