langchain/docs/examples/chains/transformation.ipynb

Transformation Chain

This notebook showcases using a generic transformation chain.

As an example, we will create a dummy transformation that takes in a super long text, filters the text to only the first 3 paragraphs, and then passes that into an LLMChain to summarize those.

In [5]:

from langchain.chains import TransformChain, LLMChain, SimpleSequentialChain
from langchain.llms import OpenAI
from langchain.prompts import PromptTemplate

In [7]:

with open('../state_of_the_union.txt') as f:
    state_of_the_union = f.read()

In [2]:

def transform_func(inputs: dict) -> dict:
    text = inputs["text"]
    shortened_text = "\n\n".join(text.split("\n\n")[:3])
    return {"output_text": shortened_text}

transform_chain = TransformChain(input_variables=["text"], output_variables=["output_text"], transform=transform_func)

In [4]:

template = """Summarize this text:

{output_text}

Summary:"""
prompt = PromptTemplate(input_variables=["output_text"], template=template)
llm_chain = LLMChain(llm=OpenAI(), prompt=prompt)

In [6]:

sequential_chain = SimpleSequentialChain(chains=[transform_chain, llm_chain])

In [8]:

sequential_chain.run(state_of_the_union)

Out [8]:

' This speech addresses the American people and acknowledges the difficulties of last year due to COVID-19. It emphasizes the importance of coming together regardless of political affiliation and encourages a sense of unity as Americans.'

In [ ]: