1

Hi! I am trying to build a workflow that iterates over contents from an Excel file, which uses assistant api retrieval. It is always processing the first row and I am not sure what is wrong. Thanks to anyone who helps.

import os
import pandas as pd
import openai
import time
from openai import OpenAI
from dotenv import load_dotenv

# Load the OpenAI API key
load_dotenv('your_env_file.env')
api_key = os.getenv('YOUR_ENV_VARIABLE_FOR_API_KEY')
openai.api_key = api_key  # Make sure to set the API key for the openai package

# Read contents from an Excel file
excel_path = 'path_to_your_excel_file.xlsx'
df = pd.read_excel(excel_path)
content_column = 'your_content_column_name'  # Replace with the name of your column

# Prepare a column in the DataFrame to store responses
df['Assistant_Responses'] = ''

# Upload a file with an "assistants" purpose
client = OpenAI()
file = client.files.create(
  file=open("path_to_your_file.json", "rb"),
  purpose='assistants'
)
print(file.id)

# Iterate over each content in the DataFrame
for index, row in df.iterrows():
    user_input = row[content_column]
    print(f"Processing row {index + 1}/{len(df)}...")

    # Create a thread
    thread = client.beta.threads.create()
    message = client.beta.threads.messages.create(
      thread_id=thread.id,
      role="user",
      content=user_input,
      file_ids=[file.id]
    )

    run = openai.beta.threads.runs.create(
      thread_id=thread.id,
      assistant_id="your_assistant_id"
    )

    while run.status != 'requires_action':
        time.sleep(1)
        run = openai.beta.threads.runs.retrieve(
           thread_id=thread.id,
           run_id=run.id
        )

    try:
        messages = openai.beta.threads.messages.list(
            thread_id=thread.id
        )
        # Assuming the last message is the response
        if messages.data:
            response = messages.data[-1].content
            df.at[index, 'Assistant_Responses'] = response
    except Exception as e:
        print(f"Error processing row {index + 1}: {e}")

# Save the updated DataFrame back to the Excel file
df.to_excel(excel_path, index=False)

print("Processing complete.")