So I think this is a case where you might to use hybrid search. You can store the dates as metadata, and have the LLM create an array with the date to filter by, and then a query for the vector search.
@yoheinakajima
-
AutoGPT Code Review: Impressive Open Source Agent
By
–
Honestly haven’t dug into the code for autoGPT so I don’t know but it looks awesome
-
Efficient Testing Methods for API Integration in Development
By
–
So I assume copy pasting the pull requests changes into a version I have w my api key is NOT the most efficient way to test these…
-
Improving AI Prompts with Context Memory
By
–
Oh shoot you're right. It should be in the prompt. I think adding something like "Take into account previous tasks you have already completed: {context}" in the prompt should work. I don't have time now, but will test and fix.
-
Context Agent Function Integration with Pinecone Vector Database
By
–
Oh I left one comment I should have deleted it seems. The context agent function pulls from pinecone!
-
Pull Request Version Management: Handling Legacy Versions
By
–
Indeed! Wait, so w the pull request – if I pull this 3.5 version what happens to the 3 version? Just exists as an older version people can look up?
-
First Pull Requests: Discovering Open Source Collaboration
By
–
Wait… so people just send you fixes to your code? Open source is amazing. Just did my first handful of pull requests!
-
Making Code Simple: A Developer’s Guide to Clarity
By
–
I often struggle to understand other ppl’s code so tried to make this dead simple!