govg

25 ms·

241.
by govg 8y ago
I think it's a very small segment of their customer base that actually requires cab hailing services in 10+ cities. Most people working day jobs will not visit more than that many cities over the course of the year, and Uber possibly s
242.
by govg 8y ago
https://github.com/jhallen/joes-sandbox/blob/master/editor-p... This is probably what you're looking for.
243.
by govg 8y ago
Even for the young employed folks this is quite the sum. The fact is though, that over the course of an average month it's pretty hard to hit these caps. Its about 800 hours of Netflix HD streaming, and you're probably not going t
244.
by govg 8y ago
That's the absolute top-end of the plans they have in Bangalore. It's around 2000INR (<30$) per month for a much more moderate 150Mbps / 875GB per month plan. And consider that a lot of the urban crowd we are talking about
245.
by govg 8y ago
I'm unsure if Spotify offers these plans elsewhere, but in India they have the following tiers : Rs. 119 per month (1.7$) Rs. 1189 per year (16.85$) And also, funnily, "top-up" plans: Rs. 13 for 1 day (18c) Rs. 39 for 1 week
246.
by govg 8y ago
You're right in that "most" Indians probably do not have broadband access / 4G access. But the urban centers certainly do,and they have enough by way of people to become a huge market. To put it in context, there are at
247.
by govg 8y ago
Usually home internet is via other providers, who also have cheap plans. For instance, ACT Giga is 1Gbps, 2.5 TB for about 100 USD per month. Post this limit, speeds usually drop to 1Mbps.
248.
by govg 8y ago
Even Xhosa has some consonants that aren't clicks, afaik. Also, your comment leads to this interesting wiki article : https://en.wikipedia.org/wiki/Whistled_language
249.
by govg 8y ago
Sure, but the plague of dark patterns and intrusive ads far far outweigh the few legitimately non intrusive ad supported websites. I agree there needs to be some way for sites to make money, but there is just too much stuff out there that s
250.
by govg 8y ago
In the US, ML/AI grad school has significant Chinese and Indian student population. It's not surprising to see this team composition.
251.
by govg 8y ago
Believe Shazam also started as such a dial in service.
252.
by govg 8y ago
But that is not helpful in commuting to Manhattan, is it? You'd still have to use the ferry.
253.
by govg 8y ago
Probability, statistics and linear algebra are recommended because all of machine learning uses these as foundations. Any graduate level course in these areas will cover what you could possibly need. For people working in the more theoretic
254.
by govg 9y ago
AI is far from solving these problems to the extent of being deployed world wide.
255.
by govg 9y ago
Maybe not exactly what you seek, but Microsoft Research China does some really state of the art work in vision. Baidu (that Andrew Ng was affiliated with) also seems to have made really big strides with stuff like speech recognition. There
256.
by govg 9y ago
But how many other examples do we have? It's a rarity to see such an occurrence. That's why Yitang Zhang's work was such a surprise.
257.
by govg 9y ago
Well, the idea behind offering insurance is that you typically don't need to account for everyone claiming it at the same time.
258.
by govg 9y ago
Well, Roombas do exist...
259.
by govg 9y ago
It's a monograph that is pretty dense with not so trivial math. Maybe there aren't many on HN who fully understand the material or have strong ideas about it.
260.
by govg 9y ago
Elseiver has "prestige". What that means is that some of their journals have become the standard destinations to publish at. It's a vicious cycle, to get acknowledged (and perhaps promoted etc) you need to publish in the loca
261.
by govg 9y ago
Reads far more about an advertisement that barely has details than an actual review or comparison article as the headline suggests.
262.
by govg 9y ago
No, this is not strictly true. a) I think you're underestimating how tough it is to actually publish at top tier conferences. Professors at the IITs rarely get more than a few a year. b) There are MS programs that are typically not as
263.
by govg 9y ago
The brain has some inherent biases and prejudices - these arise from just how we grow up and perceive objects. So optical illusions are a deviation from whatever visuals we are used to, and you can say similar events occur with stuff like t
264.
by govg 9y ago
That doesn't really say much about their skill. All Norvig says is that people who are good at competitive programming are used to quickly solving problems and move around, and that probably isn't good when it comes to real world
265.
by govg 9y ago
> For each male hired there must be a female hired. General grad CS populations are split 4:1 in favor of males. It won't end well when you cut out a significant amount of your candidate pool because you want to promote "divers
266.
by govg 9y ago
Neovim has some good documentation on how to get started. Also, spacemacs has vim bindings if you wish, afaik.
267.
by govg 9y ago
One thing you might be mixing up is population vs individual. If you hold some sort of hiring process where the skill tested is similar to that tested in marathon running, then yes, Kenyans will tend to do better as a population. That does
268.
by govg 9y ago
This is in the works, as far as I remember. They plan to plug into the UPI (financial service by the government) to bring P2P payments. Will be quite the gamechanger, especially with how popular Paytm (another p2p payment method / wall
269.
by govg 9y ago
Upgrading a PC is a one time thing you need to do once every few years. You'd be "upgrading" your console by putting it in the bin and buying a new one, so I'm not sure what the argument is against PC gaming here...
270.
by govg 9y ago
Regarding the PCA before SVM step - PCA would help reduce dimensionality w.r.t linear correlations, kernels generally are used to obtain non-linear maps. So wouldn't it be wrong to say doing PCA is redundant? Also, while kernel SVM opt

More ›