add FunctionMessage support to _convert_dict_to_message() in OpenAI chat model

[thehunmonkgroup]

Already supported in the reverse operation in _convert_message_to_dict(), this just provides parity.

@hwchase17
@agola11

[vercel]

@thehunmonkgroup is attempting to deploy a commit to the LangChain Team on Vercel.

A member of the Team first needs to authorize it.

[hwchase17]

wont we want to include additional_kwargs?

Can we add unit test?

[thehunmonkgroup]

wont we want to include additional_kwargs?

@hwchase17 none of the other elements of that function add additional_kwargs besides the function_call in the AIMessage conversion. It's not clear to me what we would even add to additional_kwargs in the FunctionMessage case.

Here's the example code from OpenAI, looks like passing a function message is just the role, name, and content to me:

{
  "model": "gpt-3.5-turbo-0613",
  "messages": [
    {"role": "user", "content": "What is the weather like in Boston?"},
    {"role": "assistant", "content": null, "function_call": {"name": "get_current_weather", "arguments": "{ \"location\": \"Boston, MA\"}"}},
    {"role": "function", "name": "get_current_weather", "content": "{\"temperature\": "22", \"unit\": \"celsius\", \"description\": \"Sunny\"}"}
  ],
  "functions": [
    {
      "name": "get_current_weather",
      "description": "Get the current weather in a given location",
      "parameters": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string",
            "description": "The city and state, e.g. San Francisco, CA"
          },
          "unit": {
            "type": "string",
            "enum": ["celsius", "fahrenheit"]
          }
        },
        "required": ["location"]
      }
    }
  ]
}

If you could confirm that the existing code is correct, I'll try to work up a unit test, first time contributor, so I'll need to fumble my way through it ;)

[thehunmonkgroup]

Can we add unit test?

@hwchase17 well there wasn't even a unit test file for chat_models/test_openai.py ;)

I think I got a correct test written and it's passing.

Let me know if you need anything else to get this PR in.

[hwchase17]

wont we want to include additional_kwargs?

@hwchase17 none of the other elements of that function add additional_kwargs besides the function_call in the AIMessage conversion. It's not clear to me what we would even add to additional_kwargs in the FunctionMessage case.

Here's the example code from OpenAI, looks like passing a function message is just the role, name, and content to me:

{
  "model": "gpt-3.5-turbo-0613",
  "messages": [
    {"role": "user", "content": "What is the weather like in Boston?"},
    {"role": "assistant", "content": null, "function_call": {"name": "get_current_weather", "arguments": "{ \"location\": \"Boston, MA\"}"}},
    {"role": "function", "name": "get_current_weather", "content": "{\"temperature\": "22", \"unit\": \"celsius\", \"description\": \"Sunny\"}"}
  ],
  "functions": [
    {
      "name": "get_current_weather",
      "description": "Get the current weather in a given location",
      "parameters": {
        "type": "object",
        "properties": {
          "location": {
            "type": "string",
            "description": "The city and state, e.g. San Francisco, CA"
          },
          "unit": {
            "type": "string",
            "enum": ["celsius", "fahrenheit"]
          }
        },
        "required": ["location"]
      }
    }
  ]
}

If you could confirm that the existing code is correct, I'll try to work up a unit test, first time contributor, so I'll need to fumble my way through it ;)

ah right sorry, got confused

thanks!