Skip to content

When using --device to select other GPUs, img2img will stuck in sleep(1) loop #166

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
basujindal merged 1 commit into from
Sep 15, 2022
Merged

When using --device to select other GPUs, img2img will stuck in sleep(1) loop #166

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 7 additions & 7 deletions optimizedSD/optimized_img2img.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -253,9 +253,9 @@ def load_img(path, h0, w0):
init_latent = modelFS.get_first_stage_encoding(modelFS.encode_first_stage(init_image)) # move to latent space

if opt.device != "cpu":
mem = torch.cuda.memory_allocated() / 1e6
mem = torch.cuda.memory_allocated(device=opt.device) / 1e6
modelFS.to("cpu")
while torch.cuda.memory_allocated() / 1e6 >= mem:
while torch.cuda.memory_allocated(device=opt.device) / 1e6 >= mem:
time.sleep(1)


Expand Down Expand Up @@ -302,9 +302,9 @@ def load_img(path, h0, w0):
c = modelCS.get_learned_conditioning(prompts)

if opt.device != "cpu":
mem = torch.cuda.memory_allocated() / 1e6
mem = torch.cuda.memory_allocated(device=opt.device) / 1e6
modelCS.to("cpu")
while torch.cuda.memory_allocated() / 1e6 >= mem:
while torch.cuda.memory_allocated(device=opt.device) / 1e6 >= mem:
time.sleep(1)

# encode (scaled latent)
Expand Down Expand Up @@ -340,13 +340,13 @@ def load_img(path, h0, w0):
base_count += 1

if opt.device != "cpu":
mem = torch.cuda.memory_allocated() / 1e6
mem = torch.cuda.memory_allocated(device=opt.device) / 1e6
modelFS.to("cpu")
while torch.cuda.memory_allocated() / 1e6 >= mem:
while torch.cuda.memory_allocated(device=opt.device) / 1e6 >= mem:
time.sleep(1)

del samples_ddim
print("memory_final = ", torch.cuda.memory_allocated() / 1e6)
print("memory_final = ", torch.cuda.memory_allocated(device=opt.device) / 1e6)

toc = time.time()

Expand Down