Nikhil Thorat

@nsthorat

TensorFlow.js // Google Brain // PAIR

Cambridge, MA
js.tensorflow.org
Vrijeme pridruživanja: siječanj 2013.
80 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @nsthorat

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

  1. prije 7 sati

    Tesla researchers authors NNs in python land and rewrite with a bare C++ implementation when deploying. This feels like a failure of our tooling / infrastructure.

    Our NN is initially in Python for rapid iteration, then converted to C++/C/raw metal driver code for speed (important!). Also, tons of C++/C engineers needed for vehicle control & entire rest of car. Educational background is irrelevant, but all must pass hardcore coding test.
    Prikaži ovu nit
    28 replies 51 proslijeđeni tweet 314 korisnika označava da im se sviđa
    Poništi
  2. 2. velj

    Hacking the features not the bugs, this is \m/etal

    99 smartphones are transported in a handcart to generate virtual traffic jam in Google Maps. Through this activity, it is possible to turn a green street red which has an impact in the physical world by navigating cars on another route!
    Prikaži ovu nit
    11 korisnika označava da im se sviđa
    Poništi
  3. 30. sij

    Pay attention to this. We’re seeing up to 3X speed ups for real world models in the tfjs WASM backend with SIMD128 enabled!

    WebAssembly brings SIMD support to the Web. What is it? and explain and demonstrate how to use it to run native code on the Web even faster! 🔥 (This feature is currently experimental and available under a flag.)
    33 proslijeđena tweeta 142 korisnika označavaju da im se sviđa
    Poništi
  4. proslijedio/la je Tweet
    29. sij

    Calling all Web and JS Developers! ☕️⚡️ The team wants YOUR valuable feedback to help shape the future of machine learning in the browser and beyond. Take the 2020 survey →

    TensorFlow.js survey 2020
    67 proslijeđenih tweetova 147 korisnika označava da im se sviđa
    Poništi
  5. 29. sij

    viscoelasticgirl

    2 korisnika označavaju da im se sviđa
    Poništi
  6. 29. sij

    flow of thoughts w/ tfjs & face-api demo here👇

    0:41
    Flow of thoughts. Built with 's face-api.js, if you are into face detection in JavaScript, go check it out!
    Prikaži ovu nit
    3 proslijeđena tweeta 21 korisnik označava da mu se sviđa
    Poništi
  7. 29. sij

    We are highly capable programs that are self aware / self-editable, emerging from billions of years of hacky legacy code that was never refactored, is littered with leaky abstractions, evolved under wildly different constraints. Yet somehow I can beam this msg to you via space..

    2 proslijeđena tweeta 19 korisnika označava da im se sviđa
    Poništi
  8. proslijedio/la je Tweet
    28. sij
    Odgovor korisnicima

    Yes! First please register at and note in 6. you’d like to see this topic discussed. I also encourage you to submit a position statement and invite you to give a talk on the topic.

    1 proslijeđeni tweet 3 korisnika označavaju da im se sviđa
    Poništi
  9. 28. sij

    Ei Wada is a legend

    0:47
    《バーコーダー》バーコードリーダーのスキャン信号をレジではなく、スピーカーに直接接続することで音を鳴らす。 いま渋谷で巨大レシート版、演奏できます┃┃┃┃_ρﾞ
    Prikaži ovu nit
    1 reply 19 proslijeđenih tweetova 73 korisnika označavaju da im se sviđa
    Poništi
  10. 28. sij

    Registration for the W3C ML on the Web workshop opened, it’s in Berlin in March. Come hang!

    13 proslijeđenih tweetova 20 korisnika označava da im se sviđa
    Poništi
  11. 27. sij

    I wonder if we're going to decide to turn off speculative execution in future computer architecture designs when downsides of this class of attack outweigh the performance gains. Anyways, awesome work to this team, this is crazy!

    Finally, the disclosure is over! We present CacheOut, a new speculative execution attack to leak data on Intel CPUs via cache eviction despite current mitigations: , Andrew Kwong, Daniel Genkin and
    7 korisnika označava da im se sviđa
    Poništi
  12. proslijedio/la je Tweet
    24. sij

    Calling all / JS ! Are you using in or ? If not, what’s stopping you? The team wants to hear from you all for our 2020 review! Send feedback below to shape the future of ML in the browser + beyond!

    15 replies 69 proslijeđenih tweetova 126 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet
    14. sij

    You can now run a TensorFlow SavedModel in Node.js directly without conversion - and you might even get a performance boost! 🏎⚡️ Read the blog to learn how ↓

    5 replies 188 proslijeđenih tweetova 476 korisnika označava da im se sviđa
    Poništi
  14. proslijedio/la je Tweet
    6. sij

    📢 Blazeface is now available in browsers with TensorFlow.js! The model detects faces and facial features in real-time 👨👩⏰ Try it out now →

    8 replies 202 proslijeđena tweeta 686 korisnika označava da im se sviđa
    Poništi
  15. proslijedio/la je Tweet
    31. pro 2019.

    Last post of the year, excited to share a post about Anomagram ... Introducing Anomagram — An Interactive Visualization of Autoencoders applied to the Task of Anomaly Detection, Built with .js .

    47 proslijeđenih tweetova 149 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  16. 26. pro 2019.

    This debate is funny to me because everybody things their way is right because it worked for them. Things work differently for different people! Don’t blindly follow one methodology :)

    Oh for fucks sake 🙄. Don’t sacrifice your 20s – or any other decade of life! – on the erroneous belief that unless you work round the clock, you’re not going to be “successful”. The world is full of people who were all work and now are all regret.
    Prikaži ovu nit
    1 proslijeđeni tweet 19 korisnika označava da im se sviđa
    Poništi
  17. 21. pro 2019.

    the link I meant to use is actually

    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  18. 21. pro 2019.

    Overall we’re really excited about this avenue for accelerating ML, more to come soon!

    3 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  19. 21. pro 2019.

    Shoutout to XNNPack and who have made this possible, we call through to this library for the meaty operations.

    1 reply 6 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  20. 21. pro 2019.

    6) This all works in Node.js! This means an ultra lite ML runtime for Node without a TensorFlow binary that runs anywhere Node does (think edge devices).

    1 reply 1 proslijeđeni tweet 7 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi

@nsthorat još nije objavio nijedan Tweet.

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:

    ·