menaerus

42 ms·

601.
by menaerus 2y ago
Even if you manage to build the product (R&D) now you have to incentivize the private sector and governments to switch to the EU product. There's only few options of making this happen and tariffs could be one of them. Natural tran
602.
by menaerus 2y ago
You're a servant to the business needs so whatever the business needs are at that moment. It's a vague answer probably not appealing to many engineers but that's what it really is. You're solving problems for your busine
603.
by menaerus 2y ago
I hope you're aware how loose of an argument this is. > Plan 9 is a research operating system, and exists to answer questions about ideas in OS design. As such, the Plan 9 experience is in essence an exploration of the interesting i
604.
by menaerus 2y ago
I am serious. C is a simple language but rather complicated to wrap your head around it since it requires the familiarity with low-level machine concepts. C++ ditto but with a difference that it is a rather complicated language with rather
605.
by menaerus 2y ago
I have professional working experience in this domain.
606.
by menaerus 2y ago
Stability and safety are the least of the concerns in data processing and database workloads. That's totally not the reason why we saw an increase of these systems during the 90s and early 00s written in Java or similar alternative lan
607.
by menaerus 2y ago
If I derive my work using multiple sources, do all the copyright holders from these multiple sources have an exclusive right on my work? How otherwise would people build a knowledge on some topic and then apply that knowledge to build a pro
608.
by menaerus 2y ago
So if you don't want to beat native engines in performance what is it that you're trying to solve but Java-based engines don't have? I think it's pretty important to set a vision upfront otherwise you're going to se
609.
by menaerus 2y ago
Rust is rather heavy on its copy/clone imposed semantics making it potentially less suitable for low-latency or large data volume processing workloads. Picking Rust for its performance potential only means that you're going to hav
610.
by menaerus 2y ago
That's a very simple and non-biased model view. In reality, many people might read your job ad as "so, your profile claims you have the skills but how come then that you don't have a job already?" aka "there's
611.
by menaerus 2y ago
I see rates around $100 all the way up to $200. Is that really considered low in the States?
612.
by menaerus 2y ago
Portability-wise this was perhaps the case ~20 years ago but today I am not aware of any platform that doesn't have a decent C++ compiler so I don't think what you're saying is generally true. I also don't agree on the s
613.
by menaerus 2y ago
Just my personal experience from using the library for at least 7-8 years. So many things and concepts are glued onto each other, making the API so much non-intuitive whenever you try to do anything more sophisticated that isn't a 1:1
614.
by menaerus 2y ago
Generally speaking, hardware found in commercial vehicles is very cheap and therefore constrained - e.g. slow CPUs and slow memory, low amount of memory, limited (flash) storage, and technology is generally lagging behind - you would for ex
615.
by menaerus 2y ago
> of complicated APIs of which the matplotlib is the embodiment. Terrible API with terrible terrible performance.
616.
by menaerus 2y ago
> the adoption of end-to-end AI within the robotics and automotive community remains low What do you think how else lane assistance, traffic sign recognition, highway and traffic jam assist, emergency braking or pedestrian or any traffic
617.
by menaerus 2y ago
libclang has been with us for ~15 years already. > I'm not so interested in being able to compile but being able to parse Chances are that you're not going to be able to walk the AST without getting errors unless the (lib)clang
618.
by menaerus 2y ago
REPL and JIT come to mind.
619.
by menaerus 2y ago
Or all you need is a database that allows anyone to skip the hard part and simply write a plugin for.
620.
by menaerus 2y ago
No, 64K is definitely not loaded in 1 load operation. Maximum amount of load operations modern CPUs can do is 2x512-bit or 128 bytes per cycle or around ~600 GB of L1 bandwidth. That said, there is no universal page size one can choose from
621.
by menaerus 2y ago
Although you didn't say anything wrong, it sounds a little bit discouraging. In reality, it requires effort as it is the case with anything else and it is really not that hard. Case in point, I hardly knew anything about the ML field a
622.
by menaerus 2y ago
> the assembly can mislead you If only metric you look for is number of instructions then yes and this is a wrong way of looking at this type of problems. One can write a 50 LoC heavily optimized SIMD code that outperforms the compiler g
623.
by menaerus 2y ago
In general, 40 samples would certainly not be enough for the experiment result to have statistical significance but perhaps in this case when the code being tested is very tight, e.g. just several tens of instructions, it may be enough. Cod
624.
by menaerus 2y ago
> and adding a single line of code can completely change the way your program is compiled without you even noticing You cannot work on SIMD micro-optimizations without assessing the final compiler codegen. Compilers will easily cancel y
625.
by menaerus 2y ago
I haven't watched that space very closely but IMO ViTs have a great potential to extract from since in comparison to CNNs they allow the model to learn and understand complex relations in the data. Where this matters, I expect it to ma
626.
by menaerus 2y ago
CNN-based OCR also have "hallucinations" and Transformers aren't that much different in that respect. This is a problem solved with domain specific post-processing.
627.
by menaerus 2y ago
It certainly isn't the same cost if expressed as a non-subsidized $$$ one needs for the Transformers compute aka infra. CNNs trained specifically for OCR can run in real time on as small compute as a mobile device is.
628.
by menaerus 2y ago
Even if we take different hardware into account we can readjust for measured vs theoretical throughput. Ceph cluster achieves 1 TiB/s / 1.7 TiB/s = 0.58% of theoretical throughput. 3FS cluster achieves 6.6 TiB/s / 9
629.
by menaerus 2y ago
> People actually have a completely skewed view of what a doctor is. It could be but treating patients also requires continuous diagnostics, result comprehension, and final assessment so this is certainly the part where AI could play the
630.
by menaerus 2y ago
You sort of implied that with your replies across the thread. And since AI already replaced part of the CS, I was wondering why do you think this would not be the case with doctors. I'm not sure I agree it's a blue collar professi

More ›