Francisco Sosa

@fjsosah

Currently down rabbit holes of technical arcana. ML | Bayesian | FP, PLT and 🐈 theory via Haskell | Data Science | AI Lecturer

Guatemala
Vrijeme pridruživanja: listopad 2011.

Tweetovi

Blokirali ste korisnika/cu @fjsosah

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @fjsosah

  1. proslijedio/la je Tweet
    prije 10 sati

    Normalisation, how to do it fast for the untyped lambda-calculus by means of evaluation, and why we care (together with some of the required type theory/PLT background): first of a series, hopefully. Feel free to yell at me in replies about mistakes etc!

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet

    I combined the content from my previous "Intermediate Streamlit" article plus some new notes I'd taken into my own "Streamlitopedia". If you're using , maybe there's something helpful in here for you.

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    2. velj

    Here are some thoughts on stats “vs” ML. Rel btwn the fields, why debates on topic are almost always fundamentally flawed, why I think there’s so much talking past each other, and why I wish we could focus instead on *specific* methods + problems 2/3

    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    1. velj

    I must highlight this great post about covariance and contravariance, and not just for . The fact that co/contravariance have an even/odd parity structure (where odd*odd = even, etc) is key to many confusions Ive had in the past in math/physics,etc

    Poništi
  5. proslijedio/la je Tweet
    2. velj

    "Machine Learning doesn’t have to be a black box anymore. What use is a good model if we cannot explain the results to others. Interpretability is as important as creating a model." A neat kernel on "Intrepreting Machine Learning models" by

    Poništi
  6. proslijedio/la je Tweet
    1. velj

    Creating a serendipitous connection, freely offering a good idea, telling someone you see their potential & you'll be in their corner... these are both cheap & costly, the currency of trust in the human project Invest this way and you'll have the world's most joyous nest egg

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    31. sij

    A few years ago TensorFlow looked indomitable. Now it's on the verge of being overtaken by Pytorch. OpenAI just announced it's standardizing on Pytorch, joining the likes of Tesla, Nvidia, and FB. Google hoped TensorFlow would cement its lead in ML. It may soon be a liability.

    Poništi
  8. proslijedio/la je Tweet
    31. sij

    An Opinionated Guide to ML Research: “To make breakthroughs with idea-driven research, you need to develop an exceptionally deep understanding of your subject, and a perspective that diverges from the rest of the community—some can do it, but it’s hard.”

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    31. sij

    If you're running xgboost, lgbm, or other forest based models in production you need to check out our new forest inference library. 40x faster predictions, cheaper than cpu and way less rack space.

    Poništi
  10. proslijedio/la je Tweet
    30. sij

    A lot gets written about how to write Haskell code, but much less about how to successfully hunt down performance bugs using profiling. That makes our latest blog post all the more interesting: “Locating performance bottlenecks in large Haskell codebases”

    Poništi
  11. proslijedio/la je Tweet

    Thanks for the invitation to talk about why I believe NLP is the most important field of Machine Learning. Who doesn't agree with me?

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    Poništi
  13. proslijedio/la je Tweet
    28. sij

    New paper: Towards a Human-like Open-Domain Chatbot. Key takeaways: 1. "Perplexity is all a chatbot needs" ;) 2. We're getting closer to a high-quality chatbot that can chat about anything Paper: Blog:

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    28. sij

    network protocols

    Poništi
  15. proslijedio/la je Tweet
    28. sij

    I just published my initial devcontainer setup for , allowing you to install a Haskell environment (via HIE) with zero effort! Please try it, and send PRs to make it better (I really have no idea what I'm doing 😅)

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    28. sij

    Introducing the new Thinc, a refreshing functional take on deep learning! 🔮 Static type checking 🔥 Mix , & ⛓️ Integrated config system 🧮 Extensible backends incl. JAX (experimental) 🧬 Variable-length sequences & more

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    27. sij

    Haskell-ide-engine and Ghcide teams are joining forces to work on a single Haskell IDE. See for details. With , and many contributors from both sides.

    Poništi
  18. 27. sij

    Any twitter friends that can invite me to ?

    Poništi
  19. proslijedio/la je Tweet
    25. sij

    GPT-2 can generate some impressive looking sentences, but claims that "anything that looks like genuine understanding is just an illusion."

    Poništi
  20. proslijedio/la je Tweet
    24. sij

    Uh this emacs region undo thing is kinda blowing my mind. I had no idea!

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·