iskander

44 ms·

421.
by iskander 12y ago
I'm really curious to find out in what situations Spark actually works for people. So far, no one in my lab seems to be having a terribly productive time using it. Maybe it's better for simple numerical computations? How large are
422.
by iskander 12y ago
>The queries I run against spark are billions of events and results are sub-second. This is interesting, I haven't gotten Spark to do anything at all in less than a second. How big is this dataset (what does each event consist of)?
423.
Spark should be better than MapReduce (if only it worked)(blog.explainmydata.com)
3 points by iskander 12y ago|0 comments
424.
Where does Scala look for implicits?(stackoverflow.com)
1 points by iskander 12y ago|0 comments
425.
by iskander 12y ago
Strange to read this while I'm at a conference on cancer therapies.
426.
by iskander 12y ago
Immune therapy for cancer is very exciting right now: there are lots of promising trials in the pipeline and cool papers coming out every month.
427.
by iskander 13y ago
This is surprisingly detailed design rationale which weighs many alternatives and gives careful consideration to possible tradeoffs. Great job by Nathaniel Smith and the numerical Python community.
428.
by iskander 13y ago
>Are there any similar devices that have been installed in people that have been more successful? When I was following the retinal implant research several years ago, you'd typically give the patient a few dozen pixels of vision and
429.
by iskander 13y ago
>And because you can now do everything the wireless way Citation please? That certainly wasn't the case last time I was reading this corner of the research literature (6ish years ago).
430.
by iskander 13y ago
I think the tone of this submission is below the typical community standards of Hacker News.
431.
Workshop on array programming (NumPy, Julia, Repa, &c)(sable.mcgill.ca)
1 points by iskander 13y ago|0 comments
432.
by iskander 13y ago
I was a lowly software engineering intern at Google and spent most of my summer writing MapReduce jobs which would traverse (1) every document in a large index of the web or (2) every book Google has scanned. Even worse, I sometimes needed
433.
by iskander 13y ago
I just started working on a project related to immuno-oncology and I think this field is (1) really fucking fascinating and (2) rapidly heating up. Two decades ago, the centrality of our immune systems in the prevention and/or evolutio
434.
by iskander 13y ago
CAR therapy requires identifying a broadly expressed target; it looks like there are a few candidate antigens being evaluated for MM but I don't know if any of the studies have published significant results yet. http://www.n
435.
Big speedup for Random Forest learning in scikit-learn 0.15(blog.explainmydata.com)
1 points by iskander 13y ago|0 comments
436.
by iskander 13y ago
> The thing achieves performance everybody else dreams of. Citation needed? In experience (implementing simpler ML algorithms across Q, Python, and Matlab), the performance of non-trivial programs (using multiple operators together) is m
437.
by iskander 13y ago
My experience with Q was somewhat less inspiring than yours. It has some powerful core primitives but is generally poorly designed and terribly implemented. Single letter function names (and error codes) don't actually make a language
438.
by iskander 13y ago
I just started working at a research hospital. One problem I've heard mentioned from several people is the difficulty of getting black patients to participate in studies or even consent to having their biological samples used for resea
439.
Libraries, Languages, and Compilers for Array Programming(sable.mcgill.ca)
1 points by iskander 13y ago|0 comments
440.
NIPS and the Zuckerberg Visit(blog.explainmydata.com)
2 points by iskander 13y ago|0 comments
441.
by iskander 13y ago
There's definitely a shortage of science research jobs in fields like biology and physics. On the other hand, having exposure to math & empirical thinking makes you ridiculously employable outside your original field. This seems to
442.
by iskander 13y ago
They're a very simple abstract interface that happens to capture a surprisingly large class of data structures and computations. You need to implement two operations to make a monad: (1) take a normal value and put it in the monad (2)
443.
Marshall, TX: Patent suits here appear to be on the rise again(marshallnewsmessenger.com)
1 points by iskander 13y ago|0 comments
444.
by iskander 13y ago
Python -> Scala = PyPy?
445.
by iskander 13y ago
Timings on my machine (2.67ghz Xeon W3520): $time ./c_compiled_with_O2 1000000 real 0m0.362s user 0m0.360s $time ./c_compiled_with_O3 1000000 real 0m0.189s user 0m0.184s I also tried dropping all the
446.
by iskander 13y ago
Thanks :-) I'm going to finish grad school soon and the job I'm going to will be pretty rich with data analysis, hopefully will lead to more blog posts.
447.
by iskander 13y ago
I think there's a big difference between explaining the essence of an algorithm and understanding all the details and decisions that go into a particular implementation. If explaining SVM, I'd stick with the linear primal formulat
448.
by iskander 13y ago
SVM prediction is pretty straight-forward, even if you're using a kernel/dual formulation: "Here are some training points from classes A and B, each with an importance weight. Compute the weighted average of your new point&#x
449.
Python on the GPU with Parakeet(vimeo.com)
1 points by iskander 13y ago|0 comments
450.
by iskander 13y ago
...but their higher quality standards are also what has allowed them to carve out a niche which youtube can never replace.

More ›