ninja3925

8 ms·

1.
by ninja3925 3mo ago
Out of curiosity, how was it discovered? You would have to look for it to find this linear combination.
2.
by ninja3925 5mo ago
These are extremely high prices. In Texas, we pay 12-15c.
3.
by ninja3925 6mo ago
So i assumed it would get crushed by OPQ (which requires training)
4.
by ninja3925 6mo ago
The story is actually quite interesting. The Serbs observed that a nighthawk would routinely fly the same route but their radar couldn’t lock on it unless the missile hatch were open, which they managed to elicit. In short, it took 2 rare e
5.
by ninja3925 6mo ago
These signs are mostly on the highway
6.
by ninja3925 10mo ago
This seems to be incorrect. The correct way to combine measurements with various degree of precision is to use the inverse variance weighting law
7.
by ninja3925 10mo ago
Large cloud providers could offer that solution but then, crawlers can also change cycle IPs
8.
by ninja3925 1y ago
Can you expand? What is “poison” referring to? Surely, burning coal as Germany’s current pace can’t be seen as a success, can it?
9.
by ninja3925 1y ago
This is not how it works. The interviewer knows 1-2 problems and there is no time for profiling since they are rushing through their day, probably focused on their day to day work. You are the least of their concern, believe me. Source: we
10.
by ninja3925 1y ago
Very likely.
11.
by ninja3925 1y ago
> Just look at health insurance companies for a prime example: they make profit by denying claims Despite being from Europe, I find this to be a shocking and erroneous interpretation. Clearly, health insurances have the duty to allocate
12.
by ninja3925 1y ago
Note that it's not done for performance reason but rather to generate clear feature maps.
13.
by ninja3925 2y ago
I are
14.
by ninja3925 2y ago
Interestingly, FAISS does exactly that before doing Product Quantization and it works very well (errors are much lower compared to no rotation). They call it “optimal PQ”. During training time, they iterate to find a good candidate and save
15.
by ninja3925 2y ago
The advised reader could devise a solution using public / private key where: - Alice communicates her public key - Bob communicates his - Alice encoders her cards with Bob’s public key - Bob does the same Problem solved!
16.
by ninja3925 2y ago
I actually think war driving progress is a strong argument. Not to use any obvious examples (e.g. internet), I would argue war has a profound effect on the psyche of nations (whether they lose or win). A few examples: - Germany rose to powe
17.
by ninja3925 2y ago
This is far fetched. It’s a very thoughtful article.
18.
by ninja3925 2y ago
If you save 32x memory with binarization, why not do a projection to a larger dimension? Say 4096 for instance. Could this actually improve performance WHILE reducing memory?
19.
by ninja3925 2y ago
> a month. No, It says a week in the article. “Each aircraft takes about a week to have its AeroSHARK film applied, which requires high-precision workmanship from our personnel.”
20.
by ninja3925 2y ago
It’s mind boggling how a country can be so rich as to spend millions of manhours on a project and never use it. Maybe we have become too rich as a species.
21.
by ninja3925 2y ago
Column misnaming/typo is indeed a problem in pandas. I think a powerful IDE could do the trick though.
22.
by ninja3925 2y ago
I wish the article made an attempt at providing a benchmark. I imagine that TSMC has a lot of prime age families so you can’t compare it against the overall population. Without it, one cannot make any inference in whether it’s low or high,
23.
by ninja3925 2y ago
And the military: https://www-users.cse.umn.edu/~arnold/disasters/patriot.html
24.
by ninja3925 2y ago
The ability for the government to spend money it doesn’t have doesn’t cease to amaze. The deficit was $1.6T last year (over 5% of the GDP) but 2024 is starting strong.
25.
by ninja3925 2y ago
This looks quite helpful actually. Thanks for sharing.
26.
by ninja3925 3y ago
It can be done for performance reason I assume. Numpy for instance does that: https://numpy.org/doc/stable/user/basics.types.html
27.
by ninja3925 3y ago
Thanks for posting this. I work in data science and I haven’t seen anybody claiming that PCA, an unsupervised technique, is a good classification (supervised) technique. I think this is the framing that the author is pursuing. That said, so
28.
by ninja3925 3y ago
Why does the US have to get involved? Wouldn’t Arab countries, be more suited to conduct such operations?
29.
by ninja3925 3y ago
True. Also, you are probably doing it wrong by turning a matrix to matrix multiplication into a for loop (over rows). The optimal solution results in better performance sim = np.vstack(df.col) @ vec
30.
by ninja3925 3y ago
We use CLIP internally (large US tech company) and it works very well at a large scale

More ›