Azure Search Service Metadata-Based Filtering for Vector Search in Langchain not working

[dilippanampilly]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain.vectorstores.azuresearch import AzureSearch
from langchain.chains.combine_documents import create_stuff_documents_chain

    vector_store = AzureSearch(
        azure_search_endpoint=azure_search_service_endpoint,
        azure_search_key=azure_search_service_key_credential,
        index_name=azure_search_service_index_name,
        embedding_function=embeddings.embed_query,
        semantic_configuration_name="default"
    )

    base_retriever = vector_store.as_retriever(search_kwargs={'filter': {'doc_Title':'Corporations Act 2001'}})

    PROMPT = PromptTemplate(
        template=PROMPT_TEMPLATE, input_variables=["context", "input"]
    )
    combine_docs_chain = create_stuff_documents_chain(llm, PROMPT)
    rag_chain = create_retrieval_chain(base_retriever, combine_docs_chain)
    response = rag_chain.invoke({"input": final_prompt})

Description

I have a vector store in Azure Search Service. the vector store is also having metadata like below. Right now i am facing issue in filtering the metadata.
base_retriever = vector_store.as_retriever(search_kwargs={'filter': {'doc_Title':'Corporations Act 2001'}})
filter is not working in above line. @dosu-bot.
"value": [
{
"@search.score": 1,
"id": "qrqra24424234",
"content": "doc_Title: Corporations Act 2001\nmandateClause_Title: Presumption in certain cases that property is offence",
"metadata": "{"doc_Title": "Corporations Act 2001"}",
"content_vector": [
-0.010611576,
-0.009440079,
....
]
}
]

System Info

azure-search-documents==11.6.0b3
azure-identity
langchain==0.1.16
langchain-openai==0.1.3

Python 3.11.9