Using OpenAI Embeddings with `get_embedding_function.py` about rag-tutorial-v2 HOT 3 CLOSED

I used something like the below - you have to update the 'embedding_function' variable in query_data.py with the model argument to get it to work though. I've pasted both below

def get_embedding_function():
embeddings = OpenAIEmbeddings(model="text-embedding-3-large")

return embeddings


in query_data.py
embedding_function = OpenAIEmbeddings(model="text-embedding-3-large")

from rag-tutorial-v2.

Yup, as @jaimeelincoln says, you can use the OpenAI embedding helper function directly rather than having to do it via the client:

https://python.langchain.com/v0.1/docs/integrations/text_embedding/openai/

OpenAIEmbeddings(model="text-embedding-3-large")

from rag-tutorial-v2.

The embedding need input but i cannot see it being passed to it in the script in populate_database.py

Also, your code is trying to create the actual embedding itself, which is why you need the input (text and model). But get_embedding_function is supposed to return the function/class for embedding (which is what something like the OpenAIEmbeddings class does).

from rag-tutorial-v2.