Skip to content

Instantly share code, notes, and snippets.

@chavinlo
Created November 25, 2022 04:43
Show Gist options
  • Save chavinlo/5bbf15af89dd66e99c3861ff03e8eba3 to your computer and use it in GitHub Desktop.
Save chavinlo/5bbf15af89dd66e99c3861ff03e8eba3 to your computer and use it in GitHub Desktop.
SO UN REAL
#!/bin/bash
#Install deps
apt-get update -y
apt-get install htop screen psmisc python3-pip unzip wget gcc g++ nano -y
#Install Python deps
wget https://gist.githubusercontent.com/chavinlo/fe8afc02e03d9cc4eb545c4c306c8a73/raw/d9a5ad446fe662dc3e6597163a1f8d5546a8a795/requirements.txt
pip install -r requirements.txt OmegaConf
pip install triton==2.0.0.dev20221120
conda install xformers -c xformers/label/dev
#Download weights
wget https://huggingface.co/hakurei/waifu-diffusion-v1-3/resolve/main/wd-v1-3-full-opt.ckpt
#Download scripts
#trainer script
wget https://gist.githubusercontent.com/chavinlo/3aff2e45c66a50c314c204a31c961ee6/raw/1920337cff138f4aeada12055822b7674ae096cf/diffusers_trainer_nodatasync.py -O train.py
#convert script
wget https://raw.githubusercontent.com/huggingface/diffusers/d9cfe325a53502641f16ce4f839391c5b0d0a684/scripts/convert_original_stable_diffusion_to_diffusers.py
#convert
python3 convert_original_stable_diffusion_to_diffusers.py --checkpoint_path wd-v1-3-full-opt.ckpt --dump_path wdiffuser
echo 'torchrun --nproc_per_node=1 train.py --model wdiffuser --run_name testrun1 --dataset datatrain --batch_size 3 --gradient_checkpointing="true" --use_8bit_adam="false" --fp16="true" --resize="true" --wandb="true" --project_id="hivemind_experiments" --use_xformers="true" --no_migration="true" --image_log_steps 500 --save_steps 1000' > run.sh
touch worked
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment