Vector store selection (Pinecone, Chroma, FAISS) in Agentic AI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to initialize a FAISS vector store.

Agentic AI

from langchain.vectorstores import FAISS
vector_store = FAISS.from_texts(texts, embeddings, index_factory=[1])

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

A"Pinecone"

B"Flat"

C"Chroma"

D"Annoy"

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'Pinecone' or 'Chroma' as FAISS index factory names.

Leaving the index_factory parameter empty.

FAISS uses index factories like "Flat" to create a simple vector index. "Pinecone" and "Chroma" are different vector store types, not FAISS index factories.

2fill in blank

medium

Complete the code to create a Chroma vector store with a persist directory.

Agentic AI

from langchain.vectorstores import Chroma
vector_store = Chroma.from_texts(texts, embeddings, persist_directory=[1])

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

A"./chroma_db"

BNone

C"/tmp/faiss"

D"pinecone_index"

Attempts:

3 left

💡 Hint

Common Mistakes

Passing None or invalid paths to persist_directory.

Using paths meant for other vector stores like FAISS or Pinecone.

Chroma vector store uses a directory path string for persist_directory to save the index. './chroma_db' is a valid local folder path.

3fill in blank

hard

Fix the error in the Pinecone initialization code by completing the missing parameter.

Agentic AI

import pinecone
pinecone.init(api_key=[1], environment="us-west1-gcp")

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

A12345

BNone

C"your-pinecone-api-key"

D"chroma"

Attempts:

3 left

💡 Hint

Common Mistakes

Passing None or numeric values as api_key.

Confusing environment names with API keys.

Pinecone requires a valid API key string to initialize. The placeholder 'your-pinecone-api-key' should be replaced with your actual key.

4fill in blank

hard

Fill both blanks to create a FAISS vector store and save it to disk.

Agentic AI

from langchain.vectorstores import FAISS
vector_store = FAISS.from_texts(texts, embeddings, index_factory=[1])
vector_store.[2]("faiss_index")

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

A"Flat"

Bsave_local

Csave_local_index

Dsave

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect method names like 'save' or 'save_local_index'.

Using wrong index factory names.

FAISS uses "Flat" as an index factory for exact search. The method to save the index locally is 'save_local'.

5fill in blank

hard

Fill all three blanks to create a Pinecone vector store client with index name and namespace.

Agentic AI

import pinecone
pinecone.init(api_key=[1], environment=[2])
index = pinecone.Index(name=[3])

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

A"your-pinecone-api-key"

B"us-west1-gcp"

C"my-index"

D"chroma"

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up environment and index name strings.

Using invalid API key formats.

To use Pinecone, you initialize with your API key and environment string, then create an Index object with the index name.