You CAN'T ignore the extra information like P(Rain|Seattle) though. What if this were the question instead: "Three friends in Seattle told you that Earth exploded and everyone died. Each has a probability of 1/3 of lying. What’s the probability that the Earth exploded and everyone died?"
I wonder how accurate these surveys are. In high school, kids I knew who actually did drugs wouldn't admit to it (even on an anonymous survey), and some other kids would answer that they do meth every day because it sounded funny to write that.
Hi, I have a question about the performance benchmarks section. The best performaing model from your benchmarks, ssd_mobilenet_v1, has a prediction time in seconds of 0.72 -- Is that the total runtime of the script? I'm wondering if I could achieve ~1 FPS running in realtime (basically looping as fast as possible against the camera input), or is there more overhead?
(edit -- made question more specific)
Disclaimer: The following suggestion only works if you only READ from their DB. Can you roll your own DB that matches theirs schema-wise for the tables you need, but doesn’t suck in terms of availability? If you can get that set up, then in places where there are writes to their DB that you care about, also write to yours. Do a one-shot migration to copy data you care about after you land your change to double write. Then, modify your app to read from your DB and ignore theirs. Finally, once everything has been running smoothly for a while, collect data showing your system is better than theirs, get them fired, and become the new DBA for your company :)
It's been a while since I've used C#, but I used to use tuples when I wanted simple composite-key dictionaries -- typically grouping things up for batch processing, etc. Before LINQ's convenient multiple-key GroupBy and other stuff, you could emulate with Dictionary<Tuple<T1,T2>,T3>.
I feel like this is basic economics... money, like most things, has diminishing marginal utility.
From a common sense standpoint, it's not so much that having money makes you happy, but having to worry about money definitely can make you unhappy.
Probably bad analogy: Eating a meal that I'd rate 9/10 doesn't make me much happier than a meal that's 8/10, but having to worry about whether I'm going to be able to eat today at all would definitely make me sad.
It's different because you can set your profile privacy to "Friends Only" so prospective employers can't look at your info. Facebook, however, has access to all that data and who knows if/how they might discriminate (doubtful it would be explicit/intentional, but it's easy to accidentally let bias creep into machine learning algorithms).
This is the sort of thing that's really neat technologically but also scares me -- what happens when [insert homophobic government here] starts to use this sort of thing to persecute people?
This is super niche, but a service that takes a video upload (or youtube link) of a professional pool/billiards match and generates overhead shot diagrams for every shot of the match. Why do I want this? Diagramming shots is the most tedious part of making a book where you analyze shots from pro matches -- this has been done by hand before and sells for ~$50 --
https://www.amazon.com/Capelle-9-Ball-Archer-vs-Reyes/dp/096... . I've played with this a bit, taking photos of a table, doing transforms to get an overhead layout (with OpenCV perspective transforms and Hough detection), but it's never come out quite right. I also want motion tracking of balls so you can generate arrows for where the balls went, but honestly, just the static "balls at rest" diagrams would be worth it.
You should consider flying to Japan/Korea to get Toric ICL if it's an option. It's not FDA approved for the US yet, but I've had friends that did it, and not one of them have any regrets. It's been a couple years since, so I wouldn't be surprised if they had even newer/better options available.
Well yeah, in practice, we can say an asset's value is what the buyer and seller agreed upon for the price, but I'm asking about valuing it while it's being held for taxation purposes. This is easy for stocks, etc, which are homogenous units that have other sales going on all the time, but for a house, for example? Comparisons are hard in practice! No house I've bought/sold has come in at the government's valuation.
Not sure what tablet/browser you're using, but several have extensions to create a whitelist of sites to always open links in new tabs. I haven't tried this one in particular, but Chrome has an "Open in new tab" extension where the description actually says "Useful for websites such as hackernews where you would ideally like to open links in new tab."
Tooling is getting better for non-Lisp languages. In C#, with Visual Studio and ReSharper, for example, you could highlight the "int x = 1;" line, and hit your keyboard shortcut for ReSharper:Introduce_Parameter, which would do pretty much exactly what you're saying. Same for turning a chunk of code into a method, etc.
This is an excellent read. I would suggest reading this also (maybe even before), as it provides one of the simplest (incomplete, but still useful) regex implementations you'll ever see: https://www.cs.princeton.edu/courses/archive/spr09/cos333/be...
It suffers from the problems remedied by the original submission (backtracking instead of DFA), but the code is shorter and simpler. They're great to compare.
From the readme:
"Relaxo is designed to scale to the hundreds of thousands of documents. It's designed around the git persistent data store, and therefore has some performance and concurrency limitations due to the underlying implementation...
Relaxo can do anywhere from 1000-10,000 inserts per second depending on how you structure the workload."
Yes, there are several legacy systems where a full rewrite is way more risky than maintenance-and-minor-enhancements-only. If you're surprised that Perl is still alive, don't go asking about how many COBOL systems are still out there!