如何修复“无效值”。使用 Azure OpenAI GPT 4o 小批量时出错？

Question

我尝试使用 Azure OpenAI GPT 4o 迷你批次。我首先需要创建批次。我跑了：

import os
from openai import AzureOpenAI

client = AzureOpenAI(
    api_key="[insert Azure OpenAI API key here]",
    api_version="2023-07-01-preview",
    azure_endpoint="https://[insert Azure OpenAI API enddpoint here].openai.azure.com/",
)


file = client.files.create(
    file=open("./dummy.jsonl", "rb"),
    purpose="batch"
)
file_id = file.id

与

dummy.jsonl

包含：

{
    "custom_id": "request-1",
    "method": "POST",
    "url": "/v1/chat/completions",
    "body": {
        "model": "gpt-3.5-turbo-0125",
        "messages": [
            {
                "role": "system",
                "content": "You are a helpful assistant."
            },
            {
                "role": "user",
                "content": "Hello world!"
            }
        ],
        "max_tokens": 1000
    }
}

我收到错误：

BadRequestError: Error code: 400 - {'error': {'code': 'invalidPayload', 'message': 'Invalid value for purpose.'}}

如何修复该错误？

Answer 1

要使代码正常工作，请升级

api_version

。例如，以下作品：

import os
from openai import AzureOpenAI

client = AzureOpenAI(
    api_key="[insert Azure OpenAI API key here]",
    api_version="2023-07-01-preview",
    azure_endpoint="https://[insert Azure OpenAI API enddpoint here].openai.azure.com/",
)

file = client.files.create(
    file=open("./dummy.jsonl", "rb"),
    purpose="batch"
)
file_id = file.id