brayevalerien

Resolution guide for image and video generation

Caution

This is very much a WIP. Not all models or resolutions are listed yet, and there are placeholder "TODO" scattered in the document. It will be updated as soon as possible.

Introduction

This documents serves as a reference cheatsheet for choosing the right resolution for the most common image and video generation AI models.

While most visual generation models support any resolution in theory, their training procedure constrains the resolution of generated images / videos to certain values only. They are the resolutions the models were trained on and that are natively supported, meaning the generation quality will be optimal.

brayevalerien

{
  "id": "monokai_classic",
  "name": "Monokai Classic",
  "colors": {
    "node_slot": {
      "BOOLEAN": "",
      "CLIP": "#e2e22e",
      "CLIP_VISION": "#56adbc",
      "CLIP_VISION_OUTPUT": "#ac6218",
      "CONDITIONING": "#cf876f",

brayevalerien

Title	Category	Original Chinese	English Translation
Soccer Scoring	Sports	一名足球运动员在球场上带球奔跑，快速传球，然后射门得分。镜头聚焦于运动员的脚下动作和射门瞬间。	A soccer player runs with the ball on the field, quickly passes, and then shoots to score. The camera focuses on the player’s footwork and the moment of the shot.
Basketball Shot	Sports	篮球运动员运球过人，跳起投篮，篮球飞入篮筐，运动员庆祝进球。画面只专注于这一个投篮动作。	A basketball player dribbles past opponents, jumps for a shot, the ball flies into the hoop, and the player celebrates the score. The scene focuses solely on this one shooting action.
Tennis Serve	Sports	网球选手准备发球，力量十足地将球击出。画面聚焦在发球的瞬间，球飞越球网，落向对手场地。	A tennis player prepares to serve, hitting the ball with full force. The shot focuses on the serving moment as the ball flies over the net and lands in the opponent’s court.
Runner Sprint	Sports	一位跑步者在跑道上快速奔跑，镜头跟随他的步伐，展示腿部肌肉的紧张和跑步的速度感。	A runner sprints on a track, with the camera following his steps, showcasing

brayevalerien

import argparse
import os

from PIL import Image
from PIL.Image import Resampling


def get_args():
    parser = argparse.ArgumentParser(
        "Bucketize",

brayevalerien

import os
import argparse
import numpy as np
from PIL import Image
from tqdm import tqdm


def add_gaussian_noise(image: Image, factor: float):
    """
    Add a given amount of white guassian noise to an image.

brayevalerien

transformer.transformer_blocks.0.norm1.linear
transformer.transformer_blocks.0.norm1_context.linear
transformer.transformer_blocks.0.attn.to_q
transformer.transformer_blocks.0.attn.to_k
transformer.transformer_blocks.0.attn.to_v
transformer.transformer_blocks.0.attn.add_k_proj
transformer.transformer_blocks.0.attn.add_v_proj
transformer.transformer_blocks.0.attn.add_q_proj
transformer.transformer_blocks.0.attn.to_out.0
transformer.transformer_blocks.0.attn.to_add_out

brayevalerien

echo off
REM Please keep up to date with https://gist.github.com/brayevalerien/0086cbb5123ceed661d7092b3dc221ae

set CONDAENV=comfyui
set COMFYPATH=C:\Users\AIXXX\comfyui\main.py

echo ====== Starting ComfyUI from %cd% ========

echo [ComfyUI Launcher] Activating Conda environment...
call conda activate %CONDAENV%

brayevalerien

plot_settings = {
    'ytick.labelsize': 16,
    'xtick.labelsize': 16,
    'font.size': 22,
    'figure.figsize': (10, 5),
    'axes.titlesize': 22,
    'axes.labelsize': 18,
    'lines.linewidth': 2,
    'lines.markersize': 3,
    'legend.fontsize': 11,

brayevalerien

import random
import numpy as np

class MarkovChainGenerator:
    def __init__(self, order=1, length=10):
        self.order = order
        self.length = length
        self.transitions = {}

    def fit(self, X):

brayevalerien

import random

def print_list(list_to_print: list, title: str = None, bullet_point: str = "•", max_items: int = None, randomize: bool = False, indent_level: int = 0) -> None:
    """
    Prints a list nicely. Handles sublists. Tupples will be displayed as items: ('e1', 'e2', ...).

    Author: Valérien BRAYE (@brayevalerien on github)

    Args:
        list_to_print (list): the list to be printed