Description
I like the elegance of your Notebooks, using Jinja Templates and Langchain from the very beginning.
However, I noticed that the get_get_completion() function in all_prompt_engineering_techniques/prompt-templates-variables-jinja2.ipynb still uses the OpenAI v0 legacy API. The requirements.txt file installs/demands openai==1.51.1.
So when I run the the class PromptTemplate: ... code cell in that Jupyter Notebook I get an error.
APIRemovedInV1:
You tried to access openai.ChatCompletion, 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`
I replaced the current code cell defining get_get_completion() function (not reproduced here) with the following code. So I know how to get along. However I don't know how many Notebooks in this repo will be affected, And I don't know if you prefer to fix everything yourself with the `openai migrate' mechanism. That's wh< I'm not sending a pull request.
### my replacement of the `get_get_completion()` function - new openAI v1 interface
import os
from openai import OpenAI
from jinja2 import Template
from dotenv import load_dotenv
load_dotenv()
def get_completion(prompt, model="gpt-4o-mini"):
''' Get a completion from the OpenAI API
Args:
prompt (str): The prompt to send to the API
model (str): The model to use for the completion
Returns:
str: The completion text
'''
messages = [{"role": "user", "content": prompt}]
client = OpenAI(api_key=os.getenv('OPENAI_API_KEY'))
response = client.chat.completions.create(
model=model,
messages=messages,
temperature=0,
)
return response.choices[0].message.content