Vector Databases

What are Vector Databases?

Vector databases are specialized data stores designed to efficiently store, index, and query high-dimensional vectors (embeddings). They enable similarity search, finding vectors that are close to a query vector in the embedding space.

Index Algorithms

Using Vector Databases

# Using FAISS
import faiss
import numpy as np

class FAISSVectorStore:
    def __init__(self, dimension=768):
        self.dimension = dimension
        self.index = faiss.IndexFlatL2(dimension)
        self.documents = []

    def add_documents(self, embeddings, documents):
        self.index.add(np.array(embeddings).astype('float32'))
        self.documents.extend(documents)

    def search(self, query_embedding, k=5):
        distances, indices = self.index.search(
            np.array([query_embedding]).astype('float32'), k
        )
        results = []
        for i, idx in enumerate(indices[0]):
            results.append({
                "document": self.documents[idx],
                "distance": distances[0][i]
            })
        return results

# Using ChromaDB
import chromadb

class ChromaVectorStore:
    def __init__(self, collection_name="documents"):
        self.client = chromadb.Client()
        self.collection = self.client.create_collection(collection_name)

    def add_documents(self, ids, documents, embeddings):
        self.collection.add(
            ids=ids,
            documents=documents,
            embeddings=embeddings
        )

    def search(self, query_embedding, k=5):
        results = self.collection.query(
            query_embeddings=[query_embedding],
            n_results=k
        )
        return results

Comparison: Vector Databases

Similarity Metrics

import numpy as np

def cosine_similarity(a, b):
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))

def euclidean_distance(a, b):
    return np.linalg.norm(a - b)

def dot_product(a, b):
    return np.dot(a, b)

Summary

Vector databases are essential infrastructure for AI applications requiring similarity search. Choose based on your scale, performance, and feature requirements.

Next: We'll explore embedding models in detail.