Azure OpenAI LangChain - (InvalidField) The vector field 'content_vector' must have the property 'vectorSearchConfiguration' set

Question

I'm trying to create an embedding vector database with some .txt documents in my local folder. In particular I'm following this tutorial from the official page of LangChain: LangChain - Azure Cognitive Search and Azure OpenAI. I have followed all the steps of the tutorial and this is my Python script:

# From https://python.langchain.com/docs/integrations/vectorstores/azuresearch


import openai
import os
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores.azuresearch import AzureSearch


os.environ["OPENAI_API_TYPE"] = "azure"
os.environ["OPENAI_API_BASE"] = "https://xxxxxx.openai.azure.com"
os.environ["OPENAI_API_KEY"] = "xxxxxxxxx"
os.environ["OPENAI_API_VERSION"] = "2023-05-15"

model: str = "text-embedding-ada-002"


vector_store_address: str = "https://xxxxxxx.search.windows.net"
vector_store_password: str = "xxxxxxx"



embeddings: OpenAIEmbeddings = OpenAIEmbeddings(deployment=model, chunk_size=1)
index_name: str = "cognitive-search-openai-exercise-index"
vector_store: AzureSearch = AzureSearch(
    azure_search_endpoint=vector_store_address,
    azure_search_key=vector_store_password,
    index_name=index_name,
    embedding_function=embeddings.embed_query,
)


from langchain.document_loaders import TextLoader
from langchain.text_splitter import CharacterTextSplitter

loader = TextLoader("C:/Users/xxxxxxxx/azure_openai_cognitive_search_exercise/data/qna/a.txt", encoding="utf-8")

documents = loader.load()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
docs = text_splitter.split_documents(documents)

vector_store.add_documents(documents=docs)




# Perform a similarity search
docs = vector_store.similarity_search(
    query="Who is Pippo Franco?",
    k=3,
    search_type="similarity",
)
print(docs[0].page_content)

Now, when I run the script I get the following error:

vector_search_configuration is not a known attribute of class <class 'azure.search.documents.indexes.models._index.SearchField'> and will be ignored
algorithm_configurations is not a known attribute of class <class 'azure.search.documents.indexes._generated.models._models_py3.VectorSearch'> and will be ignored
Traceback (most recent call last):
  File "C:\Users\xxxxxxxxx\venv\Lib\site-packages\langchain\vectorstores\azuresearch.py", line 105, in _get_search_client
    index_client.get_index(name=index_name)
  File "C:\Users\xxxxxxx\venv\Lib\site-packages\azure\core\tracing\decorator.py", line 78, in wrapper_use_tracer
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xxxxxxx\KYF\venv\Lib\site-packages\azure\search\documents\indexes\_search_index_client.py", line 145, in get_index
    result = self._client.indexes.get(name, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xxxxxx\venv\Lib\site-packages\azure\core\tracing\decorator.py", line 78, in wrapper_use_tracer
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xxxxxx\KYF\venv\Lib\site-packages\azure\search\documents\indexes\_generated\operations\_indexes_operations.py", 
line 864, in get
    map_error(status_code=response.status_code, response=response, error_map=error_map)
  File "C:\Users\xxxxxxxx\venv\Lib\site-packages\azure\core\exceptions.py", line 165, in map_error
    raise error
azure.core.exceptions.ResourceNotFoundError: () No index with the name 'cognitive-search-openai-exercise-index' was found in the service 'cognitive-search-openai-exercise'.
Code:
Message: No index with the name 'cognitive-search-openai-exercise-index' was found in the service 'cognitive-search-openai-exercise'.      

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "c:\Users\xxxxxxx\venv\azure_openai_cognitive_search_exercise\test.py", line 25, in <module>
    vector_store: AzureSearch = AzureSearch(
                                ^^^^^^^^^^^^
  File "C:\Users\xxxxxxx\venv\Lib\site-packages\langchain\vectorstores\azuresearch.py", line 237, in __init__
    self.client = _get_search_client(
                  ^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xxxxxxxx\venv\Lib\site-packages\langchain\vectorstores\azuresearch.py", line 172, in _get_search_client       
    index_client.create_index(index)
  File "C:\Users\xxxxxxx\venv\Lib\site-packages\azure\core\tracing\decorator.py", line 78, in wrapper_use_tracer
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xxxxxxx\venv\Lib\site-packages\azure\search\documents\indexes\_search_index_client.py", line 220, in create_index
    result = self._client.indexes.create(patched_index, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xxxxxxx\venv\Lib\site-packages\azure\core\tracing\decorator.py", line 78, in wrapper_use_tracer
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xxxxxx\venv\Lib\site-packages\azure\search\documents\indexes\_generated\operations\_indexes_operations.py", 
line 402, in create
    raise HttpResponseError(response=response, model=error)
azure.core.exceptions.HttpResponseError: (InvalidRequestParameter) The request is invalid. Details: definition : The vector field 'content_vector' must have the property 'vectorSearchConfiguration' set.
Code: InvalidRequestParameter
Message: The request is invalid. Details: definition : The vector field 'content_vector' must have the property 'vectorSearchConfiguration' set.
Exception Details:      (InvalidField) The vector field 'content_vector' must have the property 'vectorSearchConfiguration' set. Parameters: definition
        Code: InvalidField
        Message: The vector field 'content_vector' must have the property 'vectorSearchConfiguration' set. Parameters: definition

I have created an index manually from the Azure Cognitive Search Console, but I don't think this is the correct approach, as the script should automatically create a new index.

Answer 1

Please use pip install azure-search-documents==11.4.0b8 to ensure you are using the azure cognitive search python SDK compatible with LangChain.