Python: ACS should use Hnsw for vector indexing (microsoft#2534)

### Motivation and Context

This PR builds on microsoft#2435
- the previous PR updates azure-search to the latest package , where it
is recommended to use HnswVectorSearchAlgorithmConfiguration. The
previous implementation appears to be in a semi-deprecated state.

This new flow is the publicly recommended use for ACS according to
https://github.com/Azure/cognitive-search-vector-pr/tree/main/demo-python/code.

### Description
- update `azure-search-documents` dependency from `11.4.0b6 -> 11.4.0b8`
- index configuration changed from `VectorSearchAlgorithmConfiguration
-> HnswVectorSearchAlgorithmConfiguration`
- update `search_client.search()` call in `get_nearest_matches()` to be
consistent with the new index api

### Contribution Checklist

<!-- Before submitting this PR, please make sure: -->

- [ ] The code builds clean without any errors or warnings
- [ ] The PR follows the [SK Contribution
Guidelines](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md)
and the [pre-submission formatting
script](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md#development-scripts)
raises no violations
- [ ] All unit tests pass, and I have added new tests where possible
- [ ] I didn't break anyone 😄

---------

Co-authored-by: Kinfey <[email protected]>
Co-authored-by: Lee Miller <[email protected]>
Co-authored-by: Kinfey <[email protected]>

Author: 3 people

python/poetry.lock

@@ -56,7 +56,7 @@ psycopg = "^3.1.9"
 psycopg-binary = "^3.1.9"
 
 [tool.poetry.group.azure_cognitive_search.dependencies]
-azure-search-documents = {version = "11.4.0b6", allow-prereleases = true}
+azure-search-documents = {version = "11.4.0b8", allow-prereleases = true}
 azure-core = "^1.28.0"
 azure-identity = "^1.13.0"
 

python/pyproject.toml

@@ -8,10 +8,11 @@
 from azure.core.exceptions import ResourceNotFoundError
 from azure.search.documents.indexes.aio import SearchIndexClient
 from azure.search.documents.indexes.models import (
+    HnswVectorSearchAlgorithmConfiguration,
     SearchIndex,
     VectorSearch,
-    VectorSearchAlgorithmConfiguration,
 )
+from azure.search.documents.models import Vector
 from numpy import ndarray
 
 from semantic_kernel.connectors.memory.azure_cognitive_search.utils import (
@@ -58,7 +59,6 @@ def __init__(
         Instantiate using Async Context Manager:
             async with AzureCognitiveSearchMemoryStore(<...>) as memory:
                 await memory.<...>
-
         """
         try:
             pass
@@ -82,14 +82,14 @@ async def close_async(self):
     async def create_collection_async(
         self,
         collection_name: str,
-        vector_config: Optional[VectorSearchAlgorithmConfiguration] = None,
+        vector_config: Optional[HnswVectorSearchAlgorithmConfiguration] = None,
     ) -> None:
         """Creates a new collection if it does not exist.
 
         Arguments:
             collection_name {str}                              -- The name of the collection to create.
-            vector_config {VectorSearchAlgorithmConfiguration} -- Optional search algorithm configuration
-                                                                  (default: {None}).
+            vector_config {HnswVectorSearchAlgorithmConfiguration} -- Optional search algorithm configuration
+                                                                      (default: {None}).
             semantic_config {SemanticConfiguration}            -- Optional search index configuration (default: {None}).
         Returns:
             None
@@ -100,7 +100,7 @@ async def create_collection_async(
         else:
             vector_search = VectorSearch(
                 algorithm_configurations=[
-                    VectorSearchAlgorithmConfiguration(
+                    HnswVectorSearchAlgorithmConfiguration(
                         name="az-vector-config",
                         kind="hnsw",
                         hnsw_parameters={
@@ -403,12 +403,14 @@ async def get_nearest_matches_async(
             collection_name.lower()
         )
 
+        vector = Vector(
+            value=embedding.flatten(), k=limit, fields=SEARCH_FIELD_EMBEDDING
+        )
+
         search_results = await search_client.search(
             search_text="*",
-            vector_fields=SEARCH_FIELD_EMBEDDING,
-            vector=embedding.tolist(),
+            vectors=[vector],
             select=get_field_selection(with_embeddings),
-            top_k=limit,
         )
 
         if not search_results or search_results is None: