AttributeError: module 'giskard.llm' has no attribute 'embeddings'

[AmineDjeghri]

Issue Type

Bug

Source

source

Giskard Library Version

2.15.5

OS Platform and Distribution

Macos 14

Python version

3.11

Installed python packages

No response

Current Behaviour?

giskard.llm.embeddings.openai.set_embedding_model('text-embedding-ada-002')
    ^^^^^^^^^^^^^^^^^^^^^^
AttributeError: module 'giskard.llm' has no attribute 'embeddings'

Standalone code OR list down the steps to reproduce the issue

import openai
import pandas as pd
import tiktoken
from scipy import spatial

from giskard import scan, Dataset, Model

giskard.llm.set_llm_model('gpt-4o-mini')
giskard.llm.embeddings.openai.set_embedding_model('text-embedding-ada-002')

Relevant log output

No response

[AmineDjeghri]

okey, apparently there is a PR that was merged 45min ago

#2074

edit : still doesn't work

[henchaves]

Hey @AmineDjeghri !
Thanks for reporting this.

Could you try running this code:

from giskard.llm.embeddings.openai import set_embedding_model
set_embedding_model('text-embedding-ada-002')

for an unknown reason, it seems that giskard.llm.embeddings.openai.set_embedding_model is not working directly. I'll investigate further to see the cause of this. In the meanwhile, the previous code snippet should work.

[AmineDjeghri]

Hey
thanks it works

Do you know if the doc is deprecated ?
https://docs.giskard.ai/en/latest/reference/notebooks/LLM_QA_Winter_Olympics.html

i found this code and it seems that it doesn't work :

    query_embedding_response = openai.Embedding.create(model=EMBEDDING_MODEL, input=query)
    query_embedding = query_embedding_response["data"][0]["embedding"]

Traceback (most recent call last):
  File "/Users/aminedjeghri/PycharmProjects/ai-cloud-project-template/temp.py", line 102, in <module>
    ask('Which athletes won the gold medal in curling at the 2022 Winter Olympics?')
  File "/Users/aminedjeghri/PycharmProjects/ai-cloud-project-template/temp.py", line 83, in ask
    message = query_message(query, db, model=model, token_budget=token_budget)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/aminedjeghri/PycharmProjects/ai-cloud-project-template/temp.py", line 63, in query_message
    strings, _ = strings_ranked_by_relation(query, db)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/aminedjeghri/PycharmProjects/ai-cloud-project-template/temp.py", line 40, in strings_ranked_by_relation
    query_embedding_response = openai.Embedding.create(model=EMBEDDING_MODEL, input=query)
                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/aminedjeghri/PycharmProjects/ai-cloud-project-template/.venv/lib/python3.11/site-packages/openai/lib/_old_api.py", line 39, in __call__
    raise APIRemovedInV1(symbol=self._symbol)
openai.lib._old_api.APIRemovedInV1: 

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

You can run `openai migrate` to automatically upgrade your codebase to use the 1.0.0 interface. 

Alternatively, you can pin your installation to the old version, e.g. `pip install openai==0.28`