rq1

28 ms·

331.
by rq1 7y ago
You need to take into account the distance between the airport and the city center, the security checks, on-boarding, checking in, getting your luggage back... etc. It’s both time consuming and annoying.
332.
by rq1 7y ago
What are you trying to say? Is Apple still using the 90s’ factories and technology?
333.
by rq1 7y ago
Can you provide some references? Using vi for 5y here (switched to neovim) in dark mode, with nofrils-dark[1] theme now, I don’t feel any less productive, on the contrary I can focus for way longer than before. [1] https://github
334.
by rq1 7y ago
The football market.
335.
by rq1 7y ago
You need to take into account the added value of productivity gains: who’s getting it? If I have a better yield for a given task, then I will provide exactly what my employer asked me for the price we agreed. Eg. That means that if we agree
336.
by rq1 7y ago
https://arxiv.org/pdf/1807.06699
337.
by rq1 7y ago
> a dangerous mix of dry air, warm temperatures and gusty winds I was wondering if there were similar technics to preventive avalanches to fight fire?
338.
by rq1 7y ago
I was going to say “Loewry who?”. Cultural bias I guess.
339.
by rq1 7y ago
What I learned as developper and Data Scientist is: either you know what you’re doing and can see the big picture, either you’re walking blindly to a “good enough” solution. In the first case, I spend most of my time in architectural design
340.
by rq1 7y ago
Thank you for this. I remember my first reaction to the Neural ODEs paper was to tell my professor that they forgot to mention the parabolic PDEs case (through SDEs).
341.
by rq1 7y ago
Yes that’s exactly overfitting the bootstrapped samples, thus the high variance. The « variance » won’t just magically vanish as you average things out[1], you need to change the scale and check out the asymptotic law of your estimator (CLT
342.
by rq1 7y ago
You should show them LaTeX and its powerful features, and how VIM is such a great and way better editor than emacs.
343.
by rq1 7y ago
These are (false) claims, therefor not proofs. Deep trees will fortunately overfit your dataset. Any binary tree of depth log2(P) can completely separate your P points.
344.
by rq1 7y ago
> RF trees give unbiased fits, but they're high variance. Sounds like overfitting.
345.
by rq1 7y ago
This.
346.
by rq1 7y ago
Thank you for this reference!
347.
by rq1 7y ago
My bad I forgot the verb, I meant “are equivalent”. Edit: I don’t understand the downvotes though.
348.
by rq1 7y ago
> Regular Expressions have minimal performance impact REs and FSMs equivalent.
349.
by rq1 7y ago
I don’t think so, afaik.
350.
by rq1 8y ago
Congratulations! Keep going! Adding 20-30mn HIIT sessions fasted right before breaking the fast may help also. The hypertension may be related in some cases to sodium depletion causing aldosterone production (and cortisol and adrenaline) le
351.
by rq1 8y ago
Go for keto + IF. It helped a lot of people.
352.
by rq1 8y ago
Isn't it a generalist' ability to specialize Just In Time?
353.
by rq1 8y ago
Nice! :-)
354.
by rq1 8y ago
Nixon?
355.
by rq1 8y ago
And you don’t necessarily want a global minimum of an universal approximator...
356.
by rq1 8y ago
Nice try. :-)
357.
by rq1 8y ago
Quite natural. AI in market finance is a fraud for the moment. AI models totally fail to do what classical (and parsimonious, explainable, cheap...) methods/algos/models achieve quite easily (BS, Hawkes, RFSV, uncertainty zones, A
358.
by rq1 8y ago
http://conal.net/blog/posts/the-c-language-is-purely-functio... https://news.ycombinator.com/item?id=618817 https://news.ycombinator.com/item?id=9088676 :-)
359.
by rq1 8y ago
There is no misuse of anything. Stop being so presumptuous and shortsighted, especially on hacker news. Is there some theoretical bounds or anything preventing the compilers to be faster or is it just implementation/optimization deta
360.
by rq1 8y ago
I love it! Thanks!

More ›