jph00

64 ms·

751.
by jph00 9y ago
> Why isn't it practical? Is it because the network would have to be too deep, or too wide? Would the optimizer just get stuck in a local minima or would overfitting be inevitable? Or perhaps some combination of issues? Schmidhuber
752.
by jph00 9y ago
That would be this course. It assumes neither of those things :) (It's for dummies that are prepared to work hard over a 7 week period and that have been coding for at least a year.)
753.
by jph00 9y ago
Yeah there's a discussion of this on http://forums.fast.ai if you want the details. Basically, Andrew's course is bottom-up (i.e. start with the theory and math and simple examples) and works up from there. Ours is top
754.
by jph00 9y ago
It's free. We're a self-funded non-profit. We do this because we actually want to help people!
755.
by jph00 9y ago
My pleasure! You may need to reduce the batch size in a few lessons to fit on that card, FYI.
756.
by jph00 9y ago
There are certainly folks that fly in weekly. One student flew in from Australia for the duration! But if you're far away you'd probably be better off applying to be an international fellow, which means you watch the lesson live o
757.
by jph00 9y ago
If you remember to shut down your machine when you're done, and you do the suggested 10 hours per week, plus maybe an extra 5 hours of model training per week, at $0.45/hour for Paperspace's cheapest machine (still works grea
758.
by jph00 9y ago
Yes, I've provided a step by step walk-through here: http://forums.fast.ai/t/howto-installation-on-windows/10439 This includes setup of fastai, pytorch, cuda90, cudnn, opencv, bcolz, and much more!
759.
by jph00 9y ago
Thanks for sharing! FYI you may find resnet18 or (better still) densenet a good option for that dataset - you shouldn't need so many training images then.
760.
by jph00 9y ago
Thank you for creating that! :) BTW I just had a thought. What if instead of a txt file like you have: hi everybody welcome to practical deep 00:00:00.000 https://www.youtube.com/watch?v=IPBSB1HLNLo#t=00h00m00s
761.
by jph00 9y ago
Just jump in with this course. Your background in web dev will be really helpful, but there will be plenty of new concepts to learn - don't try to understand everything perfectly the first time through, but just try to complete each we
762.
by jph00 9y ago
Here you go: http://course17.fast.ai . (It's also linked from the 2nd paragraph of the new course FYI.) The old forum is still at the same place http://forums.fast.ai/c/part1
763.
by jph00 9y ago
> This is so uncool! Why thank you :) Being uncool is our specialty.
764.
by jph00 9y ago
Yup AWS is fine too. We have a new AMI available if you want to use it, and also a predefined conda environment to add fastai/pytorch to an existing server. Lesson 2 covers AWS usage, and there's also step by step instructions lin
765.
by jph00 9y ago
Definitely well beyond the course. It's designed to be the easiest way to create world-class models. I'll be providing a lot more information on how we're doing this in the next week or two. PS: The focus of fastai is trainin
766.
by jph00 9y ago
Definitely skip to this new course. Sorry I should have provided a link to the FAQ where this and other questions are answered: http://forums.fast.ai/t/part-1-faq/10330 The concepts and skills you learned from the
767.
by jph00 9y ago
We suggest Paperspace now. Lesson 1 walks you through getting set up. It's very easy - there's a fast.ai template there ready to go. They have the best price/performance ratio of anyone at the moment for GPUs.
768.
by jph00 9y ago
Jeremy from fast.ai here. Happy to answer any questions about the course, fastai, or anything else relevant! BTW the 2018 version of the course is being discussed in this forum, for those interested: http://forums.fast.ai/c&
769.
Practical Deep Learning for Coders 2018(fast.ai)
389 points by jph00 9y ago|61 comments
770.
by jph00 9y ago
The Australian law in question requires refunds be made available if you can show that the product was not "fit for the purpose for which they were sold". It is an entirely reasonable expectation for a purchaser to make of a produ
771.
by jph00 9y ago
Whilst there are plenty of things in CV that computers aren't super-human at yet, object classification (given 100+ examples) is not one of them. In datasets with tens of thousands of categories, humans are much worse than computers -
772.
by jph00 9y ago
Second order methods are attracted to saddle points in high dimensional spaces. The math and practice of optimizing these surfaces has a lot of nuances like this so much of the stuff you learn in your convex optimization class doesn't
773.
by jph00 9y ago
FYI this is the paper that lays the key foundation for AlphaZero, which recently got a lot of attention for easily beating the earlier Go-winning algorithms without looking at human games, and then beat the best chess algorithm with 6 hours
774.
Making Peace with Personal Branding(fast.ai)
6 points by jph00 9y ago|0 comments
775.
by jph00 9y ago
Oh BTW the "random" result means randomly initialized vectors. It's still using embeddings, just not pretraining
776.
by jph00 9y ago
I have a version that gets 94.5 on IMDB although not published yet.
777.
by jph00 9y ago
Word embeddings (or subword embeddings) are used for nearly all recent NLP algorithms, both shallow (e.g. FastText) and deep (e.g. Google Neural Translation). Unless you're using a basic bag-of-words approach, you need to translate you
778.
by jph00 9y ago
It's the notes for a 40 min talk Rachel will give at the O'Reilly AI conference. It was originally meant to be a longer tutorial, so the scope had to be cut down significantly, whilst the title remained. There's lots of links
779.
Can Neural Nets Detect Sexual Orientation? A Data Scientist’s Perspective(fast.ai)
4 points by jph00 9y ago|0 comments
780.
Medicine and the need for AI(medium.com)
2 points by jph00 9y ago|0 comments

More ›