programnature

5 ms·

1.
ML Framework Design: Reliability, Composition and Flexibility(medium.com)
4 points by programnature 9y ago|0 comments
2.
by programnature 9y ago
Its a good strategy. But its no silver bullet either. If you're exporting to a "static graph" platform, your losing a major benefit of PyTorch. If you mostly just care about shipping to production, a case can be made to just
3.
by programnature 9y ago
Fwiw this is what Soumith has said: "Internally at Facebook, we have a unified strategy. We say PyTorch is used for all of research and Caffe 2 is used for all of production." https://www.oreilly.com/ideas/why
4.
by programnature 9y ago
If you want to run in production. Even FB doesn't use PyTorch in production, and instead uses Caffe2.
5.
by programnature 10y ago
Wolfram was there in 2002. Pretty cool examples though.
6.
by programnature 10y ago
Actually not clear if there is an official affiliation with Facebook, other than some of the primary devs.
7.
PyTorch – Tensors and Dynamic neural networks in Python(pytorch.org)
447 points by programnature 10y ago|88 comments
8.
by programnature 10y ago
How?
9.
by programnature 10y ago
This is a very thick talk, one Rich's best ever IMHO. The first point is how we talk about 'change' in software, to center around what things 'provide' and 'require'. Breaking changes are changes that caus
10.
by programnature 10y ago
Its an interesting question. One reason is Wolfram is extremely talented at language design, which is necessary to build an artifact of this size without self-immolating. Another is that it is a commercial company following a plan. A third
11.
by programnature 10y ago
Yes, understandable. I encourage viewing this as part of the 'open' mandate.
12.
by programnature 10y ago
While its useful to have this kind of info, IMHO its still far from 'infrastructure for deep learning'. What about model versioning? What about deployment environments? We need to address the whole lifecycle, not just the 'tr
13.
by programnature 10y ago
Did you also appreciate slide 28? "I used to think Wolfram was wrong. Now I am not so sure."
14.
by programnature 10y ago
This is extremely similar to DevCards for Clojurescript, https://github.com/bhauman/devcards , https://www.youtube.com/watch?v=G7Z_g2fnEDg
15.
by programnature 11y ago
Theres a 160 people in the Om slack channel. There's more ways to contribute than sending patches, and theres quite a lot of community effort behind Om Next these days.
16.
by programnature 11y ago
If you have a bunch of tiny apps, how do their data models relate? Who runs them, and owns the data? Theres a lot of assumptions baked into the status quo described in the article, that technology has been developed around. Sweet spot for b
17.
by programnature 11y ago
The faster people realize this the better. What does the technology look like for achieving these kinds of goals?
18.
by programnature 11y ago
You are the target audience of this book. The Wolfram Language has gotten huge, making it harder for beginners to find their way around the basics via reference docs. That said, WL is in fact based on pretty simple primitives. If you are fa
19.
by programnature 11y ago
In the announcement they called this the "reference implementation".
20.
by programnature 11y ago
tl;dr: lol. The author is claiming Relay is a scheme to kill the open web. A simpler theory is just that Relay solves a problem Facebook (and many other people it seems) are having. The idea that REST has shortcomings for Facebook's us
21.
by programnature 11y ago
Topological data analysis is amazing, too bad all the hype around DL is leaving it in relative obscurity.
22.
by programnature 11y ago
Clojure/script already has this. Its called transients
23.
by programnature 11y ago
Fukuyama's recent book Political Order and Political Decay goes into some detail on the forest service, from its origins as a model of a modern effective government agency in the Progressive era to one captured by special interests and
24.
by programnature 11y ago
What's the rationale? Who are the target users? I go to the site and I have no idea what exactly this does that's better than the alternatives. It sounds like its trying to do everything.
25.
by programnature 11y ago
at least 2 forces come to mind - 1. Humans want to believe they are special in some way 2. Theory of evolution is under "selective pressure" to explain complexity since the watchmaker analogy
26.
by programnature 11y ago
Interesting that Wolfram predicted this in 2002, was mocked by Cosma Shalizi as having "absolutely no understanding of evolution"
27.
by programnature 11y ago
Definitely a must-read. Also: Why Nations Fail, and Violence and Social Orders. People in tech would be well-served by a deeper understanding of the systems we operate in and hope to change.
28.
ClojureScript 1.7: ClojureScript can compile itself(swannodette.github.io)
536 points by programnature 11y ago|106 comments
29.
Aeron: Open-source high-performance messaging [video](youtube.com)
9 points by programnature 12y ago|1 comments
30.
by programnature 12y ago
Imagine being a woman and coming to HN and seeing this. Clearly a place for her? We need to drop sexist article titles like this from HN.

More ›