arianvp

#!/bin/sh

#If already set (e.g. in github actions); use that instead
if [ -z "$GITHUB_TOKEN" ]; then
  nix run nixpkgs#gh auth status || nix run nixpkgs#gh auth login
  GITHUB_TOKEN="$(nix run nixpkgs#gh auth token)"
  export GITHUB_TOKEN
fi

NIX_CONFIG="access-tokens = github.com=$GITHUB_TOKEN"

fmateo05

Kazoo 2-zone cluster using Incus Containers; Open Virtual Network (OVN); Nebula Updated

This guide described how to install Kazoo with Incus containers and some other components like OVN networking and Nebula, etc.

• Create 4 instances on Digital ocean; 2 on one datacenter (zone 100 ie. San Francisco) and the other 2 on another datacenter (zone 200 ie. New York).

The idea is like 2 instances per zone or datacenter as desired; if you later would like to create another zone. The main Linux distribution is Rocky Linux 9

• Login to each server using ssh with private key

jart

#!/bin/sh
# rename-pictures.sh
# Author: Justine Tunney <jtunney@gmail.com>
# License: Apache 2.0
#
# This shell script can be used to ensure all the images in a folder
# have good descriptive filenames that are written in English. It's
# based on the Mistral 7b and LLaVA v1.5 models.
#
# For example, the following command:

Natyren

import torch
from PIL import Image
import open_clip
from transformers import CLIPImageProcessor, CLIPTokenizer, CLIPConfig, CLIPModel


config = CLIPConfig.from_pretrained("/Users/georgebredis/Downloads/MetaCLIP_b32_400m")
model = CLIPModel(config).eval()

_, _, preprocess = open_clip.create_model_and_transforms(

hawkeye217

# This script can help you determine if your PTZ is capable of
# working with Frigate NVR's autotracker.
#
# Cameras with a "YES" printed for each parameter at the end of
# the output will likely be supported by Frigate.
#
# Make sure you're using python3 with the onvif-zeep package
# Update the values for your camera below, then run:
# pip3 install onvif-zeep
# python3 ./fovtest.py

roib20

---
- hosts: localhost
  connection: local
  gather_facts: true

  tasks:
    - name: Add APT repositories
      when: ansible_os_family == 'Debian'
      become: true
      block:

cedrickchee

4 Steps in Running LLaMA-7B on a M1 MacBook

The large language models usability

The problem with large language models is that you can’t run these locally on your laptop. Thanks to Georgi Gerganov and his llama.cpp project, it is now possible to run Meta’s LLaMA on a single computer without a dedicated GPU.

Running LLaMA

There are multiple steps involved in running LLaMA locally on a M1 Mac after downloading the model weights.

JoaoLages

Maybe you've heard about this technique but you haven't completely understood it, especially the PPO part. This explanation might help.

We will focus on text-to-text language models 📝, such as GPT-3, BLOOM, and T5. Models like BERT, which are encoder-only, are not addressed.

Reinforcement Learning from Human Feedback (RLHF) has been successfully applied in ChatGPT, hence its major increase in popularity. 📈

RLHF is especially useful in two scenarios 🌟:

• You can’t create a good loss function
  • Example: how do you calculate a metric to measure if the model’s output was funny?
• You want to train with production data, but you can’t easily label your production data

BlueHippoGithub

version: '3.3'
networks:
  caddy:
services:
  portainer:
    image: portainer/portainer-ce:latest
    container_name: portainer
    restart: unless-stopped
    security_opt:
      - no-new-privileges:true

bsag

These files illustrate the way that I have set up Nix and Home Manager for standalone use on macOS on both an Intel and Apple Silicon machine, in case the examples are useful to anyone. Some of the configuration is separated out into separate files in configs/ which get imported in home.nix. Some of my existing dotfiles are placed in dotfiles/ and then symlinked in by Nix to the location that that the system expects to find them.

It is a work in progress so currently I am handling installation of R separately.