esquire_900

7 ms·

31.
by esquire_900 4y ago
I'd even argue that if you are given a bunch of code of which 20% is incorrect, it's going to take more than 20% time to fix that. If you've written 100%, you know what the code is doing, what kind of data the variables are s
32.
by esquire_900 4y ago
You are quite right, it's something I've been considering for a long time as well. The cold start problem however is so large, that it has to be an open protocol supported by many of the biggest companies for it to have even a sli
33.
by esquire_900 4y ago
I've always wondered about this without trying it myself. Isn't this be something that looks ideal to start with, but as the project progresses and the edge cases keep stacking, you end up writing a complete middle layer, which ef
34.
by esquire_900 4y ago
I've tried it a number of times, and had a similar experience. The whole stack is orders of magnitudes slower to compute compared to "simple" features (i.e. rolling averages), without showing real predictive improvements.
35.
by esquire_900 4y ago
Thanks for the list, I ran across runpod.io yesterday (no experience with them though)
36.
by esquire_900 4y ago
You might be right, though it's Linux only, and still has a "coming soon" as documentation. Worth keeping an eye on, but seems more in the experimental phase compared to bun.
37.
by esquire_900 4y ago
Your comment struck because superciliously I agree, but when thinking about it, it's quite true. Both are strongly related but quite different, just as a marathon runner (though better than average) will not set world records on a 100m
38.
by esquire_900 4y ago
Clickbait title it seems; USDT and DAI lost some market share while BUSD and USDC have grown more. > USDC has grown 20% with $10.6 billion more tokens in circulation. BUSD boosted up 22% — representing growth of $4.2 billion. USDT has sh
39.
by esquire_900 4y ago
I agree. Given the rounding (43.0 and 42.9 could theoretically just be a 0.01 difference) & measuring bias there might be no declining effect here at all.
40.
by esquire_900 5y ago
* Gun homicide is 44 times less frequent in the Netherlands than in the US: 0.28 / 100K people, per annum in NL, compared to 12.21 in the USA. So, this is much more of an extreme event over there. > The extremity here is that Peter
41.
by esquire_900 5y ago
I would say manual feature extraction? Your custom extraction could reduce the variable lengths to a uniform dimension (same number of features for every input), which can then be used by almost any algorithm. These automatic extractions ar
42.
by esquire_900 5y ago
> Apologies if this is a noob question, but could you improve it with training? Considering all other things like GPU, code quality etc. equal, the speed here mainly depends on the size of the network. Larger networks result in better pe
43.
by esquire_900 5y ago
Perhaps the question is why the Bay Area is the golden standard here? Half the salary would make you top 10% earners in almost all other places in the world, but because the salary is paid at SF levels all the "poorer" users have
44.
by esquire_900 5y ago
I was going to give a similar comment here. I get that this is more common in the US compared to Europe or asian countries, but I'd expect that 5-6k a month would be a far more reasonable salary. You're not in open source for the
45.
by esquire_900 6y ago
This isn't a joke, they are fully aware that it was going to backfire. It's just free exposure.
46.
by esquire_900 6y ago
> we found that up to 99.1% of the microbes in sediment deposited 101.5 million years ago were still alive and were ready to eat > But the real secret of their remarkable survival lies in their metabolic rate. It is just slow enough f
47.
by esquire_900 6y ago
Interesting question, wondered that myself as well. My statistics are a little rusty, so doing a quick simulation shows that the range is about .25-.29 (code sample 1). However, things like operations aren't completely random, and you
48.
by esquire_900 6y ago
You make a good point, though I'd like to add that one would expect 1/365 = .274% of procedures to be on a random day (like a birthday). .21% is not so far off that it can't be random chance.
49.
by esquire_900 6y ago
I had the almost complete opposite reaction: the user is telling something isn't working, the author interprets that he needs to fix something for the specific use case of the user. The user might have found a bug or edge case that&#x2
50.
by esquire_900 6y ago
After reading and heavily agreeing with this post, I came to the conclusion that either goodreads is not really trying, or -more likely- the data is just not good enough to make decent recommendations. There are so many biases in the review
51.
by esquire_900 6y ago
Bought a second hand GTX 1060 a while ago to play around a bit more serious with neural networks. It's a good balance between being cheap, 6GB of memory but still being serious enough to get some work done. If you are a professional re
52.
by esquire_900 6y ago
Most of them are indeed, though they specifically mention adding the issue to the roadmap, which would be some kind of acknowledgment, but didn't. ( https://github.com/ONLYOFFICE/DocumentServer/issues/19#i
53.
by esquire_900 6y ago
It's a great read and an educative insight into the working behind onlyoffice. The issue (RTL support) is dodged, promises are made ("It's on our roadmap") but none are kept. When gh user ramezrafla starts doing the heav
54.
by esquire_900 6y ago
The title doesn't really relate to its content very well; the concept of taking cold showers has some scientific backing ([1] & [2]), and is also slightly hyped. After taking cold showers and getting some (minor) benefits for some
55.
by esquire_900 6y ago
The landing page seems to be a bit confusing. This page does a better job at explaining it imo: https://tiledb.com/embedded/ Without experience with the product, it's core seems to be a wrapper around stored data
56.
by esquire_900 6y ago
It is a hyperbole; that's exactly the point that 3131s is making. It is a (quite extreme) exaggeration, and therefor factually not correct. I might indeed make some assumptions about smell. Though having smelled both durian and rotting
57.
by esquire_900 6y ago
It does smell, but by far not as repulsive as the smell of actual death (rotting meat).
58.
by esquire_900 6y ago
No problem in depending on CapRover, as long as you are at least somewhat familiar with the tools it sets up for you. Combine that with some crude generic UNIX skills (quickly analyze cpu/ram/disk usage, search in logs, transfer f
59.
by esquire_900 6y ago
And that's the problem; by then it's too late (i.e. never took the time to backup the db). It's about finding the balance, writing a PHP app shouldn't involve studying C compilers and CPU design. But I think these tools
60.
by esquire_900 6y ago
Isn't that a setup waiting for disaster to happen? Everything is happily running up to a serious production problem, at which point you miss the experience to debug and fix it.

More ›