Chroma vector store setup in LangChain - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ component_behavior

intermediate

Chroma vector store initialization output

What will be the output behavior when initializing a Chroma vector store with the following code snippet?

LangChain

from langchain.vectorstores import Chroma
from langchain.embeddings import OpenAIEmbeddings

embeddings = OpenAIEmbeddings()
vectorstore = Chroma(collection_name="test_collection", embedding_function=embeddings)
print(vectorstore._collection_name)

A Raises a TypeError because embedding_function expects a list, not an object

B Raises a NameError because OpenAIEmbeddings is not imported

C Prints the memory address of the vectorstore object

D Prints 'test_collection' indicating the collection name used

💡 Hint

Check how the collection_name attribute is stored and accessed in Chroma vector store.

The Chroma vector store stores the collection name in the _collection_name attribute. Printing it outputs the string 'test_collection'.

📝 Syntax

intermediate

Correct syntax for adding documents to Chroma vector store

Which option correctly adds a list of documents to an existing Chroma vector store instance named 'vectorstore'?

LangChain

docs = ["Hello world", "Langchain is great"]

A vectorstore.add_documents(docs)

B vectorstore.add(docs)

C vectorstore.append_documents(docs)

D vectorstore.insert_documents(docs)

💡 Hint

Check the official method name for adding documents in Chroma vector store.

The correct method to add documents to a Chroma vector store is add_documents. Other method names do not exist and cause AttributeError.

🔧 Debug

advanced

Identify the error in Chroma vector store setup

What error will this code raise when trying to create a Chroma vector store without specifying the embedding function?

LangChain

from langchain.vectorstores import Chroma

vectorstore = Chroma(collection_name="my_collection")

A TypeError: __init__() missing 1 required positional argument: 'embedding_function'

B ValueError: collection_name cannot be empty

C RuntimeError: Chroma client not found

D No error, vector store initializes with default embedding function

💡 Hint

Check the required parameters for Chroma vector store initialization.

The embedding_function parameter is required when creating a Chroma vector store. Omitting it causes a TypeError.

❓ state_output

advanced

Number of documents after adding to Chroma vector store

Given this code, what is the number of documents stored in the vector store after execution?

LangChain

from langchain.vectorstores import Chroma
from langchain.embeddings import OpenAIEmbeddings

embeddings = OpenAIEmbeddings()
vectorstore = Chroma(collection_name="col1", embedding_function=embeddings)
docs = ["doc1", "doc2", "doc3"]
vectorstore.add_documents(docs)
count = vectorstore._collection.count()
print(count)

A Raises AttributeError because _collection has no count method

B 0

C 3

D Raises RuntimeError because documents are not indexed yet

💡 Hint

Check how add_documents affects the internal collection count.

After adding 3 documents, the internal collection count method returns 3.

🧠 Conceptual

expert

Effect of persistent directory in Chroma vector store

What is the effect of specifying the 'persist_directory' parameter when initializing a Chroma vector store?

A It encrypts the vector store data in memory for security

B It saves the vector store data to disk, allowing data to persist across program runs

C It limits the vector store size to the specified directory's free space

D It disables in-memory caching, forcing all queries to read from disk

💡 Hint

Think about why you would want to save vector data to a folder.

The persist_directory parameter tells Chroma where to save data on disk so it can be reused later without rebuilding the index.