Tweetovi

Blokirali ste korisnika/cu @woc_sajit

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

  1. Prikvačeni tweet
    31. kol 2019.

    METHOD ACTING: Acting is very real, more real than the clump of habits you have come to recognize as your personality. Acting requires you to be conscious every moment, lest you stop acting and start "habiting"..

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

    "We achieved an overall accuracy of 94.5%, more than 4.5% of an increase on the previous state-of-the-art"; classifying the patterns in 18,577 Scanning Electron Microscope images

    Poništi
  3. proslijedio/la je Tweet
    26. sij

    Teaching Deep Unsupervised Learning (2nd edition) at this semester. You can follow along here: Instructor Team: , , , Wilson Yan, Alex Li, YouTube, PDF, and Google Slides for ease of re-use

    Poništi
  4. proslijedio/la je Tweet
    3. velj

    This repo is full of amazing awesomeness. I don't know of anything else like it. Independent refactored carefully tested implementations of modern CNNs

    Poništi
  5. proslijedio/la je Tweet
    31. sij

    Through this great story from James Briggs, a computational physicist turned data scientist, I came across his totally amazing deep learning notes, which are some of the best I've seen:

    Poništi
  6. proslijedio/la je Tweet
    30. sij

    Recently, I am working on moving machine learning models to production. So I made a detailed guide on how to do this properly! Instead of a minimal example, I demonstrate best practices with state of the art tools. Enjoy!

    Poništi
  7. 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
  8. proslijedio/la je Tweet
    26. sij

    Interesting mini thread on a possibly under-used approach for training models

    Poništi
  9. proslijedio/la je Tweet
    21. sij

    I wrote a blog post about the machine reading comprehension demo I recently shared. Tutorial notebook at the end if you want to train your own!

    Poništi
  10. proslijedio/la je Tweet
    22. sij

    Brought to you by our very own Mathijs, Verna, Elia and !

    Poništi
  11. proslijedio/la je Tweet

    Help me Twitter. Best (reasonably technical) math/science podcasts?

    Poništi
  12. proslijedio/la je Tweet
    19. sij
    Poništi
  13. proslijedio/la je Tweet

    Great intro to the modern landscape of Deep Learning & by . Including sweet mentions of the models inside transformers, write with transformers & 's NLP progress repo!

    Poništi
  14. proslijedio/la je Tweet
    16. sij
    Poništi
  15. proslijedio/la je Tweet
    18. sij

    "During 2018 I achieved a Kaggle Master badge (...). Very often I found myself re-using most of the old pipelines over and over again. At some point it crystallized into this repository." PyTorch extensions for fast R&D prototyping and Kaggle farming

    Poništi
  16. proslijedio/la je Tweet
    17. sij

    Distributing your workload across multiple with BlazingSQL is easy & everything you need is already installed by default! Simply pass in the Dask Client of your LocalCUDACluster into BlazingContext upon initialization and you're good to go!

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

    Tried mixed precision yet? Took 10 min to set up and my model runs almost 2x faster with same results. Vars and grads are still 32 bits so it usually doesn't affect predictive performance. E.g. in TF2, set option and make all input to your layers float16 (data, RNN states, ..):

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

    Current image datasets are getting very large and pre-training on them is time consuming. We are releasing Neural Data Server (NDS), a search engine for transfer learning data! Webservice: Paper:

    Poništi
  19. proslijedio/la je Tweet
    11. sij

    “Meet AdaMod: a new deep learning optimizer with memory” by Less Wright

    Poništi
  20. proslijedio/la je Tweet
    17. sij

    Deep learning achieves perfect anomaly detection on 108,308 retinal images including unlearned diseases by Ayaka Suzuki et al.

    Poništi
  21. proslijedio/la je Tweet
    18. sij
    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:

    ·