Skip to content

[Training] SD3 training fixes #8917

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
sayakpaul merged 4 commits into from
Jul 21, 2024
Merged

[Training] SD3 training fixes #8917

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
418729d
SD3 training fixes
sayakpaul Jul 21, 2024
9050093
rewrite noise addition part to respect the eqn.
sayakpaul Jul 21, 2024
93f40c4
styler
sayakpaul Jul 21, 2024
08d7501
Update examples/dreambooth/README_sd3.md
sayakpaul Jul 21, 2024
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
4 changes: 3 additions & 1 deletion examples/dreambooth/README_sd3.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -183,4 +183,6 @@ accelerate launch train_dreambooth_lora_sd3.py \

## Other notes

We default to the "logit_normal" weighting scheme for the loss following the SD3 paper. Thanks to @bghira for helping us discover that for other weighting schemes supported from the training script, training may incur numerical instabilities.
1. We default to the "logit_normal" weighting scheme for the loss following the SD3 paper. Thanks to @bghira for helping us discover that for other weighting schemes supported from the training script, training may incur numerical instabilities.
2. Thanks to `bghira`, `JinxuXiang`, and `bendanzzc` for helping us discover a bug in how VAE encoding was being done previously. This has been fixed in [#8917](https://github.com/huggingface/diffusers/pull/8917).
3. Additionally, we now have the option to control if we want to apply preconditioning to the model outputs via a `--precondition_outputs` CLI arg. It affects how the model `target` is calculated as well.
20 changes: 16 additions & 4 deletions examples/dreambooth/train_dreambooth_lora_sd3.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -523,6 +523,13 @@ def parse_args(input_args=None):
default=1.29,
help="Scale of mode weighting scheme. Only effective when using the `'mode'` as the `weighting_scheme`.",
)
parser.add_argument(
"--precondition_outputs",
type=int,
default=1,
help="Flag indicating if we are preconditioning the model outputs or not as done in EDM. This affects how "
"model `target` is calculated.",
)
parser.add_argument(
"--optimizer",
type=str,
Expand Down Expand Up @@ -1636,7 +1643,7 @@ def get_sigmas(timesteps, n_dim=4, dtype=torch.float32):

# Convert images to latent space
model_input = vae.encode(pixel_values).latent_dist.sample()
model_input = model_input * vae.config.scaling_factor
model_input = (model_input - vae.config.shift_factor) * vae.config.scaling_factor
model_input = model_input.to(dtype=weight_dtype)

# Sample noise that we'll add to the latents
Expand All @@ -1656,8 +1663,9 @@ def get_sigmas(timesteps, n_dim=4, dtype=torch.float32):
timesteps = noise_scheduler_copy.timesteps[indices].to(device=model_input.device)

# Add noise according to flow matching.
# zt = (1 - texp) * x + texp * z1
sigmas = get_sigmas(timesteps, n_dim=model_input.ndim, dtype=model_input.dtype)
noisy_model_input = sigmas * noise + (1.0 - sigmas) * model_input
noisy_model_input = (1.0 - sigmas) * model_input + sigmas * noise

# Predict the noise residual
model_pred = transformer(
Expand All @@ -1670,14 +1678,18 @@ def get_sigmas(timesteps, n_dim=4, dtype=torch.float32):

# Follow: Section 5 of https://arxiv.org/abs/2206.00364.
# Preconditioning of the model outputs.
model_pred = model_pred * (-sigmas) + noisy_model_input
if args.precondition_outputs:
model_pred = model_pred * (-sigmas) + noisy_model_input

# these weighting schemes use a uniform timestep sampling
# and instead post-weight the loss
weighting = compute_loss_weighting_for_sd3(weighting_scheme=args.weighting_scheme, sigmas=sigmas)

# flow matching loss
target = model_input
if args.precondition_outputs:
target = model_input
else:
target = noise - model_input

if args.with_prior_preservation:
# Chunk the noise and model_pred into two parts and compute the loss on each part separately.
Expand Down
21 changes: 17 additions & 4 deletions examples/dreambooth/train_dreambooth_sd3.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -494,6 +494,13 @@ def parse_args(input_args=None):
default=1.29,
help="Scale of mode weighting scheme. Only effective when using the `'mode'` as the `weighting_scheme`.",
)
parser.add_argument(
"--precondition_outputs",
type=int,
default=1,
help="Flag indicating if we are preconditioning the model outputs or not as done in EDM. This affects how "
"model `target` is calculated.",
)
parser.add_argument(
"--optimizer",
type=str,
Expand Down Expand Up @@ -1549,7 +1556,7 @@ def get_sigmas(timesteps, n_dim=4, dtype=torch.float32):

# Convert images to latent space
model_input = vae.encode(pixel_values).latent_dist.sample()
model_input = model_input * vae.config.scaling_factor
model_input = (model_input - vae.config.shift_factor) * vae.config.scaling_factor
model_input = model_input.to(dtype=weight_dtype)

# Sample noise that we'll add to the latents
Expand All @@ -1569,8 +1576,9 @@ def get_sigmas(timesteps, n_dim=4, dtype=torch.float32):
timesteps = noise_scheduler_copy.timesteps[indices].to(device=model_input.device)

# Add noise according to flow matching.
# zt = (1 - texp) * x + texp * z1
sigmas = get_sigmas(timesteps, n_dim=model_input.ndim, dtype=model_input.dtype)
noisy_model_input = sigmas * noise + (1.0 - sigmas) * model_input
noisy_model_input = (1.0 - sigmas) * model_input + sigmas * noise

# Predict the noise residual
if not args.train_text_encoder:
Expand Down Expand Up @@ -1598,13 +1606,18 @@ def get_sigmas(timesteps, n_dim=4, dtype=torch.float32):

# Follow: Section 5 of https://arxiv.org/abs/2206.00364.
# Preconditioning of the model outputs.
model_pred = model_pred * (-sigmas) + noisy_model_input
if args.precondition_outputs:
model_pred = model_pred * (-sigmas) + noisy_model_input

# these weighting schemes use a uniform timestep sampling
# and instead post-weight the loss
weighting = compute_loss_weighting_for_sd3(weighting_scheme=args.weighting_scheme, sigmas=sigmas)

# flow matching loss
target = model_input
if args.precondition_outputs:
target = model_input
else:
target = noise - model_input

if args.with_prior_preservation:
# Chunk the noise and model_pred into two parts and compute the loss on each part separately.
Expand Down