Azure / Azure/azure-search-vector-samples
KNN returning values that don't adhere to the filter
- Dominant language
- Jupyter Notebook
- Stars
- 910
- Forks
- 377
- PR merge metrics
- No merged PRs in 30d
Description
This may be 'by design', but I hadn't expected it.
I'm using a filter which only returns matching projectId/ownerId values, but I'm setting TopK to 100.
In the result set, I'm finding entries where the 'ownerId' isn't one of the filtered values.
Does the nearest neighbor calculation not adhere to the filter I'm assigning? i.e. can it return results which are neighbors, but may not match the filters I've assigned?
```
type eq 'Content' and projectId eq 'f9b101c3-c2e4-4239-bfb9-66de4ef37f7d' and (ownerId eq 'c1f8c39e-e3ac-5e18-a282-5cde33c9d92a' or ownerId eq 'f9b101c3-c2e4-4239-bfb9-66de4ef37f7d') and (content/vectorType eq 'Content' or content/vectorType eq 'Page' or content/vectorType eq 'Chunk' or content/vectorType eq 'Segment' or content/vectorType eq 'Phrase' or content/vectorType eq 'Image')
```
Contributor guide
No contributing guide indexed for this repository
Research direction
The issue names no file, test, or entry point. Start by reproducing the shown KNN query with TopK 100 and the supplied filter, then check whether returned ownerId values match it; done means establishing the documented filter semantics or confirming a defect.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- azure
- Domain
- cloud, search
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100