Congratulations!

Introduction to Embeddings with the OpenAI API

Emmanuel Pire

Senior Software Engineer, DataCamp

Chapter 1 - What are Embeddings?

Embeddings: vector/numerical representation of text
Capture the semantic meaning of text
Used OpenAI's Embedding model
Can use the cosine distance to find similar texts
Unlocks semantic search, recommendation engines, etc.

A plot of the 2D vector space showing that reviews with the same sentiment and topic are mapped more closely together in the vector space.

Chapter 2 - Embeddings for AI Applications

Semantic search and recommendation

An embedded search query in the vector space alongside several embedded articles on different topics.

Classification

Embedded class descriptions in the vector space, with an unknown vector assigned the Tech label.

Chapter 3 - Vector Databases

A diagram showing how a vector database, Chroma in this case, is used to store document embeddings and returned results based on an embedded query back to the user.

Where next?

Cloud-based, managed vector databases

Pinecone:
- Semantic Search with Pinecone (Code-along)
Weaviate:
- Vector Databases for Data Science with Weaviate in Python (Code-along)

Frameworks for creating applications

LangChain:
- How to Build LLM Applications with LangChain (Tutorial)
- Introduction to Large Language Models with GPT & LangChain (Code-along)

Let's practice!

Introduction to Embeddings with the OpenAI API