Using FunctionChoiceBehavior.Required in Python gets an invalid value error

[leeprtc]

Does FunctionChoiceBehavior.Required work in Python?

I tried running the function_calling_with_required_type.py sample. Instead of using OpenAI, I'm using Azure OpenAI so I swapped OpenAIChatCompletion and OpenAIChatPromptExecutionSettings with AzureChatCompletion and AzureChatPromptExecutionSettings. However, when I ran this sample, I'm getting the following error with both the v1.2.0 and v1.3.0 packages. The model I'm using is GPT-4o version 2024-05-13.

  File "C:\...\anaconda3\envs\AI\Lib\site-packages\semantic_kernel\connectors\ai\open_ai\services\open_ai_chat_completion_base.py", line 252, in _send_chat_stream_request
    response = await self._send_request(request_settings=settings)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\...\anaconda3\envs\AI\Lib\site-packages\semantic_kernel\connectors\ai\open_ai\services\open_ai_handler.py", line 51, in _send_request
    raise ServiceResponseException(
semantic_kernel.exceptions.service_exceptions.ServiceResponseException: ("<class 'semantic_kernel.connectors.ai.open_ai.services.azure_chat_completion.AzureChatCompletion'> service failed to complete the prompt", BadRequestError('Error code: 400 - {\'error\': {\'message\': "Invalid value: \'required\'. Supported values are: \'none\' and \'auto\'.", \'type\': \'invalid_request_error\', \'param\': \'tool_choice\', \'code\': \'invalid_value\'}}'))

[moonbox3]

Hi @leeprtc, we've converted this to an issue (#7463). I will respond there.