Let us start with the problem. I have a simple Next JS Application. I want it deployed as a static site. The deployment platform should be free for life and it should support CICD. Okay, now let me show you how to use Github pages to check all these boxes.
In this article, I'll show you how to train a Word2vec model using custom data. You can then use the trained model to perform tasks such as similarity search or powering a recommendation engine.
After weeks of massive effort, I finally created my portfolio website and needed a place to deploy. The deployment options I had were Vercel, s3 site on AWS and github pages. I decided to go with github pages because it was free and also simple to deploy. GitHub Pages also had simple CI/CD already integrated. This meant that after a successful setup, all I needed to do was push to production and my application would be deployed.
Keyword search sometimes is not enough. Sometimes you want to search based on meaning. This is where vector search comes in. In this article, I want to take you through how I built a Library Search Engine that is powered by vector search.