kaspermarstal

8 ms·

1. 31.▲
   by kaspermarstal 2y ago

   Yes, and it will be
2. 32.▲
   by kaspermarstal 2y ago

   I am not going to claim or report any kind of accuracy, especially with such a small model and such a specific, context dependent use case. It is the user’s responsibility to cross validate if it’s accurate enough for their use case and upg
3. 33.▲
   by kaspermarstal 2y ago

   She and one other researcher has manually classified all 7000 papers as per standard protocol. Perhaps for the next article they will measure how this tool agreed with them against them and include it in the protocol if good enough.
4. 34.▲
   by kaspermarstal 2y ago

   I don't know. This paper [1] reports accuracies in the 97-98% range on a similar task with more powerful models. With Gemma 2 2b the accuracy will certainly be lower. [1] https://www.medrxiv.org/content/10.1101&#x2
5. 35.▲
   by kaspermarstal 2y ago

   I built an Excel Add-In that allows my girlfriend to quickly filter 7000 paper titles and abstracts for a review paper that she is writing [1]. It uses Gemma 2 2b which is a wonderful little model that can run on her laptop CPU. It works su
6. 36.▲
   by kaspermarstal 2y ago

   Thanks for sharing! If you want to try it out, it will be published to github.com/kaspermarstal/cellm in the coming weeks. Will reply to this thread when it is live.
7. 37.▲
   by kaspermarstal 2y ago

   What’s your use case?
8. 38.▲
   by kaspermarstal 2y ago

   Yes I am really interested in good use-cases, send me an e-mail at kaspermarstal@gmail.com! It’s called Cellm, implemented with ExcelDna and is fully functional for Anthropic models. Currently preparing the repo for public release and writ
9. 39.▲
   by kaspermarstal 2y ago

   Thanks!
10. 40.▲
    by kaspermarstal 2y ago

    Omg instantly bookmarked. How to find "Military science fiction series" that are "Complex and Strategic" and includes "Space battles and tactics", "Action-packed sequences", "Strategic dialogue&q
11. 41.▲
    by kaspermarstal 2y ago

    I am building an Excel extension that lets me use LLMs in formulas. For example, I can write =PROMPT(A1:E1, "Extract keywords") in a cell to extract keywords from a row and drag the cell to apply the prompt to many other rows. I f
12. 42.▲
    by kaspermarstal 2y ago

    Can anyone explain what is gained by training a model? Why not use the foundational LLM for the relevance, evidence, and trajectory processes?
13. 43.▲
    by kaspermarstal 2y ago

    > Don’t ask me about covariance since I honestly don’t know the answer. I recommend reading the section on subtyping and variance and The Rustonomicon. I went through them a few times but still don’t entirely get all the details. Then, w
14. 44.▲
    by kaspermarstal 2y ago

    Very cool! A assume you use Postgres' native full-text search capabilities? Any plans for BM25 or similar? This would make Korvus the end-game for open source rag IMO.
15. 45.▲
    by kaspermarstal 3y ago

    This extension has been developed on top of pgrx and depends on the platforms that pgrx supports. From the pgrx readme: > Windows is not supported. It could be, but will require a bit of work with cargo-pgrx and figuring out how to compi
16. 46.▲
    by kaspermarstal 3y ago

    Yes, the PL/Rust code base was a very useful when developing this extension
17. 47.▲
    by kaspermarstal 3y ago

    Good suggestion, thanks
18. 48.▲
    Show HN: PRQL in PostgreSQL(github.com)

    267 points by kaspermarstal 3y ago|139 comments
19. 49.▲
    by kaspermarstal 3y ago

    Incredible. RAG will be obsolete in a year or two.
20. 50.▲
    by kaspermarstal 3y ago

    Mhm, interesting development, the paper has been withdrawn.
21. 51.▲
    by kaspermarstal 3y ago

    Can you elaborate? Harder as in writing queries or harder as in making queries performant?
22. 52.▲
    by kaspermarstal 3y ago

    None that I am aware of, and that is why I am very interested in learning about any red flags you might have found that motivated your decision to move off Postgres/Supabase and that others should be aware of.
23. 53.▲
    by kaspermarstal 3y ago

    I’m curious if you investigated the TimescaleDB extension that is built into Supabase for your usecase? And if so, what was the pros and cons?
24. 54.▲
    by kaspermarstal 3y ago

    If they were serious about emissions they should not have filed a patent.
25. 55.▲
    by kaspermarstal 3y ago

    I found that having GPT-4 return a JSON dictionary where keys are line numbers and values are their corresponding lines work the best.
26. 56.▲
    Ask HN: What do you want for Christmas that costs less than $100?

    21 points by kaspermarstal 4y ago|28 comments
27. 57.▲
    by kaspermarstal 4y ago

    Can you tell who develops Hasura Auth and Hasura Storage? You? Or did you fork it and extend it?
28. 58.▲
    by kaspermarstal 4y ago

    Sure! There is tealengine.com and you can follow github.com/tealmedical to get notified when we put it up. What are you interested in? The foreign data wrapper or the iso?
29. 59.▲
    by kaspermarstal 4y ago

    A postgres foreign data wrapper for PACS so you can send and receive medical images with SQL. DICOM, the medical imaging standard, is from the 80's and you need a bit of domain knowledge to integrate e.g. AI with hospitals. This makes
30. 60.▲
    by kaspermarstal 5y ago

    I've been looking for something like this for a long time, so cool! Great work! Does it work with TimescaleDB?