Comments on: Building “Chat with your Data” Apps using Embeddings, ChatGPT, and Cosmos DB for Mongo DB vCore https://www.relataly.com/step-by-step-guide-to-building-your-own-chatgpt-on-a-custom-knowledge-base-in-python-leveraging-mongo-db-and-embeddings/13687/ The Business AI Blog Mon, 15 Jan 2024 09:53:10 +0000 hourly 1 https://wordpress.org/?v=7.0 By: Mahdi https://www.relataly.com/step-by-step-guide-to-building-your-own-chatgpt-on-a-custom-knowledge-base-in-python-leveraging-mongo-db-and-embeddings/13687/#comment-875 Mon, 15 Jan 2024 09:53:10 +0000 https://www.relataly.com/?p=13687#comment-875 Hi Florian,
Thanks for this great tutorial.
What I couldn’t find in this article and most of the other similar tutorials is: how to keep the index updated in an efficient way for production-ready apps with a Mongo DB (Cosmos Mongo) that has regular data inputs from users through the backend and not necessarily via uploading files.
Do you have any idea how can I do that?
thanks

]]> By: Florian Follonier https://www.relataly.com/step-by-step-guide-to-building-your-own-chatgpt-on-a-custom-knowledge-base-in-python-leveraging-mongo-db-and-embeddings/13687/#comment-778 Wed, 09 Aug 2023 07:06:14 +0000 https://www.relataly.com/?p=13687#comment-778 In reply to Pradeep.

Hi Pradeep,

two options:
1) you can loop over the content and make separate API calls to generate the embeddings
2) or you pass a list to the ada embedding model and get a list of embeddings as a response. -> more information here: https://platform.openai.com/docs/api-reference/embeddings/create

]]> By: Pradeep https://www.relataly.com/step-by-step-guide-to-building-your-own-chatgpt-on-a-custom-knowledge-base-in-python-leveraging-mongo-db-and-embeddings/13687/#comment-769 Fri, 21 Jul 2023 09:43:54 +0000 https://www.relataly.com/?p=13687#comment-769 How we can load multiple documents and get there embeddings?

]]>