Find answers from the community

Updated 10 months ago

from llama_index.finetuning import (

from llama_index.finetuning import (
generate_qa_embedding_pairs,
EmbeddingQAFinetuneDataset,
)

train_dataset = generate_qa_embedding_pairs(nodes, llm=llm_model)

I was having issues with using azurechatopenai model with generate_qa_embedding_pairs function. Anyway that I can get around this?

Attachment

8 comments

WWhiteFang_Jr

How have you imported AzureChatOpenAI? i think this si not coming from llama-index

GGoat

I'm importing from

Plain Text

from langchain.chat_models import AzureChatOpenAI

WWhiteFang_Jr

Ah that's why! Import AzureOpenAI from llama-index, llm internally perfomr API call for which it does llm.complete which is not present in langchain implementation

WWhiteFang_Jr

First do this: pip install llama-index-llms-azure-openai

and then import it like this:
from llama_index.llms.azure import AzureOpenAI

GGoat

I'm getting this error

Attachment

WWhiteFang_Jr

Are you on 0.9 version ? maybe version mismatch

GGoat

LLogan M

Yea, this was solved in newer versions

Add a reply