Hello, , How to use gpt-4-turbo as llm for QA chatbot based on documents?

, I am facing this error after I update llama-index package.

Error code: 400 - {'error': {'message': "'name' is a required property - 'functions.0'", 'type': 'invalid_request_error', 'param': None, 'code': None}}

Hello, Hope you all are doing great.

How to improve response time? I have used GPTList Index, and output result was good, but it took 8mins for getting answer.

can you please advice me how to impove response time?

Hello, , how to delete document knowledge from index in llamaindex python?
My current code
storage_context = StorageContext.from_defaults(persist_dir="./gptdb/"+data['email'])
index = load_index_from_storage(storage_context,service_context=service_context)
index.delete_ref_doc(data["doc_id"], delete_from_docstore=True)
storage_context.persist(persist_dir="./gptdb/"+data['email'])

But, it doesn't delete document knowledge

Can I implement training of images from PDF in llama-index?

How to get response fastly from large CSV file?
I am using 3213row's CSV file, and could not get response with this file.
Using small CSV file, I was able to get response fastly.
Any solution?

Does Llama-Index support function calling now?
https://openai.com/blog/function-calling-and-other-api-updates

Hi, friends, How to improve response time from query in llama-index?

Hi, Is it possible to train excel files in llamaindex?

How to store metadata in llamaindex?
Please advice me

Claude API is available in LLama-Index?

Hello, @Logan M , How to set instruction prompt for index?
Before I was setting like this
index.index_struct.summary = companyPrompt

How to use AzureOpenAI with function calling?
Using OpenAI, it is working fine, but It is not working with AzureOpenAI
func_response = getGPT35Client(True).complete(func_prompt, tools=[relevance_fn])


def getGPT35Client(isAzure ):
client = OpenAI(model="gpt-3.5-turbo", api_key=os.environ.get('EXTOVATE_OPENAI_API_KEY'))
try:
if isAzure == True:
azureClient = AzureOpenAI(
openai_api_key=os.environ.get('MY_KEY'),
openai_api_base=os.environ.get("MY_ENDPOINT"),
openai_api_version="2023-12-01-preview",
openai_api_type="azure",
model="MyModelName"
)
return azureClient
else:
return client
except Exception as err:
print(err)
return client

storage_context = StorageContext.from_defaults()
index = GPTVectorStoreIndex(documents, service_context=service_context,storage_context=storage_context)

storage_context.persist(persist_dir="my directory'])

It was working fine before, but now it is not working.
How to fix?

Hello, I am facing this error
You tried to access openai.ChatCompletion, but this is no longer supported in openai>=1.0.0 - see the README at https://github.com/openai/openai-python for the API.

How to fix?

@Logan M , Hope you are doing well.
can I implement chat history with GPTVectorStoreIndex?

I have implemented QA Chatbot and want to add chathistory here

How to implement chathistory with QA chatbot?

How to use prompt templete with VecstoreStoreIndex?

Hello, can I only output JSON from Index?

Hello, Hope you all are doing great.

How to improve response time? I have used GPTList Index, and output result was good, but it took 8mins for getting answer.

can you please advice me how to impove response time?

For PDF documents, SimpleDirectoryReader makes Document Object Per page.
Before You created Document Object from full document.
Now I need to create Document Object based on only full document context.

How to do that?
@Logan M

Hello, I am using 1000page pdf document using llama-index. I need to provide full summary of this pdf, How to do that in llama-index?

Can I summarize full document using llama-Index?

Nowdays, I feel that it is only summarizing first page of document.

Look forward to hear your suggestions.

Regards

How to use function calling in llama-Index?

Can I create GPTVectorStoreIndex using Claude LLM in service context?
Before I have used OpenAI LLMs.

Bard API is available in Llama-Index?