Support Vector Search Nested benchmark #584
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Finn Roblin <[email protected]>
Signed-off-by: Finn Roblin <[email protected]>
Signed-off-by: Finn Roblin <[email protected]>
finnroblin
requested review from
IanHoang,
gkamat,
beaioun,
cgchinmay,
rishabh6788 and
VijayanB
as code owners
VijayanB
reviewed
Jul 19, 2024
osbenchmark/workload/params.py
Outdated
| @@ -884,6 +884,7 @@ class VectorDataSetPartitionParamSource(ParamSource): | |||
| def __init__(self, workload, params, context: Context, **kwargs): | |||
| super().__init__(workload, params, **kwargs) | |||
| self.field_name: str = parse_string_parameter("field", params) | |||
| self.is_nested = "." in self.field_name # in base class because used for both bulk ingest and queries. | |||
VijayanB
reviewed
Jul 19, 2024
| @@ -1155,6 +1156,16 @@ def _build_vector_search_query_body(self, vector, efficient_filter=None) -> dict | |||
| query.update({ | |||
| "filter": efficient_filter, | |||
| }) | |||
|
|
|||
| if self.is_nested: | |||
There was a problem hiding this comment.
if more than one level nested is identified, lets raise an exception and add TODO to fix that later
Signed-off-by: Finn Roblin <[email protected]>
…ct#546 Signed-off-by: Finn Roblin <[email protected]>
Signed-off-by: Finn Roblin <[email protected]>
VijayanB
reviewed
Aug 16, 2024
osbenchmark/workload/params.py
Outdated
| @@ -884,6 +884,8 @@ class VectorDataSetPartitionParamSource(ParamSource): | |||
| def __init__(self, workload, params, context: Context, **kwargs): | |||
| super().__init__(workload, params, **kwargs) | |||
| self.field_name: str = parse_string_parameter("field", params) | |||
| self.NESTED_FIELD_SEPARATOR = "." | |||
VijayanB
reviewed
Aug 16, 2024
osbenchmark/workload/params.py
Outdated
| "knn": { | ||
| self.field_name: query, | ||
| }, | ||
| } | ||
|
|
||
| if self.is_nested: | ||
| outer_field_name, _inner_field_name = self.get_split_fields() |
There was a problem hiding this comment.
Suggested change
| outer_field_name, _inner_field_name = self.get_split_fields() | |
| outer_field_name, _ = self.get_split_fields() |
|
@finnroblin looks good, please resolve conflict and rebase your change. |
Signed-off-by: Finn Roblin <[email protected]>
|
LGTM. Thanks |
VijayanB
approved these changes
Aug 22, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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 change supports nested field benchmarks for vector search. Previously k-NN maintainers would need to use the deprecated perf tool to benchmark nested fields.
Issues Resolved
This PR is part of the effort to port perf tool benchmarks to OSB.
Testing
Unit tests + tested E2E with modified nested sift euclidian dataset (workload PR forthcoming)
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.