Do you have any example data sets that are good ingesting , maybe something with not to many embeddings. I want to containerize this in docker for fun.

[JasonMarkWomack]

Do you have any example data sets that are good ingesting , maybe something with not to many embeddings. I want to containerize this in docker for fun.

[Stevenic]

@JasonMarkWomack I just checked in a new Wikipedia Sample Dataset along with detailed instructions for manually building a Vectra index using the CLI. I'll be adding RAG sample this week that shows how to use Vectra with an LLM like ChatGPT.

https://github.com/Stevenic/vectra/tree/main/samples/wikipedia#readme

Make sure you install the latest version of Vectra globally using npm install - g vectra. It has a couple of key bug fixes that will improve the over quality of the text you get back from vectra. The TextSplitter class needed a bug fix...