convoluted connor @notwa@witches.town

ah there we go
"\e[1m" -> "\e[36m"

huh, some color code functionality got fixed in mintty. one of my scripts was actually relying on it, but now i realize the old behavior didn't make any sense. now i'm trying to figure out how i wanna update my script.

i remember spending like 6 hours just micro-optimizing one track in nitronic rush against my own ghosts. i loved seeing all the subtle differences in each ghost and taking note of each to get the absolute best possible time.

my favorite part of nitronic rush and distance is simply just being able to crank the number of ghosts up to 20

got fog? https://witches.town/media/tXwxIUgV5cdhO1wEay8

oh yeah, here's something from my trackmogrify adventures the other night; probably the most points i'll ever get in one "trick" https://witches.town/media/dJ8DcvZ8p43aZaPg0hk https://witches.town/media/iDdk7dJIQdz5NE3mY44

if you're interested, the thing i use is based on this: https://github.com/Disassembler0/Win10-Initial-Setup-Script

finally got the fall update forced on my desktop, thank god for powershell scripts for disabling all the crap it adds and re-enables. that's still 2 gigs i'm not getting back though.

there's so many lasers in this trackmogrify track that they don't even all render: insanelyhazardousinsanelylaserinsanelyextremedownhillzebrafog

wow uh
the manor map is totally broken. i keep falling through the floor after the intro section. is this happening to you too, @Skirmisher ? does anyone else play this game?

you already know what it isssssssssssss

(or maybe you don't, but i find this expression too amusing) https://witches.town/media/nPgVK_P4cDoXHshpk_0

tempted to write an autohotkey script that outright prevents me from typing "though"

I kinda tune out when papers start talking about hessians in deep learning.

yume nikki is on steam now i guess

i believe i've implemented the optimizer described in: https://arxiv.org/abs/1712.03298
it seems to have comparable performance to Nesterov momentum with gradient clipping, which is my usual go-to when Adam doesn't work.

i need to stop using 'testing' and 'experimenting' interchangeably when i'm programming

just sci-hub'd the fuck out of a paper for the first time, feels good

had an intricate and nonsensical dream. today's song playing in my head as i woke up is nomad by dfa1979

oh i forgot a few points about the graph: this is fashion mnist with a 3 layer FC network (no convolution), and validation data is only measured once per epoch. the batch performance is slightly different than actual training data performance, but it's ~close enough~