Basic RAG chain with LCEL in LangChain - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the LangChain RetrievalQA class.

LangChain

from langchain.chains import [1]

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

AVectorStore

BTextSplitter

CRetrievalQA

DOpenAI

Attempts:

3 left

💡 Hint

Common Mistakes

Importing unrelated classes like TextSplitter or OpenAI instead of RetrievalQA.

The RetrievalQA class is used to create a retrieval-augmented generation chain in LangChain.

2fill in blank

medium

Complete the code to initialize the RetrievalQA chain with the retriever.

LangChain

qa_chain = RetrievalQA.from_chain_type(llm=llm, retriever=[1])

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

Aquery

Bembedding

Cdocument

Dretriever

Attempts:

3 left

💡 Hint

Common Mistakes

Passing embedding or document instead of the retriever.

The retriever object is passed to the RetrievalQA chain to fetch relevant documents.

3fill in blank

hard

Fix the error in the code to run the RAG chain and get the answer.

LangChain

result = qa_chain.[1](query)

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

Arun

Bsearch

Cpredict

Danswer

Attempts:

3 left

💡 Hint

Common Mistakes

Using search or predict instead of run to execute the chain.

The run() method executes the chain with the input query and returns the answer.

4fill in blank

hard

Fill both blanks to create a LangChain LLMChain with a prompt template and an LLM.

LangChain

from langchain.chains import [1]
from langchain.prompts import [2]
llm_chain = LLMChain(llm=llm, prompt=prompt)

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

ALLMChain

BPromptTemplate

CRetriever

DVectorStore

Attempts:

3 left

💡 Hint

Common Mistakes

Importing unrelated classes like Retriever or VectorStore here.

LLMChain and PromptTemplate are imported to create a chain with a prompt and language model.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension filtering documents by length and score.

LangChain

filtered_docs = {doc: score for doc, score in docs.items() if len(doc) [1] 100 and score [2] 0.5 and doc not in [3]

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

A<

B>

C>=

Dexcluded_docs

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong comparison operators or forgetting to exclude documents.

We filter documents shorter than 100 characters, with score greater than 0.5, and not in excluded_docs.