jph00

33 ms·

421.
The Jupyter+Git problem is now solved(fast.ai)
301 points by jph00 4y ago|163 comments
422.
Unsilence: Console Interface and Library to remove silent parts of a media file(github.com)
3 points by jph00 4y ago|0 comments
423.
by jph00 4y ago
Frankly it's not something I've been working on -- the GitLab support in v1 was added and maintained by the community. I certainly want it to work (even although I'm not a GitLab user myself), so if you try it and find it doe
424.
by jph00 4y ago
I did run into some problems with nbconvert from time to time. It's worth noting that nbdev doesn't use nbconvert at all - it uses Quarto instead (which AFAICT does everything nbconvert does plus a lot more, and is faster, more ex
425.
by jph00 4y ago
It's not focused on collaboration, but it does add some critical pieces that otherwise make Jupyter development frustrating when working with a team. Specifically: `nbdev_prepare` ensures that diffs are as small as possible, by remov
426.
by jph00 4y ago
A from-scratch rewrite of nbdev was released today -- details on what's new are in the launch post: https://www.fast.ai/2022/07/28/nbdev-v2/ Let me know if you have any questions or comments -- I&#x
427.
Shiny for Python(shiny.rstudio.com)
8 points by jph00 4y ago|0 comments
428.
by jph00 4y ago
> Howard is great with one problem: he kinda hates math I'm sorry what? I run a math study group 4x per week. Right now the book I'm reading during my rest time is a calculus book. I've co-authored a lengthy paper on matri
429.
by jph00 4y ago
No, the book is continually updated for each reprint, but there isn't a separate edition.
430.
by jph00 4y ago
I'd suggesting budgeting about 80 hours for the course given that background. That should get you to a place where you can work on practical projects that are reasonably well within standard applications of deep learning. Most practica
431.
by jph00 4y ago
I'm disappointed that Google shut down the Swift for Tensorflow project, because I do think Swift is a great option for deep learning. In some ways Jax is almost "non-python deep learning" since it's treating Python more
432.
by jph00 4y ago
I'd say it's the other way around - the videos are kinda supplemental to the book. The book has a lot more content, but doesn't have the interactive explanations in the course. Also the book is a couple of years old so is mis
433.
by jph00 4y ago
The good news is that every lesson in this course is actually run on Kaggle Notebooks, which is a free cloud environments including GPUs. So you don't need to set up anything and it runs on any computer with a modern web browser! Mac s
434.
by jph00 4y ago
Honestly I'm not an expert on autonomous driving so I'm not sure I have great insights there. I do know quite a bit about computer vision however so feel qualified to comment on that bit -- I suspect the decision by Tesla to only
435.
by jph00 4y ago
They're designed to be done in order, but yup if you know how SGD works, for instance, you could certainly skip over that bit. The videos all have youtube timestamps, so if you drag the scrollbar you'll see what each section is ab
436.
by jph00 4y ago
Both the Aurélien Géron and François Chollet TF books are absolutely terrific, and everything you learn from them will be extremely useful in becoming a deep learning practitioner, regardless of what framework you end up using. So if you&#x
437.
by jph00 4y ago
Haha yes - it's been wonderful to see how (eventually) the deep learning world has taken to transfer learning!
438.
by jph00 4y ago
Hi folks - I'm the creator/teacher of this course. I'd be happy to answer any questions that you have about the course, learning deep learning in general, or the state of deep learning in 2022.
439.
by jph00 4y ago
sixel works fine, and is better supported than a "new de facto standard" the doesn't even exist yet. It's not at all true that no popular windows s/w supports it - mintty/wsltty supports it just fine and is wid
440.
by jph00 4y ago
The approach of avoiding having to explicitly name variables is known in computer science as "point free" or "tacit" programming, and in math as "combinatory logic". Languages like APL have even less boilerplat
441.
by jph00 4y ago
Thanks you - I stand corrected.
442.
by jph00 4y ago
Perhaps -- but not "definitely". It depends on what the frequency of each of those things is. So far IIUC the frequency of critical projects being removed from pypi is 1 (in the space of a few days - although it came back), and ac
443.
by jph00 4y ago
I keep seeing this claim that open source maintainers shouldn't be upset about a 3rd party mandating we increase our maintenance burden to follow their new rules, because it's all about responsibility. But I also see people claimi
444.
by jph00 4y ago
Well this is unexpected - didn't think my little study group would appear on HN! :O Well anyhoo, all are welcome to join; details in the linked thread (including playlist of last week's videos so you can catch up). As it happens,
445.
by jph00 4y ago
Studio Lab doesn't have any GPU instances available in practice, unfortunately. They're permanently out of resources at the moment. The other two options are pretty expensive.
446.
by jph00 4y ago
PaperSpace has improved a lot in the last few months. I've been using them nearly exclusively recently and have had zero issues. They provide persistent storage and access to the full JupyterLab or Jupyter Notebook software, unlike Col
447.
by jph00 4y ago
You can use GPUs for free on Paperspace Gradient, Google Colab, and Kaggle.
448.
by jph00 4y ago
I don't think that's true - have a look at this analysis here: https://www.kaggle.com/code/jhoward/which-image-models-are-b... Those slow and inaccurate models at the bottom of the graph are the VGG mode
449.
by jph00 4y ago
Using mixed precision training you can do most operations in fp16 and just a few in fp32 where it's needed. This is the norm for NVIDIA GPU training nowadays. For instance using fastai add `.to_fp16()` after your learner call, and that
450.
by jph00 4y ago
You might want to look into Outlook Web Access. It predated gmail by many years and is the reason today we have XMLHttpRequest.

More ›