Vector data & Cognitive search + Azure Open AI

Question

Vector data & Cognitive search + Azure Open AI

Arun Srinivasan 80

I have created the docstore, vectostore and indexstore files using llama_index

Or is there a way to use Azure services to acheive this?

from langchain.chat_models import ChatOpenAI
.............
.............

    llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0, model_name="gpt-3.5-turbo", max_tokens=num_outputs))
    prompt_helper = PromptHelper(max_input_size, num_output=num_outputs, max_chunk_overlap=max_chunk_overlap, chunk_size_limit=chunk_size_limit)
 
    docs = SimpleDirectoryReader(directory_path).load_data()
    service_context = ServiceContext.from_defaults(llm_predictor=llm_predictor, prompt_helper=prompt_helper)
    vectorIndex = VectorStoreIndex.from_documents(documents=docs, service_context=service_context)
    vectorIndex.storage_context.persist(persist_dir='<<local storage>>')

Ryan Hill 30,281 Reputation points Microsoft Employee Moderator

2023-07-13T17:19:42.1366667+00:00

@Arun Srinivasan can you clarify the requirements, the goal you're trying to achieve and any error messages you're trying to overcome?
Israel Ekpo 0 Reputation points Microsoft Employee

2023-07-13T21:19:19.4066667+00:00

I think the better question is what the author is looking to accomplish and what challenges are you experiencing? From the post, it is not clear what the question is and what the author of the question is looking to accomplish from the code snippet. I dont want to make assumptions so when you have a moment, you can clarify your objectives, challenges as well as what assistance you are looking for. Thanks.
Arun Srinivasan 80 Reputation points

2023-07-14T04:38:36.41+00:00

I'm trying to use Open AI embeddings apiI'm trying to use Open AI embeddings apiI'm trying to use Open AI embeddings apiApologies for not being clear the first time.
Ryan Hill 30,281 Reputation points Microsoft Employee Moderator

2023-07-17T21:43:33.0466667+00:00

Sorry @Arun Srinivasan but that isn't really helpful for the community to best assist. Can you elaborate on what objective you're trying to achieve with Open AI with the documents, vector stores, and indices?

Your answer

Ryan Hill 30,281 Reputation points Microsoft Employee Moderator

2023-07-13T17:19:42.1366667+00:00

@Arun Srinivasan can you clarify the requirements, the goal you're trying to achieve and any error messages you're trying to overcome?
Israel Ekpo 0 Reputation points Microsoft Employee

2023-07-13T21:19:19.4066667+00:00

I think the better question is what the author is looking to accomplish and what challenges are you experiencing? From the post, it is not clear what the question is and what the author of the question is looking to accomplish from the code snippet. I dont want to make assumptions so when you have a moment, you can clarify your objectives, challenges as well as what assistance you are looking for. Thanks.
Arun Srinivasan 80 Reputation points

2023-07-14T04:38:36.41+00:00

I'm trying to use Open AI embeddings apiI'm trying to use Open AI embeddings apiI'm trying to use Open AI embeddings apiApologies for not being clear the first time.
Ryan Hill 30,281 Reputation points Microsoft Employee Moderator

2023-07-17T21:43:33.0466667+00:00

Sorry @Arun Srinivasan but that isn't really helpful for the community to best assist. Can you elaborate on what objective you're trying to achieve with Open AI with the documents, vector stores, and indices?

Share via

Vector data & Cognitive search + Azure Open AI

Your answer