263 karma · joined January 8, 2021
> This is the same IP address: 3232271615. You get that by interpreting the 4 bytes of the IP address as a big-endian unsigned 32-bit integer, and print that. This leads to a classic parlor trick: if you try to visit http://3232271615 , Chrome will load http://192.168.140.255.
murkans have been using ten digits as phone numbers for some decades now (no country code), I'm kinda bummed there isn't some saudi royalty paying four billion for a set of matching ip address and phone number.
The prepare/train/generate instructions in the github linked are pretty much it for the 'how' of training a model. You give it a task and it does it for 1 billion trillion epochs and saves the changes incrementally (or not).
Training a LoRA for an image model may be more approachable, there's more blog entries etc on this, and the process is largely similar, except you're doing it for a single slice instead of the whole network.
[edit] I'm also learning so correct me if I'm off, hn!
Sorry for your loss.
It's a false sense of urgency thing. The kids just grow calloused to it, culminating in e.g. completely ignoring the door bell.
I had a friend that decided to essentially go off the grid around 2000-2005, he's my age (gen-z), I remember him showing me a website he was developing with javascript and IE.
Now he's asking me to help with the Google Play / Mac App stuff because it makes no sense to him. It isn't an age thing.
It's also my fault when I search google images for a plastic jumper, I get molex connectors, I guess.
It's also my fault that when I get tired of not finding what I need, the last thing I clicked on gets shown to the next person.
I somehow don't think I'm at fault here.
[edit] 20 jpegs into a 150MB LoRA
Props to the webmaster (do people still say that?), I love seeing easter eggs like this.
This is the whole magic, the rest of the workflow doesn't need to unload and flush memory, causing big delays for jobs.
NYT makes for great papier mache though, nice pictures, but that's aboooout it, for over a hundred years already somehow.
Desktop class, you can render AI stuff locally on a $1,500 computer. Phones and tablets are building in neural processors. That will change. But the licensing thing won't change back.
Glad Adobe waited for the founder to pass before doing all this junk, it'd probably kill him.