Config is 2vCPU, 2 GB RAM and running Debian 7 on 64 bits.
apt-get update
apt-get upgrade
apt-get install htop
Add localadmin user
adduser localadmin
apt-get install sudo
Config is 2vCPU, 2 GB RAM and running Debian 7 on 64 bits.
apt-get update
apt-get upgrade
apt-get install htop
Add localadmin user
adduser localadmin
apt-get install sudo
require 'rubygems' | |
require 'bcrypt' | |
require 'haml' | |
require 'sinatra' | |
enable :sessions | |
userTable = {} | |
helpers do |
echo "If your SSH key has a passphrase, you might need to enter it each time. Consider using ssh-agent or removing the passphrase"; | |
while read p; do | |
echo "$p"; | |
git clone "$p"; | |
done < $1 |
tl;dr use Linux, install bitsandbytes
(either globally or in KAI's conda env, add load_in_8bit=True
, device_map="auto"
in model pipeline creation calls)
Many people are unable to load models due to their GPU's limited VRAM. These models contain billions of parameters (model weights and biases), each of which is a 32 (or 16) bit float. Thanks to the hard work of some researchers [1], it's possible to run these models using 8-bit numbers, which halves the required amount of VRAM compared to running in half-precision. E.g. if a model requires 16GB of VRAM, running with 8-bit inference only requires 8GB.
This guide was written for KoboldAI 1.19.1, and tested with Ubuntu 20.04. These instructions are based on work by Gmin
in KoboldAI's Discord server, and Huggingface's efficient LM inference guide.
<?php | |
// For now file storage requires that the file specified is created already and chmodded to allow writing. | |
/** | |
* Connection info logging script created by Xeru | |
* | |
* Website: https://xeru.me | |
* Twitter: https://twitter.com/Xeru_ | |
* GitHub: https://github.com/exec |
This gist is a note about install nvidia-docker
in Pop!_OS 20.10
. nvidia-docker
is used to help docker containers compute on GPU.
The basic installcation is in Nvidia's offical documentation. But there are a few tweaks to make it work on Pop!_OS 20.10
.
No surprise. Follow the offical documentaion should work.
/* Scale a value from one range to another | |
* Example of use: | |
* | |
* // Convert 33 from a 0-100 range to a 0-65535 range | |
* var n = scaleValue(33, [0,100], [0,65535]); | |
* | |
* // Ranges don't have to be positive | |
* var n = scaleValue(0, [-50,+50], [0,65535]); | |
* | |
* Ranges are defined as arrays of two values, inclusive |
I could not find a proper, detailed (and up-to-date) reverse-engineerment
of Omegle's text chat protocol on the internet, so here, have one made by analyzing the web app (web requests and source code).
The responses are beautified and the query strings split up and URI-decoded for
readability.
Note that "query string" refers to parameters encoded into the URL and
"form data" to parameters in the POST body which do not have to be URI-encoded.
TODO:
@echo off | |
echo Uninstalling KB3075249 (telemetry for Win7/8.1) | |
start /w wusa.exe /uninstall /kb:3075249 /quiet /norestart | |
echo Uninstalling KB3080149 (telemetry for Win7/8.1) | |
start /w wusa.exe /uninstall /kb:3080149 /quiet /norestart | |
echo Uninstalling KB3021917 (telemetry for Win7) | |
start /w wusa.exe /uninstall /kb:3021917 /quiet /norestart | |
echo Uninstalling KB3022345 (telemetry) | |
start /w wusa.exe /uninstall /kb:3022345 /quiet /norestart | |
echo Uninstalling KB3068708 (telemetry) |