-
-
Save FurkanGozukara/fdfb1c24a38509c8ec0184951ba12abf to your computer and use it in GitHub Desktop.
video of this gist : https://youtu.be/zA4LksIVas8 | |
dreambooth extension commit id : 3324b6ab7fa661cf7d6b5ef186227dc5e8ad1878 | |
sd web ui commit : 22bcc7be428c94e9408f589966c2040187245d81 | |
fuser -k 3000/tcp | |
python relauncher.py | |
pip3 install torch==1.13.1 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117 | |
pip uninstall xformers | |
pip install https://huggingface.co/MonsterMMORPG/SECourses/resolve/main/xformers-0.0.19-cp310-cp310-manylinux2014_x86_64.whl | |
pip install -r requirements.txt | |
wget https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned.ckpt | |
wget https://huggingface.co/stabilityai/sd-vae-ft-mse-original/resolve/main/vae-ft-mse-840000-ema-pruned.ckpt | |
https://github.com/d8ahazard/sd_dreambooth_extension | |
lora extended: https://github.com/KohakuBlueleaf/a1111-sd-webui-locon | |
Just curious (also having out of memory issues on Runpod with 24GB of vram): the video references v1-5-pruned-ema but your link here is for v1-5-pruned.ckpt. Is that significant or a was there a reason for the change?
edit: issue here I'm having is OOM errors during training right after a sample image is generated (50 steps). If I set "Save Preview(s) Frequency" to 0, training seems to proceed (for now).
Steps: 1%|▋ | 75/7500 [00:25<40:49, 3.03it/s, inst_loss=0, loss=0.234, lr=1e-7, prior_loss=0.311, vram=18.1]
update2: training fails during an attempt to save the model at 500 steps
Just curious (also having out of memory issues on Runpod with 24GB of vram): the video references v1-5-pruned-ema but your link here is for v1-5-pruned.ckpt. Is that significant or a was there a reason for the change?
edit: issue here I'm having is OOM errors during training right after a sample image is generated (50 steps). If I set "Save Preview(s) Frequency" to 0, training seems to proceed (for now).
Steps: 1%|▋ | 75/7500 [00:25<40:49, 3.03it/s, inst_loss=0, loss=0.234, lr=1e-7, prior_loss=0.311, vram=18.1]
update2: training fails during an attempt to save the model at 500 steps
Sorry that I am not at home. This weekend is Eid here holy day. Hopefully I can help you later hour
great please dm me from discord: MonsterMMORPG#2198