AttributeError: module 'openai' has no attribute 'Embedding'

API
, ,
1

Hello guys. I have this issue when I try to use the API. I Used it exactly 2 days ago and it was working fine. I have been running the same code with no errors at all. I am using Google Colab with LangChain and FAISS; and got my packages up-to-date. Now after the recent updates on GPT-4 Turbo I get these errors:

AttributeError Traceback (most recent call last)
in <cell line: 2>()
1 from langchain.llms import OpenAI
----> 2 llm = OpenAI(model_name=“text-davinci-003”)

3 frames
/usr/local/lib/python3.10/dist-packages/langchain/llms/openai.py in validate_environment(cls, values)
264 import openai
265
→ 266 values[“client”] = openai.Completion
267 except ImportError:
268 raise ImportError(

AttributeError: module ‘openai’ has no attribute ‘Completion’

and

AttributeError: module 'openaAttributeError Traceback (most recent call last)
in <cell line: 6>()
4 docs = text_splitter.split_documents(documents)
5
----> 6 faissIndex = FAISS.from_documents(docs, OpenAIEmbeddings())
7 faissIndex.save_local(“graduate_studies_docs”)#saves a folder to hold index.faiss and index.pkl

2 frames
/usr/local/lib/python3.10/dist-packages/langchain/embeddings/openai.py in validate_environment(cls, values)
282 import openai
283
→ 284 values[“client”] = openai.Embedding
285 except ImportError:
286 raise ImportError(

AttributeError: module ‘openai’ has no attribute ‘Embedding’

Is this because of the new updates or what? What to do please?

2

they updated “completion” to “completions” recently, and looks like “embedding” might have become “embeddings”?

3

You are gonna have to wait until langchain updates with the new changes or roll back to openai==0.28.1

4

I am sort of getting the same error

from langchain.llms import OpenAI
llm = OpenAI(model_name="text-davinci-003")

llm("what is pi?")

and the error i am getting is

AttributeError                            Traceback (most recent call last)
Cell In[29], line 4
      1 from langchain.llms import OpenAI
      2 llm = OpenAI(model_name="text-davinci-003")
----> 4 llm("what is pi?")

File ~/anaconda3/lib/python3.11/site-packages/langchain/llms/base.py:246, in BaseLLM.__call__(self, prompt, stop)
    244 def __call__(self, prompt: str, stop: Optional[List[str]] = None) -> str:
    245     """Check Cache and run the LLM on the given prompt and input."""
--> 246     return self.generate([prompt], stop=stop).generations[0][0].text

File ~/anaconda3/lib/python3.11/site-packages/langchain/llms/base.py:140, in BaseLLM.generate(self, prompts, stop)
    138 except (KeyboardInterrupt, Exception) as e:
    139     self.callback_manager.on_llm_error(e, verbose=self.verbose)
--> 140     raise e
    141 self.callback_manager.on_llm_end(output, verbose=self.verbose)
    142 return output

File ~/anaconda3/lib/python3.11/site-packages/langchain/llms/base.py:137, in BaseLLM.generate(self, prompts, stop)
    133 self.callback_manager.on_llm_start(
    134     {"name": self.__class__.__name__}, prompts, verbose=self.verbose
    135 )
    136 try:
--> 137     output = self._generate(prompts, stop=stop)
    138 except (KeyboardInterrupt, Exception) as e:
    139     self.callback_manager.on_llm_error(e, verbose=self.verbose)

File ~/anaconda3/lib/python3.11/site-packages/langchain/llms/openai.py:292, in BaseOpenAI._generate(self, prompts, stop)
    290     choices.extend(response["choices"])
    291 else:
--> 292     response = completion_with_retry(self, prompt=_prompts, **params)
    293     choices.extend(response["choices"])
    294 if not self.streaming:
    295     # Can't update token usage if streaming

File ~/anaconda3/lib/python3.11/site-packages/langchain/llms/openai.py:93, in completion_with_retry(llm, **kwargs)
     91 def completion_with_retry(llm: Union[BaseOpenAI, OpenAIChat], **kwargs: Any) -> Any:
     92     """Use tenacity to retry the completion call."""
---> 93     retry_decorator = _create_retry_decorator(llm)
     95     @retry_decorator
     96     def _completion_with_retry(**kwargs: Any) -> Any:
     97         return llm.client.create(**kwargs)

File ~/anaconda3/lib/python3.11/site-packages/langchain/llms/openai.py:81, in _create_retry_decorator(llm)
     73 max_seconds = 10
     74 # Wait 2^x * 1 second between each retry starting with
     75 # 4 seconds, then up to 10 seconds, then 10 seconds afterwards
     76 return retry(
     77     reraise=True,
     78     stop=stop_after_attempt(llm.max_retries),
     79     wait=wait_exponential(multiplier=1, min=min_seconds, max=max_seconds),
     80     retry=(
---> 81         retry_if_exception_type(openai.error.Timeout)
     82         | retry_if_exception_type(openai.error.APIError)
     83         | retry_if_exception_type(openai.error.APIConnectionError)
     84         | retry_if_exception_type(openai.error.RateLimitError)
     85         | retry_if_exception_type(openai.error.ServiceUnavailableError)
     86     ),
     87     before_sleep=before_sleep_log(logger, logging.WARNING),
     88 )

AttributeError: module 'openai' has no attribute 'error'

i tried to revert back to 0.28.1 using

!pip install openai==0.28.1

(using the ! running in the jupyter notebook) but to no avail.

Any ideas?

5

upgrade your langchain to latest version, it will work again:
pip install --upgrade langchain