Error in model execution: RetryError[<Future at 0x7efc1406b040 state=finished raised APIRemovedInV1>]

[mingjun1120]

Hi everyone, I am encountering an error while using Azure OpenAI. The code was running fine when I executed it on Google Colab, but it is raising an error when I run it on GitHub Codespace Jupyter Notebook. The code in both platforms is exactly the same. Below is my code in Jupyter Notebook:

Access to ALL Credentials

import os
from dotenv import load_dotenv, find_dotenv

load_dotenv(find_dotenv())

api_key = os.environ.get("OPENAI_API_KEY")
api_base = os.environ.get("OPENAI_API_BASE")
api_version = os.environ.get("OPENAI_API_VERSION")
api_type = os.environ.get("OPENAI_API_BASE")

print(f'API_KEY: {api_key}')
print(f'API BASE: {api_base}')
print(f'API VERSION: {api_version}')
print(f'API TYPE: {api_type}')

Define any LLM model (such as GPT-3) ✅

from promptify import Prompter, OpenAI, Pipeline, Azure

# Define the API key for the OpenAI model
# Create an instance of the OpenAI model
model = Azure(api_key=api_key, api_base=api_base, api_version=api_version, api_type=api_type, engine='gpt-35-turbo')
prompter = Prompter('multilabel_classification.jinja')
pipe = Pipeline(prompter , model)

# Example sentence for demonstration
sent = "The patient is a 93-year-old female with a medical history of chronic right hip pain, \
    osteoporosis, hypertension, depression, and chronic atrial fibrillation admitted for evaluation \
        and management of severe nausea and vomiting and urinary tract infection"
print(sent)

1: MultiLabel Text Classification Example in 2 Lines of code, with no training data required 🚀

result = pipe.fit(n_output_labels = 5,
                  domain          = 'clinical', # it could be any domain such as -> financial, education, biomedical etc
                  text_input      = sent,
                  labels          = None)

# Output
result

0%| | 0/1 [00:13<?, ?it/s]
Error in model execution: RetryError[<Future at 0x7efc140416f0 state=finished raised APIRemovedInV1>]

[st-karlos]

I got an error in the corresponding notebook for both reasons:

• nlp_prompter is used in the original code, while pipe seems the appropriate choice, as @mingjun1120 has used here.
• TypeError: Pipeline.fit() got multiple values for argument 'text_input': I cannot run even for the provided example.

Any ideas?