This project is a search solution using pgvector for an online retail store product catalog. We’ll build a search system that lets customers provide an item description to find similar items.
For more information, check this blog post, Building AI-powered search in PostgreSQL using Amazon SageMaker and pgvector (2023-05-03)
The overall architecture is like this:
- Deploy the cdk stacks (For more information, see here).
- A SageMaker Studio in a private VPC.
- An Amazon Aurora Postgresql cluster for storing embeddings.
- Aurora Postgresql cluster's access credentials (username and password) stored in AWS Secrets Mananger as a name such as
VSPgVectorStackAuroraPostgr-xxxxxxxxxxxx
.
- Open SageMaker Studio and then open a new System terminal.
- Run the following commands on the terminal to clone the code repository for this project:
git clone https://github.com/ksmin23/semantic-vector-search-with-sagemaker-pgvector.git
- Open
data_ingestion_to_pgvector.ipynb
notebook and Run it. (For more information, see here) - Run Streamlit application. (For more information, see here)