LangChain architecture overview - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the core LangChain class for building chains.

LangChain

from langchain.chains import [1]

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

AChain

BLLMChain

CPromptTemplate

DMemory

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing 'Chain' which is a base class but not commonly imported directly.

Choosing 'PromptTemplate' which is for prompts, not chains.

The LLMChain class is the core chain class that connects prompts with language models in LangChain.

2fill in blank

medium

Complete the code to create a prompt template with a variable placeholder.

LangChain

from langchain.prompts import PromptTemplate

prompt = PromptTemplate(template="Hello, {name}!", input_variables=["[1]"])

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

Aname

Buser

Cinput

Dtext

Attempts:

3 left

💡 Hint

Common Mistakes

Using a placeholder that does not match the input variable name.

Using generic words like 'user' or 'input' that are not declared.

The placeholder in the template must match the variable name in input_variables. Here, it is name.

3fill in blank

hard

Fix the error in creating an LLMChain by filling the missing argument.

LangChain

from langchain.llms import OpenAI
from langchain.chains import LLMChain

llm = OpenAI()
chain = LLMChain(llm=llm, prompt=[1])

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

Aprompt_template

BPromptTemplate

Ctemplate

Dprompt

Attempts:

3 left

💡 Hint

Common Mistakes

Using the class name instead of the variable.

Using incorrect argument names.

The LLMChain expects a prompt argument which is an instance of PromptTemplate. The variable holding the prompt is usually named prompt.

4fill in blank

hard

Fill both blanks to create a simple memory-enabled chain that remembers previous inputs.

LangChain

from langchain.memory import [1]

memory = [2]()
chain = LLMChain(llm=llm, prompt=prompt, memory=memory)

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

AConversationBufferMemory

BConversationSummaryMemory

Attempts:

3 left

💡 Hint

Common Mistakes

Confusing summary memory with buffer memory.

Using different names for import and instantiation.

The ConversationBufferMemory class stores the conversation history in a buffer. It is imported and instantiated to enable memory in the chain.

5fill in blank

hard

Fill all three blanks to create a chain that uses an OpenAI LLM, a prompt template, and conversation memory.

LangChain

from langchain.llms import [1]
from langchain.prompts import [2]
from langchain.memory import [3]

llm = [1]()
prompt = [2](template="Say hello to {name}", input_variables=["name"])
memory = [3]()
chain = LLMChain(llm=llm, prompt=prompt, memory=memory)

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

AOpenAI

BPromptTemplate

CConversationBufferMemory

DLLMChain

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up class names or forgetting to import memory.

Using the chain class as an import in place of prompt or memory.

This code imports and creates instances of OpenAI for the language model, PromptTemplate for the prompt, and ConversationBufferMemory for memory. These are combined in an LLMChain.