Handling follow-up questions in LangChain - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a LangChain ConversationChain that can handle follow-up questions.

LangChain

from langchain.chains import ConversationChain
from langchain.llms import OpenAI

llm = OpenAI(temperature=0)
conversation = ConversationChain(llm=[1])

response = conversation.predict(input="Hello, who won the world series in 2020?")
print(response)

Drag options to blanks, or click blank then click option'

AOpenAI(temperature=0)

Btemperature=0.7

COpenAI()

Dllm

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the class OpenAI instead of an instance.

Forgetting to set temperature=0 for deterministic output.

The ConversationChain requires an llm instance. We create an OpenAI llm with temperature=0 and pass it as llm parameter.

2fill in blank

medium

Complete the code to add memory to the ConversationChain for follow-up question handling.

LangChain

from langchain.memory import ConversationBufferMemory

memory = ConversationBufferMemory(memory_key="chat_history")
conversation = ConversationChain(llm=llm, memory=[1])

response = conversation.predict(input="What about the MVP?")
print(response)

Drag options to blanks, or click blank then click option'

AConversationBufferMemory()

BConversationBufferMemory(memory_key="chat_history")

Cmemory

DConversationBufferMemory(memory_key="history")

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the class instead of the instance.

Using a wrong memory_key that does not match the chain's expectations.

We create a memory instance and pass it to ConversationChain as the memory parameter to keep track of conversation history.

3fill in blank

hard

Fix the error in the code to correctly handle follow-up questions with memory.

LangChain

from langchain.chains import ConversationChain
from langchain.llms import OpenAI
from langchain.memory import ConversationBufferMemory

llm = OpenAI(temperature=0)
memory = ConversationBufferMemory(memory_key="chat_history")
conversation = ConversationChain(llm=llm, memory=[1])

response1 = conversation.predict(input="Who won the world series in 2020?")
response2 = conversation.predict(input="Who was the MVP?")
print(response2)

Drag options to blanks, or click blank then click option'

AConversationBufferMemory()

Bmemory

CConversationBufferMemory(memory_key="history")

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Passing None or a new memory instance without the correct key.

Forgetting to import ConversationBufferMemory.

The ConversationChain needs the memory instance to track conversation history and handle follow-up questions properly.

4fill in blank

hard

Fill both blanks to create a ConversationChain that uses memory and a custom prompt template for follow-up questions.

LangChain

from langchain.prompts import PromptTemplate

prompt = PromptTemplate(template=[1], input_variables=["input", "chat_history"])
conversation = ConversationChain(llm=llm, memory=memory, prompt=[2])

response = conversation.predict(input="Tell me about the 2020 World Series.")
print(response)

Drag options to blanks, or click blank then click option'

A"{chat_history}\nHuman: {input}\nAI:"

B"Human: {input}\nAI:"

Cprompt

DPromptTemplate()

Attempts:

3 left

💡 Hint

Common Mistakes

Using a prompt string instead of a PromptTemplate instance.

Omitting chat_history in the template string.

The prompt template must include chat_history and input placeholders. We pass the prompt instance to ConversationChain.

5fill in blank

hard

Fill all three blanks to build a LangChain ConversationChain that handles follow-up questions with memory and a custom prompt.

LangChain

from langchain.llms import OpenAI
from langchain.memory import ConversationBufferMemory
from langchain.prompts import PromptTemplate
from langchain.chains import ConversationChain

llm = OpenAI(temperature=0)
memory = ConversationBufferMemory(memory_key=[1])
prompt = PromptTemplate(template=[2], input_variables=["input", "chat_history"])
conversation = ConversationChain(llm=llm, memory=memory, prompt=prompt)

response = conversation.predict(input=[3])
print(response)

Drag options to blanks, or click blank then click option'

A"chat_history"

B"{chat_history}\nHuman: {input}\nAI:"

C"What is the capital of France?"

D"history"

Attempts:

3 left

💡 Hint

Common Mistakes

Using a wrong memory_key that does not match the prompt.

Passing a prompt string instead of a PromptTemplate instance.

Passing input without quotes or as a variable.

The memory_key must be 'chat_history' to match the prompt. The prompt template includes chat_history and input placeholders. The input is a question string.