Skip to content

use model.to(device) to instead of model.cuda()#142

Open
huo-ju wants to merge 1 commit intoCompVis:mainfrom
huo-ju:main
Open

use model.to(device) to instead of model.cuda()#142
huo-ju wants to merge 1 commit intoCompVis:mainfrom
huo-ju:main

Conversation

@huo-ju
Copy link
Copy Markdown

@huo-ju huo-ju commented Aug 31, 2022

load_model_from_config may use cpu to eval the model,
return the model and call eval() after the model.to(device)
can solve this problem.

load_model_from_config may use cpu to eval the model,
return the model and call eval() after the model.to(device)
can solve this problem.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant