LarsDu88

41 ms·

1. 721.▲
   by LarsDu88 2y ago

   Hamill was in multiple sci fi PC game franchises in the 90s. He voice the main protagonist in the game Starsiege as well!
2. 722.▲
   by LarsDu88 2y ago

   I think I should clarify... A multilayer perceptron can classify the red and blue dots if it uses a non-linear activation function for some or most of its layers correct? If its perceptrons all the way down, it will fundamentally reduce dow
3. 723.▲
   by LarsDu88 2y ago

   Thanks for the clarification. I'll update the post!
4. 724.▲
   All the Activations (and a brief history of deep learning)(dublog.net)

   2 points by LarsDu88 2y ago|1 comments
5. 725.▲
   by LarsDu88 2y ago

   My cited and pdf linked notes on activation functions used in deep learning and their historical progression
6. 726.▲
   by LarsDu88 2y ago

   Ok, I get what you mean now. You can build a model by plugging in any activation function into the two slots in the equation at the bottom. There's a typo in the activation function next to "otherwise" in the "Ant Pherom
7. 727.▲
   by LarsDu88 2y ago

   There are a whole lot more activation functions used nowadays in NNs https://dublog.net/blog/all-the-activations/ The author is extrapolating way too much. The simplest model of X is similar to the simplest model
8. 728.▲
   by LarsDu88 2y ago

   6/45 presidents have been shot today. Today that number is 7/45 So it went from 13% to 15%. Not only that, but firearms technology has advanced considerably. If anything, statistically speaking we've been incredibly lucky in
9. 729.▲
   by LarsDu88 2y ago

   Oh boy
10. 730.▲
    by LarsDu88 2y ago

    You make it out as if all the executives at AWS have some master plan surrounding the probably absurd number of GPUs they bought, but the likely answer is its just a bunch of fallible people bandwagoning on the latest trend.
11. 731.▲
    by LarsDu88 2y ago

    I was wondering... this post mentions that ops like sigmoid are very slow. A lot of modern LLMs use activation functions with sigmoid or soft max like SiLU, Swish, and SOLU. Does Relu take less of a performance hit, and if so, maybe it&#x27
12. 732.▲
    by LarsDu88 2y ago

    Whi os it so hard to port FA2 to the m1300 instinct?
13. 733.▲
    by LarsDu88 2y ago

    Darn, beat me too this comment!
14. 734.▲
    by LarsDu88 2y ago

    There is a much simpler approach to making a nice looking resume. Just use markdown > html > pdf using a custom css style sheet This is what I use for my resume: https://github.com/LarsDu/LarryDuMarkdownResume
15. 735.▲
    by LarsDu88 2y ago

    Python AND Rust so it doesn't run like total shiite
16. 736.▲
    by LarsDu88 2y ago

    "The first design used 8 transistors" This gave me a chuckle. I remember long ago, I read the very first Iron Man comic somewhere. I believe his repulsor blasts were powered by the newfangled and exotic "transistor" tech
17. 737.▲
    by LarsDu88 2y ago

    RISC-V is mostly a hedge against ARM at this point IMO. Part of the strategy of commoditizing your complement: https://gwern.net/complement Similar thing going on right now with the rampant open sourcing of LLMs
18. 738.▲
    by LarsDu88 2y ago

    Basically Google has the resources to write their own OS and their own alternative to Java, so they did so as an escape hatch. And Fuschia+Dart+Flutter runs perfectly fine. There's nothing wrong with them. Should Google keep pouring mo
19. 739.▲
    by LarsDu88 2y ago

    I believe Fuschia and usage of Dart for projects like Flutter grew out this back-and-forth legal rigamarole between Google and Oracle over Java. Plus, I believe at the time there was some core Chrome developers Google wanted to retain and k
20. 740.▲
    by LarsDu88 2y ago

    Classic whataboutism. Jeffrey Dahmer killed people but Matthew Broderick did too, so you should never watch "Ferris Bueller's Day Off"
21. 741.▲
    by LarsDu88 2y ago

    My experience is that both PyTorch and Tensorflow are incredibly fickle with the interplay between cuda versioning and python package versioning. You need to go straight into their websites and manually figure out what goes with what. Quite
22. 742.▲
    by LarsDu88 2y ago

    It's kind of ironic that there isn't a decent business model for folks who build tools that uplift the productivity of MILLIONS of devs. I'll just say that even though Linux is FOSS, RedHat still became a multi-million dollar
23. 743.▲
    by LarsDu88 2y ago

    Poetry does cross platform lockfiles, but the absence of cross platform lockfiles is one of the reasons why uv benchmarks significantly faster (on top of using Rust for the resolve)
24. 744.▲
    by LarsDu88 2y ago

    I mention you specifically! Thank you for your oversized contributions to the Python programming landscape. When I hear folks say you need an entire team to do X, Y, Z, I just point them to examples like Charlie Marsh!
25. 745.▲
    by LarsDu88 2y ago

    Yup that's why I wrote it. It's basically my notes dump. Wait and see my next blog post. It's going to be about Genghis Khan and a paternity dispute.
26. 746.▲
    Cartesia Sonic's Text to Speech Generator(play.cartesia.ai)

    2 points by LarsDu88 2y ago|1 comments
27. 747.▲
    by LarsDu88 2y ago

    A recruiter contacted me about potentially interviewing at this company, which was founded by the state space modelers who created Mamba. I was pleasantly surprised to see that their text to speech service is actually better than ElevenLabs
28. 748.▲
    by LarsDu88 2y ago

    In many ways I feel modern gamedev has never really attempted to redo what Spore accomplished (outside of perhaps No Man's Sky). The metaball rigging and texture generation approach feels like its still somewhat relevant. The 3d model
29. 749.▲
    by LarsDu88 2y ago

    I wrote the `so-many-python-package-managers` article and posted it earlier today. I'd like to apologize for misrepresenting `rye` in the first draft. I did not realize it had made so much progress in the span of the past year, and it
30. 750.▲
    by LarsDu88 2y ago

    Sounds like you may want to check out Hatch