TypeError: Got unknown type F

pradeepdev-1995 commented 9 months ago

System Info

langchain-0.0.291 python3.9

Who can help?

No response

Information

[X] The official example notebooks/scripts
[ ] My own modified scripts

Related Components

[X] LLMs/Chat Models
[ ] Embedding Models
[X] Prompts / Prompt Templates / Prompt Selectors
[X] Output Parsers
[ ] Document Loaders
[ ] Vector Stores / Retrievers
[ ] Memory
[ ] Agents / Agent Executors
[ ] Tools / Toolkits
[ ] Chains
[ ] Callbacks/Tracing
[ ] Async

Reproduction

import os
import openai
openai.api_type = "azure"
openai.api_base = os.getenv("OPENAI_API_BASE")
openai.api_version = "version"
openai.api_key = os.getenv("OPENAI_API_KEY")
DEPLOYMENT_NAME = 'deployment name
from langchain.chat_models import AzureChatOpenAI
llm = AzureChatOpenAI(
    openai_api_base=os.getenv("OPENAI_API_BASE"),
    openai_api_version="version",
    deployment_name=DEPLOYMENT_NAME,
    openai_api_key=os.getenv("OPENAI_API_KEY"),
    openai_api_type="azure",
    temperature=0.0
)

result = llm("Father of computer")
print(result)

Expected behavior

Expecting the answer

dosubot[bot] commented 9 months ago

🤖

Hello,

Thank you for reaching out and providing detailed information about your issue. The error message "TypeError: Got unknown type F" suggests that the function _convert_message_to_dict is receiving an argument of a type it doesn't recognize. This function is designed to handle instances of ChatMessage, HumanMessage, AIMessage, SystemMessage, and FunctionMessage. If it receives an instance of any other type, it raises a ValueError.

Based on the code snippet you provided, it seems like you're trying to use the AzureChatOpenAI class to interact with a deployed model on Azure OpenAI. However, the error might be occurring because the message you're passing to the AzureChatOpenAI instance is not an instance of the expected types.

Here is the relevant code snippet from the AzureChatOpenAI class:

def _convert_message_to_dict(message: BaseMessage) -> dict:
    if isinstance(message, ChatMessage):
        message_dict = {"role": message.role, "content": message.content}
    elif isinstance(message, HumanMessage):
        message_dict = {"role": "user", "content": message.content}
    elif isinstance(message, AIMessage):
        message_dict = {"role": "assistant", "content": message.content}
    elif isinstance(message, SystemMessage):
        message_dict = {"role": "system", "content": message.content}
    elif isinstance(message, FunctionMessage):
        message_dict = {
            "role": "function",
            "name": message.name,
            "content": message.content,
        }
    else:
        raise ValueError(f"Got unknown type {message}")
    if "name" in message.additional_kwargs:
        message_dict["name"] = message.additional_kwargs["name"]
    return message_dict

In your code, you're passing a string directly to the AzureChatOpenAI instance:

result = llm("Father of computer")

However, the AzureChatOpenAI class expects a ChatMessage, HumanMessage, AIMessage, SystemMessage, or FunctionMessage instance, not a string. You should create an instance of one of these classes and pass that to the AzureChatOpenAI instance instead. Here's an example using HumanMessage:

from langchain.chat_models import HumanMessage

message = HumanMessage(content="Father of computer")
result = llm(message)

Please try this modification and let me know if it resolves your issue. If not, please provide more details about the error, such as the full traceback, so I can assist you further.

Best regards.

Sources

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

joshuasundance-swca commented 9 months ago

@dosu-bot was close. This documentation calls the model using a list of HumanMessages. You should call the llm differently:

from langchain.schema import HumanMessage

message = HumanMessage(content="Father of computer")
result = llm([message])

result

AIMessage(content="The father of computer is considered to be Charles Babbage. He was an English mathematician, philosopher, and inventor who is best known for conceptualizing and designing an early mechanical general-purpose computer called the Analytical Engine in the 1830s. Although the Analytical Engine was never fully constructed during his lifetime, his work laid the foundation for modern computers. Babbage's contributions to the field of computing make him widely regarded as the father of computer science and the father of the computer.", additional_kwargs={}, example=False)

or use llm.predict

result2 = llm.predict("Father of computer")

result2

"The father of computer is considered to be Charles Babbage. He is known for his work on the Analytical Engine, an early mechanical computer concept. Although the Analytical Engine was never built during Babbage's lifetime, his designs and ideas laid the foundation for modern computing."

pradeepdev-1995 commented 9 months ago

@joshuasundance-swca Working the above solution. Thank you.

langchain-ai / langchain