Bla bla bla…
Model | #Total Params | #Activated Params | Context Length |
---|---|---|---|
DeepSeek-V3-Base | 671B | 37B | 128K |
DeepSeek-V3 | 671B | 37B | 128K |
Bla bla bla…
Model | #Total Params | #Activated Params | Context Length |
---|---|---|---|
DeepSeek-V3-Base | 671B | 37B | 128K |
DeepSeek-V3 | 671B | 37B | 128K |
What’s wrong? Cat got your token?
I’d have answered the same. Burning so much energy to know what date it is deserves a snarky response.
It’s easy to overlook with the omnipresent internet, but self-hosting doesn’t require internet. You could host for your fellow students on the local network. If that’s also against the Wifi rules you can either ignore that stupid rule or set up your own god damn wifi with hostapd on your machine and let students connect directly to it. It’s probably best to use a machine dedicated to the task for security reasons as you wouldn’t want curious students to accidentally erase your homework. I wouldn’t use containers or VMs for any of this, I’d just use bare metal like in the good ol’ days. You could also, without having to worry, give people shell accounts because it’s a closed network. The options are endless without all the worries of hosting on the internet.
WAT!? No internet!?
Reminded my of what happened at the MindTheTech conference half a year ago.
https://peervideo.club/w/p/i4BetLY7RZa5yeNLJriXPW?playlistPosition=3
Netscape Communicator, Netscape Communicator, KHTML, Netscape Communicator
It seems that we focus our interest in two different parts of the problem.
Finding the most optimal way to classify which images are best compressed in bulk is an interesting problem in itself. In this particular problem the person asking it had already picked out similar images by hand and they can be identified by their timestamp for optimizing a comparison of similarity. What I wanted to find out was how well the similar images can be compressed with various methods and codecs with minimal loss of quality. My goal was not to use it as a method to classify the images. It was simply to examine how well the compression stage would work with various methods.
Wait… this is exactly the problem a video codec solves. Scoot and give me some sample data!
I was not talking about classification. What I was talking about was a simple probe at how well a collage of similar images compares in compressed size to the images individually. The hypothesis is that a compression codec would compress images with similar colordistribution in a spritesheet better than if it encode each image individually. I don’t know, the savings might be neglible, but I’d assume that there was something to gain at least for some compression codecs. I doubt doing deduplication post compression has much to gain.
I think you’re overthinking the classification task. These images are very similar and I think comparing the color distribution would be adequate. It would of course be interesting to compare the different methods :)
The first thing I would do writing such a paper would be to test current compression algorithms by create a collage of the similar images and see how that compares to the size of the indiviual images.
Desktop Applications
The quote is a derivative of something Bjarne Stroustrup said himself¹.
C makes it easy to shoot yourself in the foot; C++ makes it harder, but when you do it blows your whole leg off
Intervene? I don’t think they have that kind of power ;)
Thank you.
Interesting that they’ll make it a user choice. Who would answer yes?
On 22 July 2024, Google announced that it is changing its approach to Privacy Sandbox. Instead of removing third-party cookies from Chrome, it will be introducing a user-choice prompt, which will allow users to choose whether to retain third party cookies.
Do you have a source for that excus… uehm… claim?
That’s the last stage of being a FOSS developer.
First thing I would ask the ISP to open the port. I’ve done that without problems before.
If that’s for some reason not a solution, I would, because I’m personally not very attracted to the idea of routing my selfhosting traffic though thirdparties, setup a simple static page with
<meta http-equiv="refresh" content="0;url=https://web.domain.tld:8080/" />
, somewhere and point the bare domain and www subdomain to that page and have it redirect to, like in this example, aweb
subdomain with the port number.As a last remark, I personally would not find it problematic for a different port number to be part of the host scheme and also note that most web traffic now goes to 443 and not 80 because it’s https.
Happy selfhosting!