Tweetovi

Blokirali ste korisnika/cu @georgecushen

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

  1. proslijedio/la je Tweet
    27. sij

    Handshake launches in 7days! Surprise! Get ready to save the internet :D

    Poništi
  2. proslijedio/la je Tweet

    The two best performing public stocks of the decade - Netflix (+3700%) and Domino's Pizza (+3000%) - perfectly epitomize the 2010s. You either build the world's most advanced machine learning content recommender system, or make a better pizza sauce, there's no middle ground.

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet

    .org 5.0 is out. No additions to core, but vocabulary for classification of topics for more real estate properties in development. New releases on 1st of month, changes in "Pending" section h/t

    Poništi
  4. proslijedio/la je Tweet
    25. stu 2019.

    On the just-granted Google patent, "Structured user graph to support querying and predictions" > User-Specific Knowledge Graphs to Support Queries and Predictions

    Poništi
  5. proslijedio/la je Tweet
    24. lis 2019.

    The new study by et al. provides a great overview of best practices in the current transfer learning landscape in NLP. Check out page 33 of the paper or below for the main takeaways.

    Poništi
  6. proslijedio/la je Tweet

    of Farfetch gives an overview of applying to personalising recommendations on fashion platforms

    Poništi
  7. proslijedio/la je Tweet
    22. lis 2019.

    Join us tonight at the Meetup! will be talking about Knowledge Graphs and AI to Hyper-Personalise the Fashion Retail Experience at Farfetch! AND who will covering Transforming Legal Recruitment with a Market Knowledge Graph!

    Poništi
  8. proslijedio/la je Tweet
    21. lis 2019.

    New tutorial!🚀 vs. tf.keras: What's the difference in 2.0?🤔Covers: - History of Keras & TF - Eager execution - Model subclassing - Best practices, tips, & suggestions Full tutorial w/ code here: 👍

    Poništi
  9. proslijedio/la je Tweet
    16. lis 2019.

    Need a universal fashion taxonomy? has you covered. (During this session I also learned that a down jacket is called a "puffer coat" in the UK:) 👕👖👜👠

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    16. lis 2019.

    Can't believe is nearly over (boo!) However, I'm thoroughly enjoying the chaps - a deep dive into using taxonomies and knowledge graphs to find and sell fashion items

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    16. lis 2019.

    Improving the search experience at by employing a knowledge graph, as described by at

    Poništi
  12. 15. lis 2019.

    Wow! Fantastic achievement for Academic Website Builder to have been recognized on 's Release Radar 🚀🎉 Let's celebrate all the effort from 87 contributors 👩‍💻 that has gone into making it the #1 theme/builder for 🙌

    Poništi
  13. proslijedio/la je Tweet
    Poništi
  14. proslijedio/la je Tweet
    8. lis 2019.

    This is a nice diagram by Zhengyan Zhang and that shows how many recent pretrained language models are connected. The GitHub repo contains a full list of relevant papers:

    Poništi
  15. proslijedio/la je Tweet

    🎉 The wait is over! TensorFlow 2.0 is finally here. Driven by community feedback, this release provides a complete set of tools for developers, enterprises, and researchers to easily build ML applications. Read the blog ↓

    Poništi
  16. proslijedio/la je Tweet
    28. ruj 2019.

    Steam is down so I've been talking to my wife. She seems nice.

    Poništi
  17. 28. ruj 2019.
    Poništi
  18. proslijedio/la je Tweet

    Federated learning: train machine learning models while preserving user privacy, by keeping user data on device (e.g. mobile phone) and only sending encrypted gradient updates (that can only be decrypted in aggregate) back to the server

    Poništi
  19. proslijedio/la je Tweet
    27. ruj 2019.

    Totally crazy that you can squeeze BERT down to 7 MB without much drop in performance - "Extreme Language Model Compression with Optimal Subwords and Shared Projections," Zhao et al.:

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet

    If your classifier is "99% accurate", either you're using the wrong metric (a metric this high is not informative), or you have an overfitting or leakage problem. Metrics are feedback points on the way towards better models. Not trophies to show off. They should be actionable.

    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:

    ·