Tweetovi

Blokirali ste korisnika/cu @bfarzin

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

  1. proslijedio/la je Tweet
    27. ruj 2019.

    Very interesting poll! Not many uni students I speak to who are trying to get that vital first job spend time on gh/kaggle. And few HR departments mention it in their job ads.

    Poništi
  2. proslijedio/la je Tweet
    24. ruj 2019.

    Just for the record, the quid pro quo isn’t necessary to substantiate impeachment. Simply using the power of the presidency to pressure a foreign government to investigate a political rival is egregious enough to warrant the immediate launch of impeachment proceedings.

    Poništi
  3. proslijedio/la je Tweet
    22. ruj 2019.

    If I switched my working days to Sunday through Thursday I think I’d get more done. - No one’s working at full capacity on Friday anyway - Lots of weekend trips are easier to start on Thursdays than Fridays - One workday when everyone else is off guarantees less interruptions

    Poništi
  4. proslijedio/la je Tweet
    5. ruj 2019.

    This is SO meta 🤓 We trained a generative language model on a dataset of ArXiv NLP papers. You can now get a neural net to write your papers for (with?) you 🔥. We heard from a few researchers that they're already using it in submitted papers.

    , , i još njih 6
    Poništi
  5. proslijedio/la je Tweet
    14. kol 2019.

    I highly recommend following 's ongoing series of deep learning posts - they are an excellent overview of the key issues for deep learning practitioners. (The fact that he's only 16 years old is icing on the cake! ;) )

    Poništi
  6. proslijedio/la je Tweet
    14. kol 2019.
    Odgovor korisniku/ci

    Thanks for the kind words 🙏 and made it super easy to connect it all together

    Poništi
  7. Poništi
  8. proslijedio/la je Tweet

    Wes is too humble to say it explicitly, so I will. The success of pandas (his creation) is the biggest reason why Python has become a key tool for data scientists and engineers, and in turn why it underlies so many data-driven applications. It's kind of a BFD.

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    28. srp 2019.

    Completely and passionately suspicious of zealotry in all of its incarnations. Cannabis, sports, your dog, politics—doesn’t matter. If you choose to substitute a singular personal interest for your identity, or as an organizing structure for your worldview, I wonder about you.

    Poništi
  10. proslijedio/la je Tweet

    Wicklow AI in Medicine Research Initiative (WAMRI) chaired by Lightning Talks highlighting our work with from 2018-19 now posted! Check out the videos:

    Poništi
  11. 24. srp 2019.
    Poništi
  12. proslijedio/la je Tweet
    20. srp 2019.

    This is a powerful abstraction for image data augmentation that generalizes MixUp, Cutout, CutMix, RICAP, and more, and allows annealing of the amount of augmentation during training (to support curriculum learning)

    Poništi
  13. proslijedio/la je Tweet
    16. srp 2019.

    🥁🥁🥁 Welcome to "pytorch-transformers", the 👾 library for Natural Language Processing!

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    16. srp 2019.

    🏆 Outstanding Contribution in AI for Best Course goes to , for introducing free courses for coders, software library & cutting-edge research, you can get involved here: 📚Congratulations to , and team!

    Poništi
  15. proslijedio/la je Tweet
    10. srp 2019.

    🏆 Outstanding Contribution in AI for PhD Thesis goes to Sebastian Ruder for his work in - you can read more about it here: Congratulations to , we can’t wait to see what you do next!

    Poništi
  16. proslijedio/la je Tweet
    8. srp 2019.

    Excited to share our newest course: A Code-First Introduction to Natural Language Processing All code & videos are available for free online, please check it out!

    Prikaži ovu nit
    Poništi
  17. 3. srp 2019.

    A new blog post about Label Smoothing as a Regularizer. I have had a lot of luck using it for problems where the labels are noise (which would be all problems!)

    Poništi
  18. 3. srp 2019.

    So inspiring. I dropped off a cliff myself at 40. IF has worked well for me now and it is time to get more intensity in the lifting and cardio!

    Poništi
  19. 28. lip 2019.

    I recently participated in the HAHA 2019 Challenge and came in 3rd easily surpassing the NBSVM baseline. I wrote a blog about how I did it (along with the code I used for my competition entry):

    Poništi
  20. proslijedio/la je Tweet
    11. lip 2019.

    Yesterday the HAHA 2019 competition ended. Here are the results. There were 18 participants and 226 submissions in total. Crazy stuff!! Congrats to the winner, adilism, with an F1 score of 82.1%, the runner-up with 81.6% and which is the 3rd with 81.0%.

    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:

    ·