Tweetovi

Blokirali ste korisnika/cu @singularpattern

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

  1. proslijedio/la je Tweet

    I would suggest that Mitsuku is nothing but natural language processing but yes, there's no machine learning in there. Hand written rules means I have total control over its responses and can explain any of them rather than using a black box approach.

    Poništi
  2. proslijedio/la je Tweet
    31. sij

    Why Random Forests can’t predict trends and how to overcome this problem. by

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    30. sij

    One of the best decisions we ever made Applied Deep Learning Research was to standardize on for all our research. It has made us more productive and made our work more fun. Glad to see agrees!

    Poništi
  4. proslijedio/la je Tweet
    28. sij

    I’m surprised to read again and again the same critiques about notebooks but it’s really not that complicated. With students I insist on: 1. Prototype in the notebook but as soon as a chunk of code somehow works, move it to a module. This has two advantages: 1/n

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

    Bonus: Meena often seems to put together ideas in ways that we don't manage to find matches of in the data. For example saying that "Horses go to Hayvard" in conversation we show in the blog post .

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

    I do strongly suspect that RNN-based models are being under-studied at the moment.

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

    Kornia v0.2.0 is out ! We have introduced a new data augmentation module with strong GPU support, extended the set of color conversion algorithms, supporting GPU CI tests with v1.4.0, and much more. Happy coding !

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    26. sij

    USD salary vs Year “Tech favors the young. For people with more than 15 years of experience, there’s practically no correlation between years of experience and income (corr < 0). After 15 years, you either retire, switch to management, or change career.”

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    26. sij
    Odgovor korisnicima
    Poništi
  10. proslijedio/la je Tweet
    25. sij

    "What is the future of operating systems?" Is a great question that comes up quite a bit (like yesterday). "Future of" is always tricky but one thing is always the case in software, which is everything interesting in tech moves "up the stack". Thoughts… 1/

    What is the best computer for you? magazine cover with bill gates.
    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    24. sij

    We're releasing mBART, a new seq2seq multilingual pretraining system for machine translation across 25 languages. It gives significant improvements for document-level translation and low-resource languages. Read our paper to learn more:

    Poništi
  12. proslijedio/la je Tweet

    Dismissing machine learning because it can't make sense of what it *hasn't* seen before it quite short-sighted. It is immensely valuable to be able to automatically recognize *what you are able to label* -- especially on hard pattern recognition problems, at super-human accuracy.

    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet

    Yay. Another 10 year technology contract for the NHS. It’s like we never learn from past mistakes...

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

    “You don’t need Kubernetes” Until... You’re asked to resize all your VMs cause they’re all at 10% utilization... should have used Kubernetes. /1

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    22. sij

    Our paper with & "Spatiotemporal Relationship Reasoning for Pedestrian Intent Prediction" accepted for presentation ! Stanford-TRI Intent Prediction (STIP) dataset & the paper coming out soon! Stay tuned!

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

    Now it works. Let's spatially explore the 2D images! Thanks for tips, ! And Special Kudos to - 3D Ken Burns model is his great work! Here is what happens if you exaggerate this model. And it's amazing.

    Prikaži ovu nit
    Poništi
  18. proslijedio/la je Tweet
    21. sij

    This is a very interesting paper. It shows that a tweaked ResNet50 is about as accurate as EfficientNet-B4 but >3x faster. The EfficientNet paper measured FLOPS, which is a theoretical performance measure, rather than time, which is what actually matters.

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    18. sij

    A thread of classifiers learning a decision rule. Dashed line is optimal boundary. Animations with by and . Logistic regression {stats::glm} with each class having normally distributed features. (1/n)

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    19. sij

    An issue when serving large models is that you need shared memory if one wants to optimize memory against cpu. So load once, serve concurrently, but how to achieve that on containers? Predictive CPU isolation through CFS (Completely Fair Scheduler)

    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:

    ·