LangChain Quickstart

Build secure vector search and retrieval applications with Endee and LangChain. This guide walks you through installation, setup, and creating your first vector store.

Requirements

Python 3.8 or higher
An Endee account (sign up at Endee Dashboard )
OpenAI API key (for embeddings)

Installation

Install the required packages:


pip install endee-langchain langchain langchain_openai

Setting up Credentials

Configure your API credentials for Endee and OpenAI:


import os
from langchain_openai import OpenAIEmbeddings
from endee.endee_client import Endee
 
# Set API keys
os.environ["OPENAI_API_KEY"] = "your-openai-api-key"
endee_api_token = "your-endee-api-token"
 
# Initialize Endee client
nd = Endee(token=endee_api_token)

Tip: Store your API keys in environment variables for production use.

Initialize the Embedding Model

Set up OpenAI embeddings for your vector store:


# Initialize the embedding model
embedding_model = OpenAIEmbeddings()
 
# Get the embedding dimension (OpenAI's text-embedding-ada-002 uses 1536 dimensions)
dimension = 1536

Initialize Endee Vector Store

Set up the Endee vector store integration with LangChain:


from endee_langchain import EndeeVectorStore
import time
 
# Create a unique index name with timestamp to avoid conflicts
timestamp = int(time.time())
index_name = f"langchain_demo_{timestamp}"
 
# Initialize the Endee vector store
vector_store = EndeeVectorStore.from_params(
    embedding=embedding_model,
    api_token=endee_api_token,
    index_name=index_name,
    dimension=dimension,
    space_type="cosine"  # Can be "cosine", "l2", or "ip"
)
 
print(f"Initialized Endee vector store with index: {index_name}")

Configuration Options

Parameter	Type	Description	Default
`embedding`	`Embeddings`	LangChain embedding model	Required
`api_token`	`str`	Your Endee API token	Required
`index_name`	`str`	Name of the index	Required
`dimension`	`int`	Vector dimension	Required
`space_type`	`str`	Distance metric	`"cosine"`

Distance Metrics

Metric	Best For
`cosine`	Text embeddings, normalized vectors
`l2`	Image features, spatial data
`ip`	Recommendation systems, dot product similarity

Next Steps

Now that you have your vector store set up, learn how to: