Hacker News: skierscott

Visualization of Covid-19 infection rates

skierscott — Sun, 15 Mar 2020 17:19:04 +0000

Article URL: https://stsievert.com/blog/2020/03/14/covid-19/

Comments URL: https://news.ycombinator.com/item?id=22584797

Points: 1

# Comments: 0

New comment by skierscott in "Machine Learning Confronts the Elephant in the Room"

skierscott — Sun, 23 Sep 2018 13:42:28 +0000

> a low confidence score

Neural nets should return a low confidence score. But, the popular approach (described below) ignores that. Neural nets ignore confidence because of a technique called softmax [1].

This happens as the final operation of a neural net, and is required for training.

Softmax is a tool to make an array of positive numbers look like a probability distribution:

    out = x / x.sum()

x[i] is a class prediction, but x.sum() != 1. Say if the network was uncertain, x[cat, dog] = [0.03, 0.01]. These are small values that do not imply great confidence (the network was trained on vectors with out.sum() = 1. The network would predict “dog” using softmax because out[dog] = 0.75 > 0.25 = out[cat].

But then in inference/prediction, the confidence is ignored. What if x.sum() is small? That would imply that the network is uncertain.

[1]: https://en.m.wikipedia.org/wiki/Softmax_function

New comment by skierscott in "Advanced computing with IPython"

skierscott — Tue, 05 Jun 2018 14:09:11 +0000

mpi4py is a solid implementation, but the docs aren’t the greatest, especially if you’re not familiar with MPI.

New comment by skierscott in "Chris, Nat and Satya's Calldeck Regarding Microsoft and GitHub"

skierscott — Mon, 04 Jun 2018 13:27:32 +0000

On iOS this isn’t functional. I can’t advance past the first slide.

New comment by skierscott in "Microsoft acquires Github"

skierscott — Mon, 04 Jun 2018 13:25:37 +0000

>> Going forward, GitHub will remain an open platform, which any developer can plug into and extend.

Does an “open platform” mean “open source” or “a plugin system”? I think making it open source would alleviate a lot of dev concerns.

New comment by skierscott in "DNA of every baby born in California is stored. Who has access to it?"

skierscott — Sun, 13 May 2018 07:45:57 +0000

> Some states destroy the blood spots after a year, 12 states store them for at least 21 years.

What are the other 12 states?

New comment by skierscott in "Show HN: Convert screenshots of equations to LaTeX"

skierscott — Wed, 07 Mar 2018 13:08:04 +0000

My favorite is Alan Turings quote:

> Mathpix's AI definitely passes THIS Turing test!

New comment by skierscott in "A promenade of PyTorch"

skierscott — Tue, 20 Feb 2018 23:14:30 +0000

Isn’t this what JITs do?

New comment by skierscott in "How Apple is paving the way to a ‘cloud dictatorship’ in China"

skierscott — Tue, 20 Feb 2018 03:31:04 +0000

> “If you understand and agree, Apple and GCBD have the right to access your data stored on its servers. This includes permission sharing, exchange, and disclosure of all user data (including content) according to the application of the law.”

> In other words, once the agreement is signed, GCBD — a company solely owned by the state — would get a key that can access all iCloud user data in China, legally.

What user data will this decrypt? Are iMessage and FaceTime still safe?

New comment by skierscott in "Outperforming LAPACK with C++ metaprogramming"

skierscott — Tue, 30 Jan 2018 04:33:18 +0000

There’s also a closed form for the Fibonacci numbers. It runs in O(1).

http://stsievert.com/blog/2015/01/31/the-mysterious-eigenval...

New comment by skierscott in "Project Euler"

skierscott — Mon, 11 Dec 2017 03:51:58 +0000

Not true. Infinite sums are not defined for series that don’t converge, and sum(1..n) does not converge as n => infinity.

Yes, you can play a nice trick with that sum if you ignore the fact that inf-inf is meaningless.

New comment by skierscott in "Joplin – A note-taking and to-do app with builds for desktop, mobile, terminal"

skierscott — Thu, 30 Nov 2017 14:55:50 +0000

Where does Apple Notes store it’s files? What format are they in?

New comment by skierscott in "Terrain rendering in fewer than 20 lines of code"

skierscott — Fri, 24 Nov 2017 19:54:49 +0000

Wow, great demo. Even works on mobile!

New comment by skierscott in "Deep Learning Is Eating Software"

skierscott — Fri, 17 Nov 2017 04:42:13 +0000

> 2% is HUGE at this point

Hell, on MNIST 0.14% is huge. Geff Hinton created an entire new architecture to get 0.25% error, which is far better than the baseline 0.39% error [1].

To be fair, he did reduce the error by 35%.

[1]:https://arxiv.org/pdf/1710.09829.pdf

New comment by skierscott in "Numpy: Plan for dropping Python 2.7 support"

skierscott — Wed, 15 Nov 2017 23:42:57 +0000

> what has changed in Python 3.6 that they're happy to do this now when they were pissed off about it a couple years ago?

Most of mine have to do with developer productovity and I didn’t find until Python 3.5.

– @, the matmul operator

– fstrings (f”{foo}”)

– parameter typing (foo: int = 0) which has exciting work with Cython

– other async features. I forget what library it was (tensorflow?) but for Python 3 it had better async support

TensorFlow meets PyTorch with Eager execution

skierscott — Tue, 31 Oct 2017 09:52:42 +0000

Article URL: https://medium.com/@yaroslavvb/tensorflow-meets-pytorch-with-eager-mode-714cce161e6c

Comments URL: https://news.ycombinator.com/item?id=15591987

Points: 3

# Comments: 0

New comment by skierscott in "Anaconda 5.0 Released"

skierscott — Fri, 27 Oct 2017 01:01:49 +0000

This adds support for new compilers.

This makes Dask (pure Python paralleization) run 20% faster https://twitter.com/mrocklin/status/923581208482721794 according to Matthew Rocklin, the main developer

New comment by skierscott in "Boiling lead and black art: a history of mathematical typography"

skierscott — Sun, 15 Oct 2017 15:22:06 +0000

...I read the article in Safaris Reader mode. The font is too big for mobile.

But this is a great article.

New comment by skierscott in "Diary of a Concussion"

skierscott — Wed, 11 Oct 2017 22:40:33 +0000

> Was this was the anxiety that had been mentioned in the medical literature?

I don’t know. I’ve had a difference expierence with brain injury and anxiety. My recovery resulted in classic anxiety; I experienced social anxiety and depression.

New comment by skierscott in "Core ML Community Tools"

skierscott — Thu, 05 Oct 2017 15:14:59 +0000

Turns out generating a ML model is pretty easy too, even after training in Python. There's a simple 2 or 3 line conversion from keras/sklearn to an ML model.

http://stsievert.com/blog/2017/06/11/coreml/

I verified this because I'm a graduate student studying ML and I'll likely use this.