linux Ubuntu local setup
6.5.0-44-generic 22.04.1-Ubuntu
After running command:
conda activate idm
python gradio_demo/app.py
I got such message:
/home/admin/miniconda3/envs/idm/lib/python3.10/site-packages/huggingface_hub/file_download.py:1150: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
warnings.warn(
The config attributes {'decay': 0.9999, 'inv_gamma': 1.0, 'min_decay': 0.0, 'optimization_step': 37000, 'power': 0.6666666666666666, 'update_after_step': 0, 'use_ema_warmup': False} were passed to UNet2DConditionModel, but are not expected and will be ignored. Please verify your config.json configuration file.
But after my laptop runs out of memory and get stuck
Any ideas?
Locally you need at least ~35 GB of RAM to run this model.
Locally you need at least ~35 GB of RAM to run this model.
Thank you for provided information. Interesting, is such information available in any kind of form for such model. or any other model. (At least at my first experience, I haven`t found it, while tried to setup) Is the real practice, that I need determine such characteristics by my own or as best practice, the owners/creators provide such info.
But, if such models utilize so much RAM, I can assume, that running locally such thing is not the most common occurrence.
I remember download of model I think, take a long time. For now I don't remember the downloadable size, this could be an answer.
@cryptoishere try using modal.com you can adapt your code to follow the framework, they offer free $30 so you can run and test it
