Find answers from the community

Home
Members
RAPHCVR
R
RAPHCVR
Offline, last seen 3 months ago
Joined September 25, 2024
Hi , when using llama index and ask for translating a document, or some other tasks to last gpt-4 using your api, I often encounter api timeouts (because too much retries), how to fix that ?
4 comments
R
L
R
RAPHCVR
·

Thread

58 comments
L
R
Hi , with gpt-3.5-turbo-16k-0613, I often encounter an issue:
11 comments
W
L
R
I have this error:BadRequestError: Error code: 400 - {'error': {'message': "This model's maximum context length is 16385 tokens. However, you requested 16390 tokens (14342 in the messages, 2048 in the completion). Please reduce the length of the messages or completion.", 'type': 'invalid_request_error', 'param': 'messages', 'code': 'context_length_exceeded'}}
30 comments
R
L
D
Hi , I'm back from my test, and the prompt system change for making the answer's language french is not sufficient
3 comments
R
L
Hi , when trying to use the new gpt4 turbo, I have this error: ValueError: Unknown model 'gpt-4-1106-preview'.
9 comments
R
W
Hi, when asking for summaries with AzureOpenAI llm, I got the error openai.error.InvalidRequestError: The response was filtered due to the prompt triggering Azure OpenAI's content management policy. Please modify your prompt and retry. That's happen only when asking question in french. How to fix this issue ? When I want to add an llm predictor in service_context so specify to answer in french, it says me that llm predictor can't be set when llm is
69 comments
L
s
R
sorry for disturbing again, but with azure openai, I encounter an issue like this one: https://discord.com/channels/1059199217496772688/1172580616999534593 but for the CondenseQuestion chat engine, and downgrade is not sufficient
33 comments
R
L
For the moment, when creating an index with specified fields, and putting documents in it parsed and indexed with the embedding, the llm and llama_index, it works, but the fields different than "content" are not filled. How to make these field filled automatically according to the document given ?
74 comments
R
L
Hi , when using SimpleDirectoryReader with pdf files, it shows me to install pypdf, but it is already installed, what should I do ?
5 comments
L
R
Hi, do you know how to put a routerquery engine in a chat engine ? Or to achieve the same result ?
10 comments
L
R
Hi back @Logan M , I've just tried to implement the CondensePlusContextChatEngine, but, when using summary index retriever with a large amount of documents, it shows me a token limit error
10 comments
R
L
In my case I want to use async for the summary index and the Vector store index, used in the routerquery engine, used in condense question chat engine
17 comments
R
L
R
RAPHCVR
·

Thread

77 comments
R
L
R
R
RAPHCVR
·

Azure

Hi, I'm currently always using AzureOpenAI and cognitive search as my db for retrieving context, I wanted to know why I have this error in my code:
Plain Text
documents = SimpleDirectoryReader("files").load_data()

llm = AzureOpenAI(
    model="gpt-35-turbo-16k",
    engine="gpt-35-turbo-16k",
    api_key=api_key,
    api_base=api_base,
    api_type=api_type,
    api_version=api_version,
)

# You need to deploy your own embedding model as well as your own chat completion model
embed_model = OpenAIEmbedding(
    model="text-embedding-ada-002",
    deployment_name="text-embedding-ada-002",
    api_key=api_key,
    api_base=api_base,
    api_type=api_type,
    api_version=api_version,
)


prompt_helper = PromptHelper(context_window=16384, num_output=2048)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
service_context = ServiceContext.from_defaults(
        embed_model=embed_model,
        prompt_helper=prompt_helper,
        llm=llm,
    )

summary_text=(
    "Des informations contextuelles provenant de plusieurs sources sont présentées ci-dessous.\n"
    "---------------------\n"
    "{context_str}\n"
    "---------------------\n"
    "Étant donné les informations provenant de sources multiples et sans connaissances préalables,"
    "répondre à la requête.\n"
    "Requête : {query_str}\n"
    "Réponse : ")

index1 = SummaryIndex(
    documents,
    service_context=service_context,
    storage_context=storage_context,
    summary_text=summary_text,
    response_mode="tree_summarize"
        )
        


index2 = VectorStoreIndex(
    documents,
            service_context=service_context,
            storage_context=storage_context,
        )

list_query_engine = index1.as_query_engine(response_mode="tree_summarize")
vector_query_engine = index2.as_query_engine(similarity_top_k=5)

list_tool = QueryEngineTool.from_defaults(
    query_engine=list_query_engine,
    description="Utile pour les questions de synthèse liées à la source de données",
)

vector_tool = QueryEngineTool.from_defaults(
    query_engine=vector_query_engine,
    description="Utile pour retrouver un contexte spécifique lié à la source de données",
)



# initialize router query engine (single selection, pydantic)
query_engine = RouterQueryEngine(
    selector=LLMMultiSelector.from_defaults(),
    query_engine_tools=[
        list_tool,
        vector_tool,
    ],
    service_context=service_context
)
#query_engine=index1.as_query_engine()
resp=query_engine.query("Fait un résumé complet des transcriptions d'appels, en français, à partir du contexte, avec une ou plusieurs conversations entre un assuré et un ou plusieurs opérateurs d'assistance d'assurance")
print(resp)
19 comments
R
L
W
Hi ! I'm currently trying to use the QASummaryQueryEngineBuilder with chat engine function for my LLama2 based model, but I have this error happening:
1 comment
L
Hi @Logan M , I'm currently using a specific function from llamaindex, but there is a problem with the input
80 comments
R
L
W
I've already tried llamaindex, but i does not work well : It talks about multiple sources but there is only one for example
22 comments
L
R