I’m looking into building search using OpenAI’s embeddings model. (OpenAI API)
The main area I’m getting stuck is calculating the cosine differential between the search term and the indexed database to find the most related items.
Has anyone built with embeddings in a Bubble app yet? If so, would love to learn more about how to implement it.
Me too.
You can create a plugin to do this. It’s not that complicated, since you can ask ChatGPT itself how to do it and here in javascript. Or simply take the website codes below, which are written in Python, and ask ChatGPT to rewrite them in javascript and then put everything in a plugin.
Semantic text search using embeddings
How’d you end up?
How large of a document are you trying to embed? BestBubbleDev just released a video on how to search with Typesense
This topic was automatically closed after 70 days. New replies are no longer allowed.
