synparb

12 ms·

61.
PyData 2012 talks: videos(vimeo.com)
4 points by synparb 14y ago|0 comments
62.
Obama-Romney Facial Recognition(continuum.io)
3 points by synparb 14y ago|0 comments
63.
by synparb 14y ago
I'm coming from the scientific computing side of python, so beyond the BDFL's suggestions, I use the following protocol to make python fast. Note, that I'm doing numerical analysis and manipulations, so this isn't a general workflow for all
64.
Why doctest.js is better than Python’s doctest(blog.ianbicking.org)
11 points by synparb 14y ago|1 comments
65.
We Need an Open Economist(gist.io)
2 points by synparb 14y ago|0 comments
66.
John Hunter 1968-2012: Reflections from Travis Oliphant(technicaldiscovery.blogspot.com)
1 points by synparb 14y ago|0 comments
67.
by synparb 14y ago
One of the top givers is currently workforpie, which is a company, so it is already happening.
68.
by synparb 14y ago
There is an issue for this already: https://github.com/whit537/www.gittip.com/issues/221 In general Chad is very receptive to suggestions and discussions and there is quite an amazing log of ideas surrounding Gittip in the github issues a
69.
by synparb 14y ago
I don't know how much Armin drinks, but he's already pulling down $50 a week which seems to be a pretty healthy Friday beer fund.
70.
by synparb 14y ago
"Now if you look at the numbers it's not doing super amazing currently but I believe that's because not enough people back it at the moment and for a while that was mostly because getting money out of there was not yet easy enough." The iss
71.
With payouts, Gittip is minimally viable(blog.gittip.com)
14 points by synparb 14y ago|0 comments
72.
by synparb 14y ago
There is also an example of Numba in action in the comments of the following post about benchmarking Cython's memoryview for array slicing: http://jakevdp.github.com/blog/2012/08/08/memoryview-benchma...
73.
Numba and LLVMPy(technicaldiscovery.blogspot.com)
1 points by synparb 14y ago|1 comments
74.
Cython memoryviews for fast and clean array access(jakevdp.github.com)
1 points by synparb 14y ago|0 comments
75.
by synparb 14y ago
This piece also has some nice numpy specific information for scientific computing.
76.
Intro to writing C-extension for Python(danfm.ca)
89 points by synparb 14y ago|7 comments
77.
by synparb 14y ago
I'll be interested to see how this compares to numba and how easy it is to fit it into the scientific python ecosystem. I'm also curious about performance versus numba and my current go to solution, cython.
78.
Does the compiler know best? (PyPy, LLVM and Python)(pwang.wordpress.com)
2 points by synparb 14y ago|0 comments
79.
by synparb 14y ago
The problem is of course that very few pieces of software beyond the benchmarks actually can scale to 1.6 million cores, so I wonder what the actual functional speed-up is for most research codes moving from the past generation to this gene
80.
by synparb 14y ago
And interestingly, this appears to have generated no discussion on the Freesurfer mailing list: https://mail.nmr.mgh.harvard.edu/pipermail//freesurfer/2012-... There was a single post dated June 2nd that didn't receive any replies: https
81.
by synparb 14y ago
The original article (freely accessible via PLoS One): http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjourna...
82.
by synparb 14y ago
I don't think one should do something just because it is familiar syntactically. Having left MATLAB for python, there were enough compelling reasons to ditch the familiar.
83.
by synparb 14y ago
As a grad student I was definitely a polygot programmer, but I've since really settled into an almost 100% python workflow, except when I used highly optimized simulation packages that run on massively parallel machines. The reason why pyth
84.
by synparb 14y ago
I don't mind writing cython to judiciously optimize my scientific python code when needed, but efforts along these lines, as well as Travis Oliphant's Numba project ( https://github.com/ContinuumIO/numba ) seem like promising alternatives.
85.
Julia, Python and Cython(groups.google.com)
107 points by synparb 14y ago|15 comments
86.
The need for an embedded array expression compiler for NumPy(wesmckinney.com)
2 points by synparb 14y ago|0 comments
87.
by synparb 15y ago
I just looked at Microsoft's offering, and while it has a lot of nice features it seems like the citation counts are wildly wrong when compared to the numbers produced by Web of Science and Google Scholar. Perhaps they have not completed th
88.
Zeromq and the Crossroads fork(imatix.com)
18 points by synparb 15y ago|0 comments
89.
Good, Cheap, and Fast: Pedantic Code Construction Contract for OSS(unprotocols.org)
2 points by synparb 15y ago|0 comments
90.
by synparb 15y ago
The best thing to do is to take a look at the Cython docs ( http://cython.org/ ) and check out some of the whitepapers they have posted. There is also a bunch of cython code on github. If you search "cdef extern" you should find examples wh

More ›