Skip to content

Dreambooth: fix #1566: Use autocast to correct full vs half precision error #1575

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

Closed
DrewWalkup wants to merge 4 commits into from
Closed

Dreambooth: fix #1566: Use autocast to correct full vs half precision error #1575

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
c5d4e87
Use autocast to correct precision error
DrewWalkup Dec 6, 2022
03acabd
Fixed code quality check
DrewWalkup Dec 6, 2022
2a0e74b
Remove comments
DrewWalkup Dec 6, 2022
49d8a1b
Remove code comment
DrewWalkup Dec 6, 2022
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
5 changes: 4 additions & 1 deletion examples/dreambooth/train_dreambooth.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,8 @@
import torch.utils.checkpoint
from torch.utils.data import Dataset

from torch.cuda.amp import autocast

from accelerate import Accelerator
from accelerate.logging import get_logger
from accelerate.utils import set_seed
Expand Down Expand Up @@ -629,7 +631,8 @@ def collate_fn(examples):
encoder_hidden_states = text_encoder(batch["input_ids"])[0]

# Predict the noise residual
model_pred = unet(noisy_latents, timesteps, encoder_hidden_states).sample
with autocast():
model_pred = unet(noisy_latents, timesteps, encoder_hidden_states).sample

# Get the target for loss depending on the prediction type
if noise_scheduler.config.prediction_type == "epsilon":
Expand Down