achompas

17 ms·

121.
by achompas 9y ago
Dumb question — you bill for this “non-programming value-added” work, right? Even jumping into a truck for a few hours? I fully understand the value in “doing more” but still occasionally have to fight the feeling that I need to be coding a
122.
by achompas 9y ago
It's almost certainly a prop. At worst, a previously-used wallet that has since been emptied.
123.
by achompas 9y ago
Sorry, where do you see this? I see a lot of slides devoted to the "in-between" of random forests, perceptrons, etc. The jump from supervised to unsupervised to RL also makes sense, since RL is a different learning paradigm from t
124.
by achompas 9y ago
> I'm not an expert on water resistance but I'm guessing something can be both resealable and water resistant. I’m not either, but you’d have to figure out an extremely special way to create a water-resistant phone with battery
125.
by achompas 9y ago
> There is plenty of resources and Python 2 support is neither a burden nor this burden in any way increasing. Aside from what others have said, NumFOCUS is woefully underfunded. If you're interested in seeing continued development
126.
by achompas 9y ago
> I'm glad this thing is still working like new and I don't have huge urges to upgrade because I suddenly feel inferior, like some other tech companies try to pull... I don't get this. Your feelings of inferiority are your
127.
by achompas 9y ago
This is the thesis of Cal Newport’s “Deep Work.” It’s an interesting book, I recommend reading it.
128.
by achompas 9y ago
My understanding from speaking with sleep specialists is that you'd need to take a large amount every day for weeks or months to develop a tolerance. They thus recommend cycling off melatonin bi-weekly or monthly for a little while. (S
129.
by achompas 9y ago
> Looks like a watch I hear this constantly and I don't get this. There's a very long tradition of non-circular watches in horology. See the Reverso, the Tank, and the Monaco.
130.
by achompas 9y ago
Now you're just trolling. :)
131.
by achompas 9y ago
Yep, we arrived at my larger point: if you care about interpretability, NNs are horrible and Bayesian techniques are pretty damn great.
132.
by achompas 9y ago
And you don't have computational efficiency issues with NNs? We're also ignoring the benefits of a posterior distribution, which is useful for understanding the data-generating process.
133.
by achompas 9y ago
> And what is a data scientist then, if their work does not involve analysing data and presenting their analysis? This is a dead giveaway that you have no idea what you’re talking about. You’ve captured about 5% of my work. The rest of t
134.
by achompas 9y ago
> You immediately arrive at computationally intractable sums/integrals. So we instead sample from that posterior. Unless you think MCMC is also obsolete, in which case I’ll see myself out.
135.
by achompas 9y ago
> Yes, it's largely conceptually obsolete. I'm sorry, what? How did you arrive at a point where you believe this is true? This is like calling compilers "obsolete." Is it because you believe deep learning has "ta
136.
by achompas 9y ago
This is absurd and false. This person is an analyst of some sort. Maybe this holds in the consulting world? It definitely does not hold in the tech world, IME.
137.
by achompas 9y ago
What do you get out of your subscription? I love Ben's podcast (exponent.fm) but cannot bring myself to pay the monthly subscription.
138.
by achompas 9y ago
Perfect, not a problem. How about viewing graphs?
139.
by achompas 9y ago
Do you write software using your iPad Pro? I've thought about using one with an AWS instance for modeling/prototyping, but it doesn't seem like we're there yet.
140.
No, Soylent Isn’t Healthy. Here’s Why(nateliason.com)
1 points by achompas 9y ago|0 comments
141.
by achompas 9y ago
Ahh, I see. I was confused about whether semi-supervised approaches rely on using predictions on the unlabeled data to improve model performance. Wikipedia seems to suggest this is a key component which isn't mentioned in OP: > Semi
142.
by achompas 9y ago
How so? I don't see reference to sampling with replacement or a suggestion to re-use the unsupervised results to further improve the model. Seems like I'm missing something...
143.
by achompas 9y ago
I haven't seen this, although I hear it's also excellent. Thanks for the rec!
144.
by achompas 9y ago
Very surprised to pop in and see discussion of Dune without mention that Denis Villeneuve is rebooting Dune. [0] Villeneuve is also working on the Blade Runner sequel scheduled for release in October. For anyone who has seen Arrival, Sicari
145.
by achompas 10y ago
Day One for general notes, 2Do for notes on tasks. (I use Mac and iOS.)
146.
by achompas 10y ago
It's not the final valuation though, there are other factors. Time value of money, dilution through failed attempts to raise more money, etc. From OP's perspective, he could go to Japan with his wife now, or maybe get his money
147.
by achompas 10y ago
But we don't know their private valuation as of right now. So OP isn't just out $4k+, he's also going to pay taxes on the private valuation of his shares.
148.
by achompas 10y ago
Where would you suggest someone begin with Foucault?
149.
by achompas 10y ago
I'm 31 and I use my iPad Air 2 constantly. Great for reading textbooks/papers, as a second screen for TV shows, or for streaming (whether at home or on trips). It is very much a consumption device, though -- the most I "creat
150.
by achompas 10y ago
A good friend -- advertising-focused designer with lots of Photoshop and Flash experience -- has been exploring this transition. I also have a "unicorn" title myself (data scientist, lol), and think this would be an interesting tr

More ›