cf

54 ms·

211.
by cf 14y ago
Except many of the things he is talking about don't require lots of money. The company just needs to give a shit and treat people like people.
212.
Voice-controlled AR Drone(lemondronor.com)
3 points by cf 14y ago|0 comments
213.
by cf 14y ago
Well they have Thad Starner http://www.cc.gatech.edu/~thad/ who has been doing wearable computing nearly as long.
214.
by cf 14y ago
You can subscribe with http://zinkov.com/rss.xml . I get this question a lot, what's the standard way this is solved?
215.
AAAI 2012 talk: How to Grow a Mind(videolectures.net)
1 points by cf 14y ago|0 comments
216.
by cf 14y ago
The trouble with probabilistic programming is it requires very unnatural control flow to run inference on the models. Now there any solutions like Factorie, Figaro and Infer.NET where they are just a library you call into your code. The tro
217.
by cf 14y ago
Anyway, I wrote this post as a precursor to a more substantial set of writings explaining how to practically use these upcoming languages. Most of what I am trying to show here is that while we use specialized algorithms for most of our mac
218.
Minireviews for popular Machine Learning books(zinkov.com)
1 points by cf 14y ago|0 comments
219.
by cf 14y ago
They are for different usecases. Storm is more for realtime data like analyzing twitter or bidding on ads. Spark is very much a in-memory map-reduce and designed for batch computations. Spark makes sense when you just have a lot of data you
220.
by cf 14y ago
That's why they include which features they used, which is educational.
221.
by cf 14y ago
The issue is that news in its raw form lack context. You just read the event and only have the news you immediately read before it for context. Most information is better understood from a larger window. Mainstream news does not bother to e
222.
Inferno Part 2: Let's Make a Cluster(debu.gs)
2 points by cf 14y ago|0 comments
223.
by cf 14y ago
Largely, what I need is integration with JAGS http://mcmc-jags.sourceforge.net/ which is a niche that pymc fills. I am working on a PR to get in the other stuff I need. I bring up R more as a point that there is more to a library than sup
224.
by cf 14y ago
Since it isn't likely going to be said otherwise, I really think this is the best library for doing Machine Learning out there right now. This assessment isn't based on breadth of algorithms supported, since R beats it here. It has nothing
225.
by cf 14y ago
The technique I like to do is read the foreign language Wikipedia. I try to pick articles that exist in both languages. It helps to know the subject a bit so you aren't constantly flipping back and forth.
226.
by cf 15y ago
Could you change the title to mention that its Kaggle? The company has been featured on Hacker News.
227.
by cf 15y ago
Yes, that is how you make fast libraries in Python. But, nltk isn't written using C extension modules. All of this NLP is done in pure python. You could rewrite what needs to be fast with C extensions, but then what's the point of using nlt
228.
by cf 15y ago
Anna Karenina is the better book.
229.
Two miami kids make millions of dollars being international arms dealers(rollingstone.com)
7 points by cf 16y ago|0 comments
230.
by cf 16y ago
Los Angeles seems like a glaring omission.
231.
Datasets available from Yahoo(webscope.sandbox.yahoo.com)
2 points by cf 16y ago|0 comments
232.
by cf 16y ago
The closest approximation to this is how many people are watching a project. Sometimes when I can't figure out which fork is the main one, I use this.
233.
Scientific Computing in the cloud(opani.com)
3 points by cf 16y ago|2 comments
234.
Million Song Dataset of audio metadata(labrosa.ee.columbia.edu)
41 points by cf 16y ago|4 comments
235.
How to Read a Paper(zinkov.com)
6 points by cf 16y ago|0 comments
236.
by cf 16y ago
Yea, kayak doesn't do that at all.
237.
by cf 17y ago
No, http://en.wikipedia.org/wiki/Ion_(window_manager) is the Father of modern tiling WMs
238.
by cf 17y ago
Nonnegative matrix factorization. It was the advantages of keeping a sparse matrix sparse, can be iteratively created and generally the bellkor team's matrix factorization of choice.
239.
by cf 17y ago
Don't most people use NNMF instead these days?
240.
by cf 17y ago
But wouldn't they also know when someone that didn't know was in the house?

More ›