Radim

67 ms·

751.
by Radim 12y ago
I don't know what these acronyms mean, but I'd hate NOT being assigned full intellectual property rights, as a client hiring a freelancer. Is that really common in your field? What's your field's price tag between doing
752.
Streaming large files from/to S3 [Python](github.com)
1 points by Radim 12y ago|0 comments
753.
by Radim 12y ago
...and I have no plans to add NLP tools in gensim. The connection between gensim and tokenizing/tagging/parsing libs is intentionally loose and flexible. I'm a fan of "do one thing, do it well". Having said that, it
754.
by Radim 12y ago
You should be asking questions about your tax domicile, double taxation treaties, your personal legal status w.r.t. to this company, cost of extracting personal profits from this company, can you get a bank account as non-resident, VAT rule
755.
The Cranfield tests(blog.codalism.com)
6 points by Radim 12y ago|0 comments
756.
Making sense of word2vec (with code)(radimrehurek.com)
2 points by Radim 12y ago|0 comments
757.
by Radim 12y ago
On that exact note -- I can recommend "Whipping star" , a sci-fi novel by Frank Herbert (of Dune fame).
758.
by Radim 12y ago
The top EU folk get a massive monthly budget for assistants and consultants = expert help (for EU MPs, this "extra staff budget" is to the tune of 30k USD per month [1], though often abused to fund family members/friends, j
759.
by Radim 12y ago
In their recent KDD article [1], Etsy used 200 SVD dimensions. For those interested in trying this out in Python: * `gensim` contains stochastic SVD for large data (fast online model training) [2] * I wrote a benchmark of (approximate) near
760.
by Radim 12y ago
Mandatory link to the "What colour are your bits?" article: http://ansuz.sooke.bc.ca/entry/23 (law vs. computer science perspective on information; go read if you haven't yet!)
761.
by Radim 12y ago
That part I agree on. I guess my point is that pragmatic progress in science is hardly ever as earth-shattering and all-changing as the academic circles flaunt it to be (understandably so, in these "publish or perish" times). &quo
762.
by Radim 12y ago
Now you're just being silly. Yes, other models may perform better, but Naive Bayes is a simple model that performs quite well, especially in the fields mentioned. The statement is completely uncontroversial.
763.
by Radim 12y ago
A sad read on many levels. "I investigated a little. As it turns out, this particular manufacturer first copies the iOS file system onto the Windows partition that their software is running on, and then pulls the timestamp information
764.
by Radim 12y ago
Not to rain on the parade, but such glorification of "heroes" is... just awkward. Wasn't there an article on Herbert's Dune on the HN front page, just yesterday? In particular, von Neumann was an "amazing human&qu
765.
by Radim 12y ago
This is much better advice than the grandparent. Indeed, if you found Dune too difficult and political and expansive, don't go for God Emperor of Dune . But if you do like that sort of stuff, someone telling you to stop at Dune wo
766.
by Radim 12y ago
How do you feel about LGPL?
767.
by Radim 12y ago
Indeed. ElasticSearch is super useful, but its docs used to be of the frustrating variety: "Explain minute API details, as if a passing note from the lead dev to himself, assuming all context and concepts are understood and obvious. D
768.
by Radim 12y ago
On a related note, Herbert's "God Emperor of Dune" is an amazing book. Best in the series, in my opinion. Pretty much every scene/dialogue in it is a worthwhile quote: on leadership, duty, love, humanity... Check out som
769.
Efficient Data Streaming & API Bondage in Python(radimrehurek.com)
1 points by Radim 13y ago|0 comments
770.
by Radim 13y ago
Related: benchmarks of available k-NN libs using cosine similarity, in Python. http://radimrehurek.com/2013/12/performance-shootout-of-near...
771.
Show HN: Demo of word2vec on GoogleNews(radimrehurek.com)
3 points by Radim 13y ago|0 comments
772.
by Radim 13y ago
:-) Hopefully the SciPy & BLAS dependencies will only get easier to install from now on... Continuum Analytics received shit loads of money and some of it is going towards better scientific Python packaging, I believe.
773.
by Radim 13y ago
Liberating approach Daniel! I'm still in the naive do-it-well phase, but seeing the downvotes, it may be time to join the hipsters. Or at least shut up ;)
774.
by Radim 13y ago
If it's obvious to you, then you're not the target audience of my disclaimer :) But HN responses to posts like these overwhelmingly suggest it's far from obvious. So, it is a fun and satisfying exercise. I agree. Perha
775.
by Radim 13y ago
Great point. Often overlooked by people who only know what I call "drive-by machine learning" (finished an online ML course or something). There's a multitude of problems with real-world texts that a robust guesser must deal
776.
by Radim 13y ago
"We need a revolution instead of a technology evolution." I think we need a system that pays more attention to the reality and results, and less to ideologies and ideas. And whether that improved process is based on techn
777.
Money, startups, fame(radimrehurek.com)
2 points by Radim 13y ago|0 comments
778.
by Radim 13y ago
Now, that concept I'm familiar with. In fact, I think that's a rather common feeling, not even exclusive to chess or pros. Intuition (whatever that is) plays a great role, especially at the top level.
779.
by Radim 13y ago
Occult drivel. Where on earth did you read that about Magnus Carlsen? To say good chess players (not to say grandmasters, let alone world champions) don't calculate moves borders on insulting.
780.
by Radim 13y ago
Sacrifice personal sense of comfort and shame, to get better results faster? ;) I've already trained my brain to filter buzzwords like "hustling" or "hacking" or "disrupt" etc, so this doesn't bother

More ›