plingamp

Specifically when the document number reaches around 10k+, a phenomenon called "Semantic Collapse" occurs.

https://dho.stanford.edu/wp-content/uploads/Legal_RAG_Halluc...

Shared this in another comment, but my luggage was in a locked truck, nothing visible from the back windows. They broke in by smashing the windows, unlocking the door and using the latch to fold the back seats down to expose the trunk.

For what it's worth, everything was in a locked truck with no visible way of seeing any items.

From what I heard from others, apparently the thieves have a device that allows them to detect electronics (I had two laptops, cellphone, and a few other devices). I'm not sure how accurate this is, but i'm not sure why my car was the only one on the street that was targeted as there were no visible signs of valuables in the car (nothing visible from windows etc.) Funny part is a few weeks later nothing was found except for my Kindle which a kind citizen found and returned to me. Apparently thieves don't like to read?

My car got broken into in Oakland, California. Multiple pieces of luggage stolen (yes, my fault for leaving it in the car in the first place). Luckily I had an AirTag that showed the exact location of the stolen items. I called the police but they said they couldn't do anything. Apparently, even if I had the location the thief would have to invite them in. Regardless, I was put on a waiting list, they finally called me back 3 days later. I promptly left the state a few months later.

Sorry not related - your blog is awesome. Cool to see you here on HN!

I’m building PaperDrop it's a research workspace that turns PDFs and new arXiv papers into something you can actually work with (notes, questions, cross paper comparisons). Would love feedback from people who read a lot of papers: https://paperdrop.xyz

It's still early prototype / beta, but wanted to share it anyway!

Indian and Chinese are not languages

According to Crunchbase, Loopt raised $39.1M.

From the github issue referenced in the FAQ, I think they mean that because TensorFlow only natively supports CUDA, TensorFire may outperform TensorFlow on computers that have non Nvidia GPUs, such as the new MacBook Pro.

I may have used the term hyperparameter too loosely. Yes, this project does a good job on taking a configuration first approach, but even they set some defaults. For example, they set relu as their default layer activation function. I haven't had time to see what other such defaults are being set.

In the past, i've found that these higher level libraries built on top of TF are useful for quick model building, but should be used cautiously. By having default hyperparameters it can be easy to blindly build semi-working models without knowing what's happening. I would recommend either reading the associated papers or implementing the models in code (at least once) before using these pre-built models. That being said, i'm really excited by this project! I think it'll save researchers a bunch of time.

I think he maybe referring to this:

Bannon responded: “When two-thirds or three-quarters of the CEOs in Silicon Valley are from South Asia or from Asia, I think . . . ” he didn’t finish his sentence. “A country is more than an economy. We’re a civic society.”

http://www.theverge.com/2016/11/16/13653490/steve-bannon-tru...

I have a old Chromebook that I am using as my primary dev machine (at home). Was hoping to upgrade it to a MBP, but now i'm not sure. I know they are two different classes of machines, but what really struck me in that I can literally buy 10 Chromebooks for 1 MBP.

Hmm, I also have uBlock Origin on, and the website renders perfectly for me (on chrome). I think this is might be a firefox specific issue.

All ML networks are inherently biased towards its creators. My colleague recently described this issue to me as the "Old, white, male" problem. This is why most voice recognition services drastically fail when they are shown foreign accents.

There are so many theories on what is happening, let me propose yet another. Satoshi Nakamoto is akin to the Dread Pirate Roberts. Who knows how many Satoshi's there have been? Some may have died, some may have lost interest, but the torch continues to burn, and must be passed on.

Saving as PDF was not working on OSX Chrome. However, using Safari with "show no footers" option, generated a perfect PDF.

If there is clear evidence that your clothes were ruined, we will work with both parties, and reimburse you for any damages. For more detailed information, email us at contact@laundromatchapp.com

The pricing is $10 per load to do wash/dry/fold and extra $10 if you want pickup/delivery. For example, 2 loads with pickup would be $20 (for laundry) + $10 (for delivery), total of $30.

Hi HN, I'm one of the co-founders of Laundromatch, I would be happy to answer any questions!

I think they have potential, they are very clunky right now. But they have most of the same features at the Microsoft ones. It was really cool to "touch" a 3d object.

How is this different than the Meta AR glasses? I have a pair, and I think they did a great job with the display and interaction.

Yea man! It's my favorite pizza place in NYC :)

No, like you said Grunt/Gulp is useful when you are automating multiple tasks that are done repeatedly. The title of this article is total click bait.

I wish you could combine topics, example: Design + Tech News.

Looks cool, but this is still my favorite theme for Sublime: http://kkga.github.io/spacegray/

Scraping is always such a pain, this looks incredibly well done. I've personally have had a really good experience with PhantomJS. What does your backend look like? (What happens when a API request is made)

Very interesting project! Can you explain what some of the differences are between this library and PassportJS?

Just one (small) design thought, the sidebar has the appearance that it could be hidden. Or maybe i've just been on mobile for far too long. Looks great, nonetheless!

looks awesome! it's always great to see hardware stuff on hn :)