Open joowon-dm-snu opened 2 years ago
I did
pip install pytorch-lightning==1.5
and it works fine.
pip install pytorch-lightning==1.6.1
for me
This worked for me (although I needed to re-install pytorch with cuda too), but now txt2img.py gives me an Entry Point Not Found error for torchvision_C.pyd , as seen here: https://github.com/CompVis/latent-diffusion/issues/72#issuecomment-1138729437
I did
pip install pytorch-lightning==1.5
and it works fine.
This worked for me, but only after removing the conda environment, changes the yaml to pytorch-lightning==1.5, and creating the conda environment based on the new yaml.
Created new conda env here with pytorch-lightning==1.6.1
and it seems to work, after running conda env remove -n ldm
and recreating the env with new .yml as pointed out by @Tuxius
Thanks for great research, I found out conda env settings error while using scripts below.
Error Log
Problem was
pytorch-lightning==1.4.2
automatically importsfrom torchmetrics.utilities.data import get_num_classes as _get_num_classes
but that function was dropped by this PR.So yaml file should be changed by updating
pytorch & torchvision & pytorch-lightning
or add explicittorchmetric
versionI solved it by updating pytorch-lightning.