Chroma vector store setup in LangChain - Mini Project: Build & Apply

Chroma Vector Store Setup

📖 Scenario: You are building a simple application that stores text data as vectors using the Chroma vector store from LangChain. This will help you quickly find similar texts later.

🎯 Goal: Create a Chroma vector store with some sample documents and set it up to be ready for similarity search.

📋 What You'll Learn

Create a list of documents with exact texts

Set a collection name for the Chroma vector store

Initialize the Chroma vector store with the documents and collection name

Add the final step to persist the vector store

💡 Why This Matters

🌍 Real World

Vector stores help applications find similar texts quickly, useful in chatbots, search engines, and recommendation systems.

💼 Career

Understanding how to set up vector stores is important for roles in AI development, data science, and software engineering working with natural language processing.

1

Create the documents list

Create a list called documents with these exact strings: 'Hello world', 'LangChain is awesome', and 'Vector stores are useful'.

LangChain

# Create the documents list with exact texts
# Your code here

Need a hint?

Use a Python list with the exact strings inside quotes.

2

Set the collection name

Create a variable called collection_name and set it to the string 'my_collection'.

LangChain

documents = ['Hello world', 'LangChain is awesome', 'Vector stores are useful']
# Set the collection name variable
# Your code here

Need a hint?

Assign the exact string 'my_collection' to the variable collection_name.

3

Initialize the Chroma vector store

Import Chroma from langchain.vectorstores and OpenAIEmbeddings from langchain.embeddings. Then create a variable called embedding as an instance of OpenAIEmbeddings(). Finally, create a variable called vectordb by initializing Chroma with embedding, documents, and collection_name as the collection name.

LangChain

documents = ['Hello world', 'LangChain is awesome', 'Vector stores are useful']
collection_name = 'my_collection'
# Import Chroma and OpenAIEmbeddings
# Create embedding instance
# Initialize Chroma vector store with documents and collection_name
# Your code here

Need a hint?

Use Chroma.from_texts with the exact parameters: texts=documents, embedding=embedding, and collection_name=collection_name.

4

Persist the vector store

Call the persist method on the vectordb variable to save the vector store data.

LangChain

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

documents = ['Hello world', 'LangChain is awesome', 'Vector stores are useful']
collection_name = 'my_collection'

embedding = OpenAIEmbeddings()
vectordb = Chroma.from_texts(texts=documents, embedding=embedding, collection_name=collection_name)
# Persist the vector store
# Your code here

Need a hint?

Call persist() on the vectordb variable to save the data.