stabbles

33 ms·

571.
by stabbles 6y ago
FWIW, MaBLAS currently does not depend on LoopVectorization.jl, the code to generate kernels is all handwritten.
572.
by stabbles 6y ago
> I don't think writing eg ARPACK performantly in Julia is impossible Not sure what you mean, ARPACK just wraps a bunch of calls to LAPACK and BLAS, that's all. It does not have any low-level linalg kernels of its own. Also, it
573.
by stabbles 6y ago
You're probably confusing ARPACK with BLAS / LAPACK. Pure julia ARPACK already exists, e.g. https://github.com/haampie/ArnoldiMethod.jl/ . A competive BLAS-gemm is implemented here https://gith
574.
by stabbles 6y ago
StaticArrays.jl is just linear algebra operations for tuples, and they have always been stack allocated as they are immutable and have a size known at compile-time, so Julia 1.5 will not bring anything new for it. The deal with 1.5 is that
575.
by stabbles 6y ago
Inverse problems are ill-posed, you cannot just invert the kernel, information is lost.
576.
by stabbles 6y ago
There is a very beautiful way to do it using much less memory. You need O(sqrt(n)/log(n)) memory, not O(n), if you keep track of the list of primes found so far, and a fixed-size interval over which you sieve: [1, k], [k+1, 2k], [2k+1,
577.
by stabbles 6y ago
Nice, maybe std::string has changed in GCC 10, I was using GCC 9. Shouldn't C++20 have constexpr containers?
578.
by stabbles 6y ago
Neat, the way I benchmarked was through BenchmarkTools.jl, which runs the functions sufficiently many times to get a statistically relevant result. It will not measure compilation time, but I did not include that for gcc / clang either
579.
by stabbles 6y ago
This is a fun benchmark in C++, where you can see that GCC has a more restrictive small string optimization. On my desktop the main python example runs in 3.1s. Then this code void example() { for (int64_t i = 1; i <= 20; ++i
580.
by stabbles 6y ago
Gitlab solved this properly by making runners private to (private and public) organizations and repos.
581.
by stabbles 7y ago
The point of the article is to explain the happy statistical exception
582.
Show HN: Ldd as a Tree and Much More(github.com)
2 points by stabbles 7y ago|0 comments
583.
Show HN: Libtree – ldd as a tree and much more(github.com)
1 points by stabbles 7y ago|0 comments
584.
Show HN: Ldd as a Tree with an Option to Deploy(github.com)
4 points by stabbles 7y ago|0 comments
585.
by stabbles 7y ago
On Android you have to specify a hostname for DNS (a bit ironic), which is 1dot1dot1dot1.cloudflare-dns.com. However, it does not seem 1dot1dot1dot3.cloudflare-dns.com works yet.
586.
by stabbles 7y ago
Did that really happen to you or are you confusing it with this issue on MacOS? https://twitter.com/BenoitLetondor/status/939127296266588160
587.
by stabbles 7y ago
When I was in Barcelona an ATM had its keypad numbers flipped (7 8 9; 4 5 6; 1 2 3), which I only realized after the first attempt. Second attempt I got my PIN wrong, and I did not dare to try it a third (and final) attempt. Apparently it&#
588.
by stabbles 7y ago
C++ does not have multiple dispatch, though. Historically in C++ it was considered to be more performant to dispatch function calls on the base type of its arguments, not the concrete types. Consider for simplicity member functions A::f(x..
589.
by stabbles 7y ago
This is hilarious! Turns out 18 year old websites were mobile friendly after all
590.
by stabbles 7y ago
In Europe it is rather easy :)
591.
by stabbles 7y ago
I've started working with Flux [1] in Julia, and it's so elegant and such a great experience :). Just look at this definition of a U-net model for image segmentation: https://gist.github.com/haampie/bceb1d59fd
592.
by stabbles 7y ago
So what version will apt install then... I wonder how much software will break if apt quickly adopts new minor versions.
593.
by stabbles 7y ago
I wonder how many people would know about using return_type = typename std::result_of<F(Args...)>::type though.
594.
by stabbles 7y ago
That's an interesting question. In Kuhn's view on science the tree would be replaced after a paradigm shift, meaning there is enough low-hanging fruit in a period of normal science (between two major breakthroughs). I wonder if Ku
595.
by stabbles 7y ago
"achieves a geometric mean runtime speedup of 1.015× on the NAS benchmark suite when compared to LLVM’s SLP vectorizer" sounds negligible?
596.
by stabbles 7y ago
Yeah, it creates a jump table, so basically an O(1) operation in assembly rather than O(n) comparison worst case for every case. However, this means switch _only_ works with integers, which is rather limited and confuses newcomers. For inst
597.
by stabbles 7y ago
Of course there is no guarantee truths about the universe are simple. Newtonian mechanics is extremely simple, yet it is only true as a limiting case of relativistic mechanics, which is not as simple. Maybe relativistic mechanics is only tr
598.
by stabbles 7y ago
I'm a bit confused it took Protonmail more than a year yo develop ProtonCalendar. Is it really that difficult to develop?
599.
by stabbles 7y ago
Marshall headphones have a 9mm jack on the headphones themselves to solve that problem.
600.
by stabbles 7y ago
> I only have one device. I'm happy I can easily use my headphones with a 9mm plug on my crappy work laptop as well

More ›