VHRanger

59 ms·

781.
by VHRanger 5y ago
See how well that went for El Salvador: https://www.nber.org/papers/w29968 (not well)
782.
by VHRanger 5y ago
That wouldn't work. You can always make more LLCs to dilute your numbers. Ultimately we need to weaken corporate ownership of real estate
783.
by VHRanger 5y ago
99.9999% to be precise At least it's a tax writeoff for the person selling it, I guess?
784.
by VHRanger 5y ago
Intel had toxic culture since Brian Krzanich took the lead in 2013, around the switch to 14nm. His mistakes weren't seen clearly until almost 2019, when it became obvious 10nm was a disaster, and 14nm++++++++ was their only workable pr
785.
by VHRanger 5y ago
The reason time *feels slow* in these catastrophes, AFAIK is that memory formation kicks into high gear during it Your brain isn't somehow over locking itself, you just remember everything about the event in detail, so it feels like it
786.
by VHRanger 5y ago
Check the blog post - They wanted the same language for implementation and user API - They wanted a single machine, CPU bound runtime where the executable is the model Given these, D and Go make sense - easy to pick up, compiles fast and pe
787.
by VHRanger 5y ago
They should be sparse -- the input to Word2Vec is one-hot encodings of words in the models dictionary
788.
by VHRanger 5y ago
Banks didn't expect to be bailed out. That question has been studied Bank upper management just didn't care that they'd be bailed out or not because their incentives were on fairly short term horizons.
789.
by VHRanger 5y ago
The 2008 bubble was largely driven by the deregulation of financial instruments (and non-regulation of new instruments) in the late 1990 and early 2000.
790.
by VHRanger 5y ago
Don't try to short a fraud. You'll get squeezed.
791.
by VHRanger 5y ago
I've gone across Canada in a NV2500, and it went well but in retrospect I should have gone with a Ford or GMC If you're in the middle of nowhere and something breaks, a Ford or GMC will be easy to get parts for and repair vs anoth
792.
by VHRanger 5y ago
Have you checked for ADHD in your pre-teen that has impulse control issues? It can be eye opening if the behavior has a neurological component, and it's always better to intervene early in these cases
793.
by VHRanger 5y ago
The fact there are vector languages in subsets of python (numpy, pandas, etc.) and R. And these already have great large columnar dataset support (eg. Apache Arrow) And an open source community intent on developing and maintaining the ecosy
794.
by VHRanger 5y ago
Russia tried to execute on their normal "Deep Battle" doctrine where they attack from multiple fronts simultaneously and culminate towards the same point. The problem was execution. Military analysts expected exactly that, but Rus
795.
by VHRanger 5y ago
More data doesn't help if the additional data points don't add information to the dataset. At some point it's better to add features than simply more rows of observations. Arguably text and images are special cases here becau
796.
by VHRanger 5y ago
Network node embeddings are the best for tabular data. I maintain a library on it here, but there's plenty of good alternatives: https://github.com/VHRanger/nodevectors
797.
by VHRanger 5y ago
For sparser data, you should just do normal network node embeddings. Look into node2vec libraries for instance
798.
by VHRanger 5y ago
This had been discussed a lot of times. All of those proposed uses are better served by regular databases When crypto proposes something that *is*, rather than something that *could be* people will stop considering it as a big casino.
799.
by VHRanger 5y ago
Gary Marcus hasn't changed his tune since 2016. Yet a bunch of progress has been made since then! Is DL a reasonable path to strong AI? Probably not. DeepMind is making cool stuff in game playing, but it's still "dumb" A
800.
by VHRanger 5y ago
FWIW this is a blog aimed at statisticians and economists, so intermediate undergraduate statistics is more or less assumed in the authors language
801.
by VHRanger 5y ago
Because checking off features matter more than usability when a manager is the one buying it instead of the person using it Next question
802.
by VHRanger 5y ago
This post is not so subtly pointing the finger at it being a likely false flag attack orchestrated to consolidate power. This American Life #763 goes into much more details: https://www.thisamericanlife.org/763/the-othe
803.
by VHRanger 5y ago
Finance is weaponized insofar that this large web of interdependent economic relationships is extremely beneficial to be a part of. Trade creates giant win-win situations where parties can specialize and use the economic surplus they produc
804.
1999 Russian Apartment Bombings(en.wikipedia.org)
4 points by VHRanger 5y ago|1 comments
805.
Institute for Study of War: Ukraine Asseessment 28/02(understandingwar.org)
2 points by VHRanger 5y ago|0 comments
806.
John Conway: The Free Will Lectures(youtube.com)
3 points by VHRanger 5y ago|0 comments
807.
You shouldn't hire a real estate agent to sell your home(singlelunch.com)
2 points by VHRanger 5y ago|0 comments
808.
by VHRanger 5y ago
There are a few oddly shaped lakes in the inconceivably large northern part of Quebec. One of my favorites is lake mistassini[1], an immense lake shaped like claw marks 1. https://en.wikipedia.org/wiki/Lake_Mistassini
809.
by VHRanger 5y ago
Also pandas
810.
by VHRanger 5y ago
You're aware pandas and most of scipy is cython, right? I like numba, but cython is clearly used more in the popular packages

More ›