• 0 Posts
  • 18 Comments
Joined 10 months ago
cake
Cake day: December 30th, 2023

help-circle
  • Yeal I get it, I find it hard to rely on online recommendations too. But still, I find it funny that I personally haven’t seen any kagi comment and you have seen enough already to grow skeptical. Because of that, and because I have enjoyed using it, I wanted to share it. Could have gone deeper on the reasons I like it but it would sound more forced/advertisy imo (e.g. “For only xyz per month you get 100% private reliable results with xyz features. I have used it for xyz months and it has never failed” etc).

    Still kinda annoying to get implied to be a bot/paid rando when sharing something cool but that’s the web

    Edit: I took a second looks to other comments and I can see that kagi is indeed mentioned quite a bit so nvm that. I’m glad to hear more people are starting to use it now :)











  • alehc@slrpnk.nettoScience Memes@mander.xyzName & shame. :)
    link
    fedilink
    English
    arrow-up
    1
    ·
    5 months ago

    That’s not necessarily true. General-purpose 3rd party models (chatgpt, llama3-70b, etc) perform surprisingly good in very specific tasks. While training or finetuning your specialized model should indeed give you better results, the crazy amount of computational resources and specialized manpower needed to accomplish it makes it unfeasible and unpractical in many applications. If you can get away with an occational “as an AI model…”, you are better off using existing models.







  • Sounds like a cool project! I have some questions just out of curiosity.

    • What type of model are you considering? Would it potentially fit in mobile devices? Would love to open an app or smth after meeting a random cat in the street to estimate its age lol.
    • How many recordings are you planning to work with? I don’t have experience with audio but with NLP and many other domains NNs require huge amounts of data to work correctly. What are you thinking to mitigate this issue?
    • Do you plan to open-source the model or notify us again when its ready somehow?

    Good luck with the project! My cats very rarely meow but if I get to record them I’ll send it to ya.