In the name of God
This gist contains steps to setup Ubuntu 22.04
for deep learning.
# avoid setting the timezone during installation in a docker | |
apt update | |
DEBIAN_FRONTEND="noninteractive" | |
TZ="Europe/Berlin" | |
ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone | |
apt install python3.9 -y |
['Nigeria: Director General National Emergency Management Agency (NEMA) flags off Distribution of Relief items and livelihood support items in Jigawa state', '412 words', '25 September 2023'] | |
{"primary_classification":"H", "event_type":"NNC" } | |
['Rescued from the earthquake in Morocco, midwife Maria delivers babies and saves lives as catastrophe unfolds around her', '629 words', '25 September 2023'] | |
{"primary_classification":"G", "event_type":"EQ"} | |
['Two Qatari Planes from Qatari Air Bridge Arrive at Benina Airport in Libya to Aid Flood-Hit People', '192 words', '25 September 2023'] | |
{"primary_classification":"H", "event_type":"F"} |
# to use this, first install python and exllamav2 (https://github.com/turboderp/exllamav2) | |
# load a model, rearrange the layers as you like, set generation parameters, and chat with it | |
# duplicate layers need no extra memory | |
# WARNING: duplicate layers share the same cache, even though they shouldn't. | |
# This makes the model even more demented than a frankenmerge - use at your own risk. | |
# public domain - Silphendio | |
import random | |
import sys |
# to use this, first install python and exllamav2 (https://github.com/turboderp/exllamav2) | |
# load a model, rearrange the layers as you like, set generation parameters, and chat with it | |
# duplicate layers need no extra memory | |
# WARNING: duplicate layers share the same cache, even though they shouldn't. | |
# This makes the model even more demented than a frankenmerge - use at your own risk. | |
# public domain - Silphendio | |
import copy | |
import math | |
import random |