Skip to content

Instantly share code, notes, and snippets.

View victorchall's full-sized avatar

Victor Hall victorchall

View GitHub Profile
@victorchall
victorchall / gist:67bc53472f86641aef1ebee1e154f5d1
Created October 11, 2022 18:50
Quick start on caption training with kanewallmann repo
Using this repo:
https://github.com/kanewallmann/Dreambooth-Stable-Diffusion
Folder structure, using a project name of "ff7r" for example but you can name it however you want
/reg/man/ (all your regularization images of men)
/training_samples/ff7r/man (all your images of men to train)
/reg/woman/ (all your regulaization images of women)
/training_samples/ff7r/woman (all your images of women to train)
{
"last_node_id": 65,
"last_link_id": 114,
"nodes": [
{
"id": 4,
"type": "CheckpointLoaderSimple",
"pos": [
-48,
96
@victorchall
victorchall / SD3_nsfw_negative_testing.json
Created June 18, 2024 19:58
SD3 2B "nsfw negative" A/B/C test workflow
{
"last_node_id": 67,
"last_link_id": 131,
"nodes": [
{
"id": 8,
"type": "VAEDecode",
"pos": [
1200,
96
@victorchall
victorchall / tarup.py
Created June 30, 2024 02:26
Python script to create 1GB tar files for webp/txt/json tuple dataset
import os
import tarfile
import json
from collections import defaultdict
def group_files(root_dir):
file_groups = defaultdict(list)
for subdir, _, files in os.walk(root_dir):
for file in files:
if file == "urls.txt" or file == "caption_cog_params.txt":