How can I combine outputs from multiple LangChain prompts using SequentialChain in Python?

When you use a SequentialChain, the final output is determined by the output variables you specify when constructing the chain. In your code you only set output_variables=["dob"], so only the result from chain3 is returned. To get the results from all chains and combine them, you need to do two things:

1. Include all the outputs you need (e.g., "person", "country", "dob") in the output_variables argument of SequentialChain.
2. Format or combine those outputs as desired when displaying the result.

Below is an updated version of your code:

PYTHONCopied
import os
from constent import openai_key

from langchain_community.llms import OpenAI
import streamlit as st
from langchain import PromptTemplate
from langchain.chains import LLMChain, SequentialChain

os.environ["OPENAI_API_KEY"] = openai_key

st.title('Search about person')

input_text = st.text_input("Search about the person")

llm = OpenAI(temperature=1.0)

# Prompt Engineering.
prompt1 = PromptTemplate(
    input_variables=["name"],  # placeholder for the input text
    template="Give the full name of {name}"
)
chain1 = LLMChain(llm=llm, prompt=prompt1, verbose=True, output_key='person')

prompt2 = PromptTemplate(
    input_variables=["person"],  # placeholder for the person output from chain1
    template="Give the name of the country that {person} belongs to?"
)
chain2 = LLMChain(llm=llm, prompt=prompt2, verbose=True, output_key='country')

prompt3 = PromptTemplate(
    input_variables=["person"],
    template="What is the date of birth of {person}?"
)
chain3 = LLMChain(llm=llm, prompt=prompt3, verbose=True, output_key='dob')

# Include all outputs from each chain
combined_chain = SequentialChain(
    chains=[chain1, chain2, chain3],
    input_variables=["name"],
    output_variables=["person", "country", "dob"],
    verbose=True
)

if input_text:
    outputs = combined_chain.run({"name": input_text})
    # Combine the outputs into a single message as expected
    full_output = (f"{outputs['person']} and the date of birth of {outputs['person']} is {outputs['dob']}. "
                   f"He is from {outputs['country']}.")
    st.write(full_output)

Explanation

• Output Variables: By setting output_variables=["person", "country", "dob"], the SequentialChain now returns a dictionary containing the output of each chain.
• Formatting the Final Result: The combined result is formatted into one output string which includes the full name (from chain1), the date of birth (from chain3), and the country (from chain2).

Now when you pass "Dhoni" as the input text, you should receive an output similar to:

"Mahendra Singh Dhoni and the date of birth of Mahendra Singh Dhoni is July 7, 1981. He is from India."