Update integrated vectorization

Azure-Samples · Oct 16, 2024 · d4e40b8 · d4e40b8
1 parent e99f6e2
commit d4e40b8
Show file tree

Hide file tree

Showing 4 changed files with 27 additions and 24 deletions.
diff --git a/app/backend/prepdocslib/integratedvectorizerstrategy.py b/app/backend/prepdocslib/integratedvectorizerstrategy.py
@@ -6,16 +6,15 @@
 )
 from azure.search.documents.indexes.models import (
     AzureOpenAIEmbeddingSkill,
-    AzureOpenAIParameters,
-    AzureOpenAIVectorizer,
     FieldMapping,
     IndexProjectionMode,
     InputFieldMappingEntry,
     OutputFieldMappingEntry,
     SearchIndexer,
     SearchIndexerDataContainer,
     SearchIndexerDataSourceConnection,
-    SearchIndexerIndexProjections,
+    SearchIndexerDataSourceType,
+    SearchIndexerIndexProjection,
     SearchIndexerIndexProjectionSelector,
     SearchIndexerIndexProjectionsParameters,
     SearchIndexerSkillset,
@@ -67,6 +66,7 @@ async def create_embedding_skill(self, index_name: str):
         skillset_name = f"{index_name}-skillset"
 
         split_skill = SplitSkill(
+            name=f"{index_name}-split-skill",
             description="Split skill to chunk documents",
             text_split_mode="pages",
             context="/document",
@@ -82,17 +82,20 @@ async def create_embedding_skill(self, index_name: str):
             raise ValueError("Expecting Azure Open AI instance")
 
         embedding_skill = AzureOpenAIEmbeddingSkill(
+            name=f"{index_name}-embedding-skill",
             description="Skill to generate embeddings via Azure OpenAI",
             context="/document/pages/*",
-            resource_uri=f"https://{self.embeddings.open_ai_service}.openai.azure.com",
-            deployment_id=self.embeddings.open_ai_deployment,
+            resource_url=f"https://{self.embeddings.open_ai_service}.openai.azure.com",
+            deployment_name=self.embeddings.open_ai_deployment,
+            model_name=self.embeddings.open_ai_model_name,
+            dimensions=self.embeddings.open_ai_dimensions,
             inputs=[
                 InputFieldMappingEntry(name="text", source="/document/pages/*"),
             ],
             outputs=[OutputFieldMappingEntry(name="embedding", target_name="vector")],
         )
 
-        index_projections = SearchIndexerIndexProjections(
+        index_projection = SearchIndexerIndexProjection(
             selectors=[
                 SearchIndexerIndexProjectionSelector(
                     target_index_name=index_name,
@@ -114,7 +117,7 @@ async def create_embedding_skill(self, index_name: str):
             name=skillset_name,
             description="Skillset to chunk documents and generate embeddings",
             skills=[split_skill, embedding_skill],
-            index_projections=index_projections,
+            index_projection=index_projection,
         )
 
         return skillset
@@ -132,25 +135,14 @@ async def setup(self):
         if self.embeddings is None:
             raise ValueError("Expecting Azure Open AI instance")
 
-        await search_manager.create_index(
-            vectorizers=[
-                AzureOpenAIVectorizer(
-                    name=f"{self.search_info.index_name}-vectorizer",
-                    kind="azureOpenAI",
-                    azure_open_ai_parameters=AzureOpenAIParameters(
-                        resource_uri=f"https://{self.embeddings.open_ai_service}.openai.azure.com",
-                        deployment_id=self.embeddings.open_ai_deployment,
-                    ),
-                ),
-            ]
-        )
+        await search_manager.create_index()
 
         # create indexer client
         ds_client = self.search_info.create_search_indexer_client()
         ds_container = SearchIndexerDataContainer(name=self.blob_manager.container)
         data_source_connection = SearchIndexerDataSourceConnection(
             name=f"{self.search_info.index_name}-blob",
-            type="azureblob",
+            type=SearchIndexerDataSourceType.AZURE_BLOB,
             connection_string=self.blob_manager.get_managedidentity_connectionstring(),
             container=ds_container,
             data_deletion_detection_policy=NativeBlobSoftDeleteDeletionDetectionPolicy(),

diff --git a/app/backend/prepdocslib/searchmanager.py b/app/backend/prepdocslib/searchmanager.py
@@ -4,6 +4,8 @@
 from typing import List, Optional
 
 from azure.search.documents.indexes.models import (
+    AzureOpenAIVectorizer,
+    AzureOpenAIVectorizerParameters,
     HnswAlgorithmConfiguration,
     HnswParameters,
     SearchableField,
@@ -174,12 +176,21 @@ async def create_index(self, vectorizers: Optional[List[VectorSearchVectorizer]]
                         VectorSearchProfile(
                             name="embedding_config",
                             algorithm_configuration_name="hnsw_config",
-                            vectorizer=(
+                            vectorizer_name=(
                                 f"{self.search_info.index_name}-vectorizer" if self.use_int_vectorization else None
                             ),
                         ),
                     ],
-                    vectorizers=vectorizers,
+                    vectorizers=[
+                        AzureOpenAIVectorizer(
+                            vectorizer_name=f"{self.search_info.index_name}-vectorizer",
+                            parameters=AzureOpenAIVectorizerParameters(
+                                resource_url=f"https://{self.embeddings.open_ai_service}.openai.azure.com",
+                                deployment_name=self.embeddings.open_ai_deployment,
+                                model_name=self.embeddings.open_ai_model_name,
+                            ),
+                        ),
+                    ],
                 ),
             )
             if self.search_info.index_name not in [name async for name in search_index_client.list_index_names()]:

diff --git a/app/backend/requirements.in b/app/backend/requirements.in
@@ -7,7 +7,7 @@ tiktoken
 tenacity
 azure-ai-documentintelligence
 azure-cognitiveservices-speech
-azure-search-documents==11.6.0b5
+azure-search-documents==11.6.0b6
 azure-storage-blob
 azure-storage-file-datalake
 uvicorn

diff --git a/app/backend/requirements.txt b/app/backend/requirements.txt
@@ -52,7 +52,7 @@ azure-monitor-opentelemetry==1.6.1
     # via -r requirements.in
 azure-monitor-opentelemetry-exporter==1.0.0b28
     # via azure-monitor-opentelemetry
-azure-search-documents==11.6.0b1
+azure-search-documents==11.6.0b6
     # via -r requirements.in
 azure-storage-blob==12.22.0
     # via