mattip

21 ms·

181.
PyPy 5.0.1 Released(morepypy.blogspot.com)
2 points by mattip 11y ago|0 comments
182.
by mattip 11y ago
Personally I prefer wxwindows, which doesn't need upgrading because it just works, but people keep telling me to use qt (which requires a commercial license) and to keep upgrading it too. So dunno who is right.
183.
by mattip 11y ago
We (PyPy release manager here) adopted a release policy where we update the major version about every three months. Bug fixes, if needed, will be 5.0.1, and 5.1 should be released in a month or so
184.
PyPy 5.0 Released(morepypy.blogspot.com)
284 points by mattip 11y ago|62 comments
185.
by mattip 11y ago
Has this been done for any project? I am interested in seeing the setup used, resulting reports, and changes instituted
186.
Calling c from numba uaing cffi(nbviewer.jupyter.org)
3 points by mattip 11y ago|0 comments
187.
Research shows connection between content clarity and business credibility(precisioncontent.com)
2 points by mattip 11y ago|0 comments
188.
by mattip 11y ago
Out of 8 million data points the top book got around 50 references. I wonder how much significance should be attached to that, it looks to me to be down in the noise level.
189.
by mattip 11y ago
Is there more info around about this style of project management? My google-foo is not getting any hits
190.
by mattip 11y ago
Many other solutions implement both loop fusion and shortcut deforestation, for instance theano. In PyPy we had this for numpy expressions, we called it "lazy evaluation", we have temporarily removed it until we achieve more funct
191.
Python C API, PyPy and the road into the future(lostinjit.blogspot.com)
178 points by mattip 11y ago|55 comments
192.
by mattip 11y ago
yes. However the delegation only happens if you use the numpy.linalg module, or if numpy has chosen internally to call the numpy.linalg module (i.e. logic in numpy.dot will call out for large ndarrays) Our vectorizing JIT can use SIMD seman
193.
PyPy 4.0.0 Released – A Jit with SIMD Vectorization and More(morepypy.blogspot.com)
303 points by mattip 11y ago|144 comments
194.
by mattip 11y ago
That make 6 Linux Foundation Fellows; Linus Torvalds, Greg Kroah-Hartman, Till Kamppeter, Richard Purdie, Janina Sajka, and now Thomas Gleixner
195.
PyPy 2.6.1 released(morepypy.blogspot.com)
5 points by mattip 11y ago|2 comments
196.
by mattip 11y ago
Performance improvements have been driven by user-reported slowdowns. The more users that report performance problems, the more the entire community profits. There are more performance improvements on the way: STM [1] will do away with the
197.
by mattip 11y ago
PyPy 2.6.0 - "Cameo Charm" since PyPy looks best in profile
198.
PyPy 2.6.0 released(morepypy.blogspot.com)
147 points by mattip 11y ago|28 comments
199.
by mattip 11y ago
Why only for cyclists? Why not collect data from multiple sources: reports from pedestrians, driver, security cameras facing intersections...
200.
by mattip 12y ago
Born April 15, 1452, this week 563 years ago
201.
by mattip 12y ago
That was probably a troll, but I'll bite. See, that's what I mean. Anyone can code a quick sort or a sobel filter. But when to avoid the sort in the first place, or how to build an image processing pipeline for CPU,GPU,FPGA, asic
202.
by mattip 12y ago
Over 50 programmer here. What I bring to the table is not only programming skills, but life experience, people skills, and a desire to mentor others since retirement is on the horizon. Keeping up-to-date is important, but just being around
203.
Linalg support in pypy/numpy(morepypy.blogspot.com)
3 points by mattip 12y ago|0 comments
204.
Please do my Homework(blog.aclark.net)
4 points by mattip 12y ago|0 comments
205.
by mattip 12y ago
You mean "I wish you could run scons with PyPY". It looks to me like scons does contents = open(file).read() where they should be doing with open(file) as fid: contents = fid.read()
206.
by mattip 12y ago
PyPy dev here. Could you give us some info on how to help you more, for instance, a small benchmark of where we are still slow? We love to hear how PyPy is being used "in the real world"
207.
Would type annotations help PyPy’s performance?(doc.pypy.org)
2 points by mattip 12y ago|1 comments
208.
by mattip 12y ago
My take away from "Do just enough" is the flip side - throw it away and start over. The second part is just as important as the first or you end up with crap prototypes in the heart of your solution. The first time doesn't ha
209.
by mattip 12y ago
Maybe they stop to run up the meter more
210.
by mattip 12y ago
You should certainly be proud of this work, it it beautiful. But be careful of patent-protected feature detectors like SIFT, you could get burnt[0] There are alternatives that perform as well, also mentioned in the link [0] http:/&#x2F

More ›