how to pass multiple inputs to langserve api in playground

[arvindkrbhatt]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain.schema.runnable import RunnableParallel, RunnablePassthrough
import os

from dotenv import load_dotenv
load_dotenv()  #

from langchain.globals import set_debug
from langchain.globals import set_verbose

set_debug(True)
# set_verbose(True)

#
from langchain_core.prompts import ChatPromptTemplate, PromptTemplate
from langchain_core.output_parsers import StrOutputParser

assistant_template = """
You are {sentiment} assistant that responds to user comments,
using similar vocabulary than the user.
User:" {customer_request}"
Comment:
"""
#Create the prompt template to use in the Chain for the first Model.
assistant_prompt_template = ChatPromptTemplate(
    input_variables=["sentiment", "customer_request"],
    template=assistant_template
)

output_parser = StrOutputParser()


from langchain_community.chat_models import AzureChatOpenAI
model = AzureChatOpenAI(deployment_name="V123",azure_endpoint="https://az123.openai.azure.com",openai_api_key="2123",api_version="2024-02-15-preview")

assistant_chain = assistant_prompt_template | model | output_parser

# Add typing for input
from langchain_core.pydantic_v1 import BaseModel

class Question(BaseModel):
    __root__: str

chain = assistant_chain.with_types(input_type=Question)

Description

1. I am trying to pass 2 input parameter in langserve api
2. I am unable to pass 2 input parameter 'sentiment' and 'customer_request' from api playground

System Info

Windows
langchain 0.1.11
langchain-cli 0.0.21
langchain-community 0.0.25
langchain-core 0.1.29
langchain-experimental 0.0.53
langchain-google-genai 0.0.9
langchain-openai 0.0.8
langchain-text-splitters 0.0.1
langchainhub 0.1.15
langserve 0.0.46
langsmith 0.1.19
Python 3.9.11

[dosubot[bot]]

Hey @arvindkrbhatt, great to see you here again! Hope you're doing well.

Based on the code you've provided, it seems like you're trying to pass two parameters, 'sentiment' and 'customer_request', to the assistant_prompt_template instance of ChatPromptTemplate. However, the with_types method is being called on assistant_chain with input_type=Question, where Question is a class with a single __root__ attribute. This might be the source of the issue, as the with_types method is expecting a Pydantic model that matches the structure of the input data.

To pass multiple parameters, you should define a Pydantic model that has attributes for each of the parameters you want to pass. Here's an example:

from pydantic import BaseModel

class AssistantInput(BaseModel):
    sentiment: str
    customer_request: str

Then, you can use this model in the with_types method:

chain = assistant_chain.with_types(input_type=AssistantInput)

Now, when you call the chain with an instance of AssistantInput, both 'sentiment' and 'customer_request' should be correctly passed to the assistant_prompt_template.

result = chain(AssistantInput(sentiment="positive", customer_request="Hello, assistant!"))

Please note that the with_types method is used to enforce type checking and conversion for the input and output of a Runnable (in this case, assistant_chain). The input_type and output_type parameters should be subclasses of Pydantic's BaseModel, and they should match the structure of the input and output data, respectively.

I hope this helps! If you have any other questions or if something is not clear, please let me know.

Sources

• libs/langchain/langchain/chains/api/openapi/chain.py
• libs/langchain/langchain/agents/load_tools.py
• libs/langchain/langchain/chains/retrieval_qa/base.py
• libs/langchain/langchain/chains/openai_functions/citation_fuzzy_match.py
• libs/core/langchain_core/runnables/passthrough.py

About Dosu

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.

[latifs]

@dosu-bot,
how would I access these newly passed values in my chain?