Exercise 7

: Creating applications using different types of LangChains – Simple Sequential, Sequential and map reduce

Ex1 : Simple Sequential LLM CHains - Chains with single input and single output

#Using LLM Chains Suggest a good name for a company that makes a particular product

#Also generate a catchy phrase or a logo for this company

###############################################################################

from langchain_openai import OpenAI

from langchain.prompts import PromptTemplate

from langchain.chains import LLMChain

from langchain.chains import SimpleSequentialChain

llm = OpenAI(temperature = 0.9)

template = "Suggest a good name for a company that makes {product}"

prompt1 = PromptTemplate.from_template(template)

chain1 = LLMChain(llm=llm, prompt = prompt1)

template = "Generate a catchy phrase for this company {company}"

prompt2 = PromptTemplate.from_template(template)

chain2 = LLMChain(llm=llm, prompt = prompt2)

final_chain = SimpleSequentialChain(

chains = [chain1,chain2],

verbose = True)

response = final_chain.run("Noodles")

print(response)

SEQUENTIAL CHAIN

EX2 : USING A SEQUENTIAL LANGCHAIN, FIND THE YEAR WHEN “FORD” WAS LAUNCHED AND LIST
THE TOP 5 CAR MODELS OF THAT YEAR

#Sequential chain with single input and single output variable

from langchain_openai import OpenAI

from langchain import PromptTemplate

from langchain.chains import LLMChain

from langchain.chains import SequentialChain

llm = OpenAI(temperature = 0.5)

prompt1 = PromptTemplate(

input_variables = ['name'],

template = "Tell me when the car model {name} was launched"

chain1 = LLMChain(llm = llm, prompt = prompt1, verbose = True, output_key = 'year')

prompt2 = PromptTemplate(

input_variables = ["year"],

template = "Tell me the top 5 car model in the year {year}"

chain2 = LLMChain(llm = llm, prompt = prompt2, verbose = True, output_key = 'top 5')

final_chain = SequentialChain(

chains = [chain1,chain2],

input_variables = ['name'],

output_variables = ['year', 'top 5'], verbose = True)

response = final_chain({"name": "Ford"})

print(response)

****************************************************************************
Ex3 FIND THE YEAR WHEN “FORD” LAUNCHED “ECO-SPORT”. FIND SIMILAR PRODUCTS LAUNCHED
IN THAT YEAR

# With multiple input Variables

from langchain_openai import OpenAI

from langchain import PromptTemplate

from langchain.chains import LLMChain

from langchain.chains import SequentialChain

llm = OpenAI(temperature = 0.5)

prompt1 = PromptTemplate(

input_variables = ['company', 'product_name'],

template = "Tell me when the {company} launched the product {product_name}?"

chain1 = LLMChain(llm = llm, prompt = prompt1, verbose = True, output_key = 'year')

prompt2 = PromptTemplate(

input_variables = ["year"],

template = "List four similar products launched in the year {year}"

chain2 = LLMChain(llm = llm, prompt = prompt2, verbose = True, output_key = 'four similar')

final_chain = SequentialChain(

chains = [chain1,chain2],

input_variables = ['company', 'product_name'],

output_variables = ['year', 'four similar'], verbose = True)

response = final_chain({"company": "Ford", "product_name":"Eco_sport" })

print(response)

Ex4 : Sequential Chains - Chains with multiple inputs and Outputs

PROBLEM: TRANSLATE THE GIVEN REVIEW To ENGLISH. SUMMARIZE IT. FIND THE LANGUAGE IN
WHICH THE REVIEW IS WRITTEN. WRITE A FOLLOW UP RESPONSE TO THE SUMARY IN A GIVEN
LANGUAGE.

from langchain_openai import OpenAI

from langchain.prompts import PromptTemplate

from langchain.prompts import ChatPromptTemplate

from langchain.chains import LLMChain

from langchain.chains import SimpleSequentialChain

from langchain.chains import SequentialChain

llm = OpenAI(temperature = 0.8)

Review = """Les ordinateurs portables GamersTech impressionne par ses

performances exceptionnelles et son design élégant. De sa configuration

matérielle robuste à un clavier RVB personnalisable et un système de

refroidissement efficace, il établit un équilibre parfait entre prouesses

de jeu et portabilité."""

#Promp Template 1 : Translate the review to English

prompt1 = ChatPromptTemplate.from_template(

"Translate the following to English:"

"\n\n{Review}"

#Chain 1 : Input : Review Output : Eng_review

chain1 = LLMChain(llm = llm, prompt = prompt1, output_key = "Eng_review")

#Promp Template 2 : Summarize the translated review

prompt2 = ChatPromptTemplate.from_template(

"Summarize the following review in 1 sentence"

"\n\n{Eng_review}"

#Chain 2 : Input : Eng_review Output : Summary

chain2 = LLMChain(llm = llm, prompt = prompt1, output_key = "Summary")

#Promp Template 3 : Find the language in which the review is written

prompt3 = ChatPromptTemplate.from_template(

"What is the language in which the review is written"

"\n\n{Review}"

)
#Chain 3 : Input : Review Output : language

chain3 = LLMChain(llm = llm, prompt = prompt3, output_key = "language")

#Promp Template 4 : Write a follow up response to the summary in the language given

prompt4 = ChatPromptTemplate.from_template(

"Write a follow up response to the following summary"

"In the specified language"

"\n\nSummary : {Summary}\n\n Language : {language}"

#Chain 4 : Input : Summary, Language, Output : followup_message

chain4 = LLMChain(llm = llm, prompt = prompt4, output_key = "followup_message")

final_chain = SequentialChain(

chains = [chain1,chain2,chain3,chain4],

input_variables = ["Review"],

output_variables = ["Eng_review", "Summary", "language", "followup_message"],

verbose = True)

response = final_chain(Review)

print("Language of the given review is", response['language'])

print("\nReview translated to English is:", response['Eng_review'])

print("\nSummary of the review is", response['Summary'])

print("\nA follow-up message of the review summary is", response['followup_message'])