Does langchain support Input Parsing?

[blacksmithop]

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

It is currently possibly to define custom output parsers for Runnables and Agents

Description

I was wondering if there any similar support for the input to the LLM, Chain or Agent.

System Info

langchain==0.2.2
langchain-community==0.2.3
langchain-core==0.2.4
langchain-openai==0.1.8
langchain-text-splitters==0.2.1

Windows
Python 3.11.5

[dosubot[bot]]

Yes, LangChain supports input parsing for Runnables, Chains, or Agents, similar to how it supports custom output parsers. This is evident from the usage of the inputs dictionary in various methods, which allows for the parsing and handling of input data. For example:

inputs = {**kwargs, **{"intermediate_steps": intermediate_steps}}
final_output: Any = None
if self.stream_runnable:
    async for chunk in self.runnable.astream(
        inputs, config={"callbacks": callbacks}
    ):
        if final_output is None:
            final_output = chunk
        else:
            final_output += chunk
else:
    final_output = await self.runnable.ainvoke(
        inputs, config={"callbacks": callbacks}
    )
return final_output

This demonstrates that LangChain can handle and parse inputs for its components, similar to how it handles custom output parsing ^[1].

To continue talking to Dosu, mention @dosu.