stellalo

7 ms·

31.
by stellalo 1y ago
Ideally though, the compiler lowers all domain specific logic into simple byte-moving, just after having checked that types add up. Or maybe I misunderstood what you meant?
32.
by stellalo 1y ago
Career shortens life
33.
by stellalo 1y ago
Oh that’s exactly what the book’s author blog mentions: https://norasandler.com/2017/11/29/Write-a-Compiler.html
34.
by stellalo 1y ago
From what the blog author says (I haven’t looked into the book), the approach reminds me of > Abdulaziz Ghuloum, 2006, An Incremental Approach to Compiler Construction http://scheme2006.cs.uchicago.edu/11-ghuloum.pdf
35.
by stellalo 1y ago
> At 12.8 deaths per 100,000 people, it is double that of Greece, triple that of Austria, and six times more than Japan. Should this not be measured in terms of fatal accidents every million of km driven? To normalize by how much people
36.
by stellalo 2y ago
Luckily the founding fathers thought of this
37.
by stellalo 2y ago
To be precise (because I was confused initially): the building being shown in pictures is the current “main hall” of KU Leuven in Naamsestraat (second picture), nearby Oudemarkt (first picture), and not the university library in Ladeuzepl
38.
by stellalo 2y ago
So, screw the planet?
39.
by stellalo 2y ago
All “high risk” uses listed in the article (“Systems that are used in sectors like education, employment, law enforcement, recruiting, and essential public services, Systems used in certain kinds of products, including machinery, toys, lif
40.
by stellalo 2y ago
Looks like this makes (at least initially in training) the “negative” attention term smaller in the early layers (smaller l) compared to later layers (larger l). Which I guess makes sense: you probably want to attend a little bit to everyth
41.
by stellalo 2y ago
> The researchers […] say the issue can be fixed by formatting Excel columns as text and remaining vigilant—or switching to Google Sheets, where gene names are stored exactly as they're entered. I think the safest fix is to avoid sp
42.
Chronos: Pretrained (Language) Models for Probabilistic Time Series Forecasting(github.com)
8 points by stellalo 3y ago|3 comments
43.
by stellalo 3y ago
> If I include the time for Julia to start up and compile the script, my implementation takes 354 ms total, on the same level as Mojo's. I don’t think the article mentions it explicitly, but I suppose the timing is from Julia 1.10:
44.
by stellalo 3y ago
What is the capital of France?
45.
by stellalo 3y ago
> and found Julia to be a lot of talk but seemed clunky to me This is a bit vague, any concrete example?
46.
by stellalo 3y ago
> Here is what has changed: we no longer trust developers as much and we are re-introducing the complexity that we were fighting. What I think has changed instead: Python is used for large code bases, unlike 15 years ago, and without typ
47.
by stellalo 3y ago
Probably via retrieval augmented generation (RAG) https://www.promptingguide.ai/techniques/rag
48.
by stellalo 3y ago
> Programming languages do matter but not as much as many people think f.ex. the HN crowd has a soft spot for LISP, and most of the don't even have proper parallel execution I don’t understand how the argument about LISP would imply
49.
by stellalo 3y ago
It’s not one run: they are using BenchmarkTools that provides macro to compute exactly the kind of statistics that you expect. One can say anything, but the Julia community takes performance and benchmarking really seriously.
50.
by stellalo 3y ago
> The least amount of effort to get something done much faster wins by default. And then you’re stuck with a proprietary language “subscribe to download”? Not sure it’s the least amount of effort honestly.
51.
by stellalo 3y ago
I guess “advances in technology” are not only new CPUs and GPUs, but also many other kinds of hardware and software
52.
by stellalo 3y ago
I’m not sure looking at the absolute number of incidents is very informative. How did the total number of miles-per-day from autonomous vehicles vary in the same timespan?
53.
by stellalo 3y ago
If the problem is remembering wether to add or subtract 1 after complementing, then one just needs to take decimal 0 = 0x00 as example: complement (0xFF) then necessarily add 1 (0x00) to get -0 = 0.
54.
by stellalo 3y ago
If there’s a company that can make the VR/AR headset thingy fly, that is Apple. I’m not sure it will fly (I’m skeptical) but I’m happy that we will finally find out.
55.
by stellalo 3y ago
> I actually would prefer it over Python for small scripts, but the compilation overhead is too long Looks like this release reduces that by a lot, see the first section in the OP on caching native code, modulo adoption of good precompil
56.
by stellalo 3y ago
The library is called Plots because it plots, what is confusing about it?
57.
by stellalo 3y ago
> Tbh, from the subsequent manual listing it's not even clear it's a math-focused programming language It’s true that the majority of the Julia community comes from a scientific or engineering background, and this is reflected
58.
by stellalo 4y ago
Meta jumped up 5% at their new layoffs announcement last week
59.
by stellalo 4y ago
Nitpick “2.5 Mutable Global State Avoid mutable global state. […] 2.5.2 Pros Occasionally useful.” So, are these pros of mutable global state or of avoiding it? Elsewhere pros and cons appear to refer to the description of the guideline, no
60.
by stellalo 4y ago
So 2022 was better than average?

More ›