dreness

#!python

"""
Configure the arguments to the process_directory call at the bottom.
Start additional instances of this script until your hw is full. If you
have multiple GPUs, you're responsible for setting CUDA_VISIBLE_DEVICES.
Using the 'medium' model, a V100 is mostly utilized with two instances.

This still has some bugs probably...
"""

dreness

/*
 By Andre LaBranche, with extensive help from your friend and mine, ChatGPT.
 In other words, assume this code eats babies.

 This program attempts to be a reasonably inexpensive replacement for
 /usr/bin/pidpersec.d in macOS, which still ships, but is broken by default
 (i.e. when SIP is enabled).

 compile with: clang -o pidpersec pidpersec.c
 */

dreness

#!python -u

import subprocess
import plistlib
import sys

"""
Print selected power metrics from /usr/bin/powermetrics at intervals, suitable
for graphing.
"""

dreness

#!/usr/bin/env python3

# Instructions:
# 1) install svgpathtools with the command:
#    pip install svgpathtools
# 2) Run this script with no arguments:
#    python keynote_shape_to_svg.py

from svgpathtools import parse_path, svg2paths, wsvg
import sqlite3

dreness

#!/usr/bin/env python
# dre at mac dot com
# Graphical and statistical analysis of frame rate in a video
# Requires ffmpeg and gnuplot

# Example output:
# https://youbeill.in/scrap/Screen-Recording-2018-09-07-at-3.17.33-PM.mov.png

from sys import platform
from subprocess import check_output, DEVNULL

dreness

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import json  # Native JSON Library
import argparse  # Native Argument parser Library

# This is a slightly updated version of a script found on ye cyberwebs to
# add batch conversion support. Just don't specify an image ID to export the world.

# ==============================#

dreness

Algodoo

... is pretty great. If you somehow arrived here without knowing what Algodoo is, I implore you to go try it right now.

I had a hard time finding good examples of Thyme code on ye interwebs, and this document tries to remedy that. I downloaded a bunch of public compositions by the talented matto and then made a Jupyter notebook to extract interesting Thyme snippets from the project files (which are just zip archives after all). For this purpose, I define "interesting" as:

dreness

# -*- coding: utf-8 -*-
from __future__ import print_function, unicode_literals

import os
import socket
from datetime import datetime
from urllib import parse
from klein import Klein
from twisted.web.static import File
from twisted.python.filepath import FilePath

dreness

#!/bin/zsh
#
# Batch transcribe audio files with whisper.cpp and ffmpeg
# This script wants:
# - ffmpeg installed and accessible via $PATH
# - whisper.cpp built locally:
#       git clone https://github.com/ggerganov/whisper.cpp
#       cd whisper.cpp
#       make
#   The compiled binary will be called 'main' in the whisper.cpp directory.

dreness

#!/usr/bin/env python
import numpy as np

p1 = "portrait of a robot cyberpunk cat,"
p2 = "a cyborg steampunk dog,"
common = "octane render, 8k, high detail, global illumination, artstation, [out of frame]"
blendSteps = 300

f = open('prompts.txt', 'w')