Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Sample code in RunnableWithMessageHistory not working through langserve #630

Open
harris opened this issue May 4, 2024 · 3 comments
Open

Comments

@harris
Copy link

harris commented May 4, 2024

I am using the code from the sample section of the RunnableWithMessageHistory. it works perfectly if I invoke the chain manually by calling in python code directly. It DOES NOT work when calling through the langserve API. hopefully I am not mistaken but i believe the dictionary parsing in langserve probably not able to recognize the dict properly. Attach is the error I am encountering also.

This works:

print(chain_with_history.invoke(  # noqa: T201
    {"ability": "math", "question": "What's its inverse"},
    config={"configurable": {"session_id": "foo"}}
))

with the samples from the RunnableWithMessageHistory file

from typing import List
from langchain_openai.chat_models import ChatOpenAI

from langchain_core.chat_history import BaseChatMessageHistory
from langchain_core.messages import BaseMessage, AIMessage
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_core.pydantic_v1 import BaseModel, Field
from langchain_core.runnables.history import RunnableWithMessageHistory

def example_history(settings):
    class InMemoryHistory(BaseChatMessageHistory, BaseModel):
        messages: List[BaseMessage] = Field(default_factory=list)

        def add_messages(self, messages: List[BaseMessage]) -> None:
            self.messages.extend(messages)

        def clear(self) -> None:
            self.messages = []

    # Here we use a global variable to store the chat message history.
    # This will make it easier to inspect it to see the underlying results.
    store = {}

    def get_by_session_id(session_id: str) -> BaseChatMessageHistory:
        if session_id not in store:
            store[session_id] = InMemoryHistory()
        return store[session_id]

    history = get_by_session_id("1")
    history.add_message(AIMessage(content="hello"))
    # print(store)  # noqa: T201

    prompt = ChatPromptTemplate.from_messages([
        ("system", "You're an assistant who's good at {ability}"),
        MessagesPlaceholder(variable_name="history"),
        ("human", "{question}"),
    ])
    model = ChatOpenAI(temperature=0, openai_api_key=settings.OPENAI_API_KEY)

    chain = prompt | model

    chain_with_history = RunnableWithMessageHistory(
        chain,
        # Uses the get_by_session_id function defined in the example
        # above.
        get_by_session_id,
        input_messages_key="question",
        history_messages_key="history",
    )
    return chain_with_history

However, it will fail if call through the api through langserve.

await fetch(`localhost:8000/example_history/invoke`, {
			method: 'POST',
			headers: { accept: 'application/json', 'Content-Type': 'application/json' },
			body: JSON.stringify({
				input: { ability: 'math', question: "What's its inverse?" },
				config: {
					configurable: {
						session_id: "something_random",
					}
				}
			}),
		});

This produces the following errors:

ne 65, in __call__
    await wrap_app_handling_exceptions(self.app, conn)(scope, receive, send)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 64, in wrapped_app
    raise exc
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/routing.py", line 756, in __call__
    await self.middleware_stack(scope, receive, send)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/routing.py", line 776, in app
    await route.handle(scope, receive, send)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/routing.py", line 297, in handle
    await self.app(scope, receive, send)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/routing.py", line 77, in app
    await wrap_app_handling_exceptions(app, request)(scope, receive, send)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 64, in wrapped_app
    raise exc
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/starlette/routing.py", line 72, in app
    response = await func(request)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/fastapi/routing.py", line 278, in app
    raw_response = await run_endpoint_function(
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/fastapi/routing.py", line 191, in run_endpoint_function
    return await dependant.call(**values)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langserve/server.py", line 530, in invoke
    return await api_handler.invoke(request)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langserve/api_handler.py", line 833, in invoke
    output = await invoke_coro
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 4529, in ainvoke
    return await self.bound.ainvoke(
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 4529, in ainvoke
    return await self.bound.ainvoke(
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 2536, in ainvoke
    input = await step.ainvoke(
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 4529, in ainvoke
    return await self.bound.ainvoke(
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 2536, in ainvoke
    input = await step.ainvoke(
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/prompts/base.py", line 143, in ainvoke
    return await self._acall_with_config(
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/runnables/base.py", line 1677, in _acall_with_config
    output = await coro
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/prompts/base.py", line 117, in _aformat_prompt_with_error_handling
    _inner_input = self._validate_input(inner_input)
  File "/Users/harris/code/meow/.venv/lib/python3.10/site-packages/langchain_core/prompts/base.py", line 103, in _validate_input
    raise KeyError(
KeyError: "Input to ChatPromptTemplate is missing variables {'ability'}.  Expected: ['ability', 'history', 'question'] Received: ['question', 'history']"

@harris
Copy link
Author

harris commented May 4, 2024

one thing I notice is the input in BasePromptTemplate::invoke would get parsed properly with ability while BasePromptTemplate::ainvoke would not include it.

    def invoke(
        self, input: Dict, config: Optional[RunnableConfig] = None
    ) -> PromptValue:
        config = ensure_config(config)
        if self.metadata:
            config["metadata"] = {**config["metadata"], **self.metadata}
        if self.tags:
            config["tags"] = config["tags"] + self.tags
        return self._call_with_config(
            self._format_prompt_with_error_handling,
            input,
            config,
            run_type="prompt",
        )

    async def ainvoke(
        self, input: Dict, config: Optional[RunnableConfig] = None, **kwargs: Any
    ) -> PromptValue:
        config = ensure_config(config)
        if self.metadata:
            config["metadata"].update(self.metadata)
        if self.tags:
            config["tags"].extend(self.tags)
        return await self._acall_with_config(
            self._aformat_prompt_with_error_handling,
            input,
            config,
            run_type="prompt",
        )

@harris
Copy link
Author

harris commented May 4, 2024

I am able to bypass the input_schema validation so the data would flow through in this PR: #631

please let me know the proper fix.

@ZohaibRamzan
Copy link

looking for the solution!!! Anyone able to manage session id based chat history(each user specific) or pipeline state(langgraph) if we deploy through langserve

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants