Rezultati pretraživanja
  1. 4. stu 2019.

    Just implemented in a batch size finder from OpenAI and wrote an article on Medium ! Thank you and for the great work, and bottom line : bs matters, having a bs of 512 instead of 64 can lead to a 4x speedup on Rossmann !

    5 replies 57 proslijeđenih tweetova 233 korisnika označavaju da im se sviđa
  2. 17. svi 2019.

    Introducing our new library FastBert. It will helps developers and data scientists develop and deploy BERT based model. Thank you and

    285 proslijeđenih tweetova 899 korisnika označava da im se sviđa
  3. 29. ožu 2019.

    Discover ULMFit in library & be inspired by a super effective transfer learning method for any NLP task...yes! and without having to train models from scratch! Read more at Brilliant work from the duo of and

    1 reply 23 proslijeđena tweeta 60 korisnika označava da im se sviđa
  4. 19. lip 2019.

    I had never thought that big O notation would one day help me classify dog images faster. A reminder that the order of matrix multiplication can matter a lot! A small change reduces run time by 1/3.

    17 proslijeđenih tweetova 84 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
  5. 12. pro 2018.

    Today was the last class of Deep Learning for Coders at . There’s so much we’ve covered in the last 2 months and it’s been incredible being part of the community. Thanks & for making ML so enjoyable, accessible and practical 🙏

    13 proslijeđenih tweetova 74 korisnika označavaju da im se sviđa
  6. 17. lis 2019.

    Day 1 Watched Lecture 5 of Practical Deep Learning for Coders, v3 course involving Back propagation; Accelerated SGD; Neural net from scratch.

    1 reply 8 proslijeđenih tweetova 11 korisnika označava da im se sviđa
  7. 28. lis 2019.

    "What's growing there?" - a blog post we stumbled across by chance, noticing used and our to build crop classification in . These kind of stories clearly demonstrate the power of in !

    1 reply 37 proslijeđenih tweetova 122 korisnika označavaju da im se sviđa
  8. 3. velj

    On the road again. This time Cali for and – mjesto: Boston Logan International Airport (BOS)

    2 korisnika označavaju da im se sviđa
  9. 3. velj

    1 new repository created in - 1 fastpages

    1 proslijeđeni tweet
  10. 31. sij

    Computational Linear Algebra for Coders: This course was taught in the University of San Francisco's, Masters of Science in Analytics program, (free) More training ->Wiki->Education at:

  11. 31. sij

    This is how CS education should be: accessible, exciting and grounded in the real world + yes ethics as key component of pretty much any AI/DL/CS course...

    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:
    1 reply 10 korisnika označava da im se sviđa
  12. 31. sij

    Deep Learning course by is just amazing!

    4 korisnika označavaju da im se sviđa
  13. 29. sij

    Alright after hours of struggle I managed to integrate angular 7 with and . On the backend I created an App with . Works pretty well together. Next I'm gonna model basic game entities and processing.

    9 proslijeđenih tweetova 17 korisnika označava da im se sviđa
  14. 23. pro 2019.

    The new blog post is out! 🤖📚 👉 In collaboration with A gentle introduction to ULMFit from (thanks & !) 💻🔥Google colab available Have data fun!

    24 proslijeđena tweeta 72 korisnika označavaju da im se sviđa
  15. 31. lis 2019.

    Wrote casually how to use optuna, a hyperparam tuner with for a better tabular model on Rossman dataset! Acknowledgement: for fabulous library and lecture. for

    I just published Get Better fastai Tabular Model with Optuna
    1 reply 12 proslijeđenih tweetova 67 korisnika označava da im se sviđa
  16. 7. lis 2019.

    In addition to the already tagged folks, huge props to for developing an excellent algorithm and teaching us all and for supporting and supplying open satellite imagery

    Can you apply ML on satellite imagery at scale ? Yes. Can you do it fast ? For sure —> talking about `Rapid Earth AI` cc
    3 proslijeđena tweeta 22 korisnika označavaju da im se sviđa
  17. 27. ruj 2019.

    Thanks for reaching out!😁✨ and created over a dozen phenomenal machine learning notebooks: And for learning the language, check out 's rad book, "Swift in Depth":

    Odgovor korisniku/ci @DynamicWebPaige
    Hi Paige, are there any resources you would recommend to learn Swift? Want to slowly move towards S4TF. Thanks! 🙂
    16 proslijeđenih tweetova 58 korisnika označava da im se sviđa
  18. 18. ruj 2019.

    Why it's essential to see what is detecting with saliency maps. Correct classification of IVC filter but totally missed the filter!

    1 reply 12 proslijeđenih tweetova 29 korisnika označava da im se sviđa
  19. 13. ruj 2019.

    You can do machine learning. No, really, you can. If you're at the team will show how ML can help your investigations • Today! 11 am at the InterContinental.

    8 proslijeđenih tweetova 14 korisnika označava da im se sviđa
  20. 13. kol 2019.

    My first blog post on Deep Learning - Text Classification in Polish with and in Google Colab.

    6 proslijeđenih tweetova 36 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit

Nema rezultata.

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.