Jeremy Howard

@jeremyphoward

Deep learning researcher & educator. Founder: fast.ai; Faculty: USF & Singularity University; // Previously - CEO: Enlitic; President: Kaggle; CEO Fastmail

San Francisco
যোগদান করেছেন আগস্ট ২০১০

মিডিয়া

  1. ১১ ঘন্টা আগে

    Python friends: I'm trying to wrap an iterator (plus __len__) with map(). But I want to be able to loop through it multiple times. Is there a more concise or idiomatic way to do this?

  2. ১৬ জুলাই

    "An Opinionated Introduction to AutoML and Neural Architecture Search" by

  3. ১৪ জুলাই

    When Tim told me his student project was going to be pretty cool, I still didn't quite expect he meant "nominated for an Emmy" level of cool! :) This is so well-deserved; Tim worked *very* hard on the "Not Hot Dog" project and the result was brilliant.

  4. ৫ জুন

    It's great to simplify complex ideas for a lay audience. But that's not a license to over-claim. When you take a pre-written color inversion tool, provide it as an option to an optimizer, and it gets picked, that's not "learning to invert colors"

    এই থ্রেডটি দেখান
  5. ৩১ মে

    Want to get notified when your fastai training is finished? Here's a little callback that does just that!

  6. ২৪ মে

    Want to sound like a pretentious bore? Well now you can, thanks to this adroit scholarly monograph from the erudite , introducing the "pretentiousness vector".

  7. ২৪ মে

    Fantastic work from student showing how to create a spelling checker based on word vectors. Includes link to full jupyter notebook code walk-thru.

  8. ২১ মে

    One of our students has started on a nice overview of the fastai library

  9. ১৮ মে

    My friend Arshak Navruzyan and I have started commercializing this deep learning labeling idea I demoed a few years ago (see video below). We're looking to raise a seed round to bring our team on full-time and get to revenue. Ping me if interested!

    এই থ্রেডটি দেখান
  10. ১৭ মে

    Oh my. This is fascinating and amazing... A totally practical approach to building a new kind of spell checker, just invented in a forum post! :)

  11. ১৭ মে

    Is this use of the equals sign (in numbered points 1-3) unusual? Incorrect? I can't quite get past the idea that an approximate model "equals" y. What's the best way to write these equations?

  12. ১৬ মে

    If you're looking for an interesting collaborative research project in computer vision, here's one that I think could be really cool

  13. ১৫ মে

    New work with : "Universal Language Model Fine-tuning for Text Classification", with code and pre-trained models. Text classification with ~20% better accuracy, or with ~100x less labeled data

    এই থ্রেডটি দেখান
  14. ১৪ মে

    We now have a general Deep Learning forum that you can use to discuss papers, projects, or anything else - regardless of whether it's related to or not! I hope that this might become a useful resource for all practitioners

    এই থ্রেডটি দেখান
  15. ১১ মে

    “From Idea to Open-Source in 12 days” by Great initiative from a student who has written a handy image annotation web app

  16. ৭ মে

    Apparently the Uber fatality was caused by Uber being crap at building autonomous vehicles. Who could have guessed that would turn out to be the case?

  17. ৭ মে

    As part of the new deep learning course we released today, we've included a section on data ethics, highlighting work from a number of terrific people including and more

  18. ৫ মে

    Our final lesson covered super-resolution - one student decided to try it on satellite imagery, and results are looking pretty good!

  19. ৫ মে

    The National University of Singapore has a course based on our materials, and it looks great! They cover the all 14 lessons of parts 1 & 2, and every student does a project as part of the course. h/t

  20. ১ মে

    Completed the last lesson of the new in-person course last night. Exhausting, but all the kind comments from students make the hard work more than worth it.

লোড হতে বেশ কিছুক্ষণ সময় নিচ্ছে।

টুইটার তার ক্ষমতার বাইরে চলে গেছে বা কোনো সাময়িক সমস্যার সম্মুখীন হয়েছে আবার চেষ্টা করুন বা আরও তথ্যের জন্য টুইটারের স্থিতি দেখুন।

    আপনিও পছন্দ করতে পারেন

    ·