Skip to content
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

Jump to bottom

fix pynoise code bug #489

Merged
merged 8 commits into from
May 11, 2023
Merged

fix pynoise code bug #489

Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
e3a5d63
fix pynoise
sdbds May 8, 2023
6d21922
Update custom_train_functions.py for default
sdbds May 9, 2023
c023d50
Update custom_train_functions.py for note
sdbds May 9, 2023
ca79915
Update custom_train_functions.py for default
sdbds May 11, 2023
2780021
Revert "Update custom_train_functions.py for default"
sdbds May 11, 2023
483577e
Update custom_train_functions.py for default
sdbds May 11, 2023
867a812
Revert "Update custom_train_functions.py for default"
sdbds May 11, 2023
cfbcb60
default value change
sdbds May 11, 2023
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
10 changes: 5 additions & 5 deletions library/custom_train_functions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -346,14 +346,14 @@ def get_weighted_text_embeddings(


# https://wandb.ai/johnowhitaker/multires_noise/reports/Multi-Resolution-Noise-for-Diffusion-Model-Training--VmlldzozNjYyOTU2
def pyramid_noise_like(noise, device, iterations=6, discount=0.3):
b, c, w, h = noise.shape
def pyramid_noise_like(noise, device, iterations=6, discount=0.4):
b, c, w, h = noise.shape # EDIT: w and h get over-written, rename for a different variant!
u = torch.nn.Upsample(size=(w, h), mode="bilinear").to(device)
for i in range(iterations):
r = random.random() * 2 + 2 # Rather than always going 2x,
w, h = max(1, int(w / (r**i))), max(1, int(h / (r**i)))
noise += u(torch.randn(b, c, w, h).to(device)) * discount**i
if w == 1 or h == 1:
wn, hn = max(1, int(w / (r**i))), max(1, int(h / (r**i)))
noise += u(torch.randn(b, c, wn, hn).to(device)) * discount**i
if wn == 1 or hn == 1:
break # Lowest resolution is 1x1
return noise / noise.std() # Scaled back to roughly unit variance

Expand Down