improbable22

12 ms·

91.
by improbable22 7y ago
The default Array does not have a size as part of its type. There's a package StaticArrays which does this, typically faster below about 100 elements. But this isn't useful for catching mistakes before you run it, obviously. Plott
92.
by improbable22 7y ago
> will break naive min/max calculations which start accumulator with +/-inf. Wouldn't the equivalent pattern just be to start at +/- typemax(Posit) or whatever? As you would for integers, which lack infinity.
93.
by improbable22 7y ago
Like I said, what's most readable is problem-dependent. Sometimes it's about modular arithmetic, but sometimes it's about not having to say Γ(n+1) too often. The number line concerns real numbers, not elements.
94.
by improbable22 7y ago
Ever heard someone talk about their zeroth kid, without making a programming joke? English counts objects one two three. I've never heard of a natural language doing otherwise, we would surely translate what they say when pointing at t
95.
by improbable22 7y ago
Indeed, and for a little longer than 12 years! Flipping through a few of his notebooks, I guess he's from the part of academia that was core Matlab-land, engineering-style number-crunching. Mathematica is mostly people doing symbolic t
96.
by improbable22 7y ago
You should think that the cost is basically per floor area. The cars are already as light as they are willing to go (for crash safety etc). The minimum acceptable space gets you to a price that a trainload of people per day can afford. The
97.
by improbable22 7y ago
Indeed it's subsidised, IIRC "about half" is a decent ballpark figure. This is worth bearing in mind when people complain about Ryanair getting not-full-price landing spots, not paying road tax, or whatever. Tourists pay 25%
98.
by improbable22 7y ago
Yes to all this. I also wonder whether cars ought to have something less loud than the horn, for communicating with people who aren't encased in metal. Like trams (many places) have a little bell.
99.
by improbable22 7y ago
Anyone else used the wired ones like TFA, or the numerkeyless variant? Presumably these are simpler to get right than bluetooth (which I presume you're describing).
100.
by improbable22 7y ago
You can buy spreads which are just hazelnuts and chocolate. But they cost several times as much as nutella, unsurprisingly.
101.
by improbable22 7y ago
LyX is also structured in the sense you describe. But somehow I never realised that TeXmacs is actually not TeX at all, I assumed it was another front-end. How on earth did it get such a confusing name?
102.
by improbable22 7y ago
I guess "easier" is a relative term; welding something together in your garage is never going to be as easy & efficient as buying the Toyota... but what the binding constraints on the field's progress are I don't cla
103.
by improbable22 7y ago
A related fun fact, when slicing several dimensions: >>> a = np.arange(9).reshape(3,3) # a matrix >>> a[0:3,0:3] # ranges are treated independently array([[0, 1, 2], [3, 4, 5],
104.
by improbable22 7y ago
A warning sign that your faith in 0-based indexing may be faltering -- catching yourself writing comments like this :) # element at the top right. i.e. (1, 2) aka (0, 1) in python A[0, 0] * B[0, 1] + A[0, 1] * B[1, 1]
105.
by improbable22 7y ago
There is this, though shorter: https://julia.guide/broadcasting
106.
by improbable22 7y ago
It would be great if it were quicker. But right now, interactive use is pretty good, like 5ms for a simple plot. It's calling julia from the command line, and thus starting cold, which is more than 5s.
107.
by improbable22 7y ago
I thought it had got better, but also I've just adjusted to work around it. Here are some timings today, Julia 1.1, cold start to first plot: $ julia -e '@time (using GR; plot(rand(20)))' 3.931433 seconds (10.38
108.
by improbable22 7y ago
There are indeed frustrating lags due to JIT, but they have got better lately & are being worked on -- if I understand right this is now one of the priorities, after focusing on getting the breaking changes done before 1.0. Here's
109.
by improbable22 7y ago
My tip: on recent macs, if you have two keyboard layouts, there's an easy option to make caps lock switch them. (Great so long as you got a mac with a real escape key too...)
110.
by improbable22 7y ago
Also works in the Julia REPL, and jupyter notebooks. I'd hope that the tooling for any language which allows unicode variable names would support this.
111.
by improbable22 7y ago
I don't know, but you may enjoy this neat paper on such matters: https://arxiv.org/abs/1603.00299
112.
by improbable22 7y ago
OK, I'm glad to hear... you are literally the first person ever to tell me law school wasn't as bad as it sounds!
113.
by improbable22 7y ago
Good catch, I did in fact get a B in grammar. Is there a term for this? "This nonsense should cease" does differ from "you should go" in that it demands an action on an inanimate thing, while avoiding saying who acts.
114.
by improbable22 7y ago
It certainly won't stop just because we wish it away. Note the passive voice in "This nonsense should cease." Who, exactly, should do this? Some junior professor, who's up for tenure based on research output, and teachin
115.
by improbable22 7y ago
One downside of class ranking is that it incentivises destructive behaviour. Law schools often do this, and I heard many stories of things like (in the days of physical books...) moving the perfect references, a week before the deadline, to
116.
by improbable22 7y ago
Maybe you know this, but LyX is somewhere in the middle. You can see roughly what you are doing while writing, which makes some things much easier. Especially typing in large formulae (and double-checking that you didn't make a mistake
117.
by improbable22 7y ago
Thanks. Do I understand right that the Bayesian gaussian-process things people do here use only the fully trained loss as input, i.e. just one number L(W), being minimised over W? As opposed to something more detailed about the model, viewe
118.
by improbable22 7y ago
Thanks. What's DFO? And what do you usefully do with multiple objectives, besides minimise some total?
119.
by improbable22 7y ago
Is there a good reason not to regard this as a standard few-parameter no-gradient optimisation problem, and use something like Nelder-Mead on it?
120.
by improbable22 7y ago
Cool, I think we're on the same page. I know what you mean by "orientation of the manifold along the path" but this isn't precisely the standard usage. (My vague memory is that you can define an orientation without needi

More ›