cfcf14

8 ms·

61.
by cfcf14 5y ago
I lived and worked in Edmonton and Calgary for about 8 years before moving to London (UK) and I can confidently say that I have no intention of ever going back. To me, those two cities are exemplars for the concept of 'soul-less glass
62.
by cfcf14 5y ago
Wonderful explanation, thank you.
63.
by cfcf14 5y ago
I'd really like to read the whole text, are there any non-paywalled versions available. The topic seems intriguing.
64.
by cfcf14 5y ago
This is bewildering and concerning. I absolutely applaud the authors for releasing their code and asking the public to have a look, as well as raising the alarm. Given the degree of due diligence they've performed, it seems too much to
65.
by cfcf14 5y ago
This is a relatively common trick in ML - mapping temporal data to image form so one can use the relatively efficient convolutional operations on it. An example would be mapping audio data via a short-time fast fourier transform ( https:&#x
66.
by cfcf14 5y ago
I'm so exhausted and demoralised by this endless cascade of things-getting-worse. It's taking a really big toll on my mental health and there's no resources available in my country to offer support (and there's nothing t
67.
by cfcf14 5y ago
This is a very strong condemnation of their efforts, and I'm skeptical that you're making this claim in good faith.
68.
by cfcf14 5y ago
From Clippy to this technology, in just under 25 years! Very impressive. I wonder how much impact this tool might have as a teaching aid, as well?
69.
by cfcf14 5y ago
Thanks for sharing this lovely anecdote, and for the positive sentiment. Unfortunately I do agree with the person you're replying to - these days, it is vanishingly unlikely to find a partner past your early 30s. Only a tiny proportion
70.
by cfcf14 5y ago
This is a really baffling decision from Apple - normally they're very good about having tightly integrated ecosystems where everything just 'works' with everything else. That's kinda the entire value proposition in my ey
71.
by cfcf14 5y ago
Have you been successful in implementing non-trivial computational code in numba/numpy? I've always found it starts to really break for anything which isn't really trivial, and the errors are mostly non-prescriptive and highl
72.
by cfcf14 6y ago
I found the literary style of this article to be really aggravating and distracting. The content is interesting, but the author is showing off their prose more than actually communicating anything.
73.
by cfcf14 6y ago
As silly as this is from a chess point of view, I really genuinely enjoyed watching Hikaru Nakamura's face as he sees the opening. Needed a good laugh!
74.
by cfcf14 6y ago
Preface: I think this is probably giving the colleague too much credit, but with that thought in mind... I learned the hard way once that even a (seemingly) innocent comment about a friend's tan could be perceived very adversely, due
75.
by cfcf14 6y ago
Aside from the primary content regarding the hijacking the registrar, I really enjoyed reading about the methodological approach they adopted for tracking information and contacts during the crisis. But to the primary content - I've be
76.
by cfcf14 7y ago
The research published in this case (the original word2vec paper) is quite literally not what it claims to be. It is factually incorrect. Is that really the responsibility of the individual attempting to reproduce the work to identify and
77.
by cfcf14 7y ago
Mostly machine learning for distributed sensor networks (ie: smart meters). Deal with lots of time-series data, state space modelling, some recurrent neural nets (etc). Our shop has a 'golang only' thing going, which means that I
78.
by cfcf14 7y ago
I completely agree. When I first started writing Go (coming from Python/C#) I complained an awful lot about what felt like pointless hamstringing of functionality. Go is a simple language, and you don't get many toys. It also feel
79.
by cfcf14 7y ago
I've been really interested in learning Fortran for a while- I work doing a lot of computational stats/ML and typically my tools are python, R, or Golang. Is there any use-case for Fortran in production settings or should I just a
80.
by cfcf14 8y ago
:) Personally I've never used any serious complex analysis in my job (I'm very grateful too, because I always struggled a bit with it). The closest thing I've seen, which I did run into recently, is the use of complex numbers
81.
by cfcf14 8y ago
Hamiltonian mechanics, along with many other seemingly out of place 'advanced' maths, show up in modern Bayesian statistics pretty frequently. Hamiltonian Monte Carlo/Riemannian Manifold Monte Carlo are pretty cutting edge (a
82.
by cfcf14 8y ago
I've used Gaussian Processes to great effect in the field of energy regression/forecasting for commercial buildings, and have found them to be generally superior to other approaches due to the richness of prior information you can
83.
by cfcf14 8y ago
It goes without saying that SV pays dramatically more than London, and certainly more than mainland EU (Paris, Frankfurt, etc). My point is more that as a Canadian you can go pretty much anywhere else and be better paid (even if only a litt
84.
by cfcf14 8y ago
Canadian living in London for about 4 years now - of course people are leaving. The pay is incredibly low in major Canadian cities like Vancouver & Toronto; I was looking at 50-60k/yr there, ~150k in the US, or about 60-70k GPB in