Skip to content

Add v-pred support for SDXL train #1505

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
merged 1 commit into from
Aug 24, 2024
+6 −2
Merged

Add v-pred support for SDXL train #1505

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
8 changes: 6 additions & 2 deletions sdxl_train.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -590,7 +590,11 @@ def fn_recursive_set_mem_eff(module: torch.nn.Module):
with accelerator.autocast():
noise_pred = unet(noisy_latents, timesteps, text_embedding, vector_embedding)

target = noise
if args.v_parameterization:
# v-parameterization training
target = noise_scheduler.get_velocity(latents, noise, timesteps)
else:
target = noise

if (
args.min_snr_gamma
Expand All @@ -606,7 +610,7 @@ def fn_recursive_set_mem_eff(module: torch.nn.Module):
loss = loss.mean([1, 2, 3])

if args.min_snr_gamma:
loss = apply_snr_weight(loss, timesteps, noise_scheduler, args.min_snr_gamma)
loss = apply_snr_weight(loss, timesteps, noise_scheduler, args.min_snr_gamma, args.v_parameterization)
if args.scale_v_pred_loss_like_noise_pred:
loss = scale_v_prediction_loss_like_noise_prediction(loss, timesteps, noise_scheduler)
if args.v_pred_like_loss:
Expand Down
Loading