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

remove in-place addition for dreambooth train with text encoder #8825

Merged
merged 4 commits into from
Apr 29, 2024
Merged

remove in-place addition for dreambooth train with text encoder #8825

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
bf13ac8
remove in-place addition for dreambooth train with text encoder
Victor49152 Apr 4, 2024
7f22cac
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Apr 4, 2024
38c1f79
Merge branch 'main' into mingyuanm/dreambooth_fix
Victor49152 Apr 5, 2024
b8277df
Merge branch 'main' into mingyuanm/dreambooth_fix
Victor49152 Apr 29, 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
3 changes: 2 additions & 1 deletion nemo/collections/multimodal/models/text_to_image/dreambooth/dreambooth.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,6 @@
from torch._inductor import config as inductor_config

from nemo.collections.multimodal.data.dreambooth.dreambooth_dataset import DreamBoothDataset
from nemo.collections.multimodal.modules.stable_diffusion.attention import LinearWrapper
from nemo.collections.multimodal.modules.stable_diffusion.distributions.distributions import (
DiagonalGaussianDistribution,
)
Expand Down Expand Up @@ -647,6 +646,8 @@ def load_from_checkpoint(
return checkpoint

def _check_and_add_adapter(self, name, module, peft_name, peft_cfg, name_key_to_mcore_mixins=None):
from nemo.collections.multimodal.modules.stable_diffusion.attention import LinearWrapper

if isinstance(module, AdapterModuleMixin):
if isinstance(module, LinearWrapper):
peft_cfg.in_features, peft_cfg.out_features = module.in_features, module.out_features
Expand Down
2 changes: 1 addition & 1 deletion nemo/collections/multimodal/modules/stable_diffusion/encoders/modules.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -718,7 +718,7 @@ def forward(self, text):

def encode_with_transformer(self, text):
x = self.model.language_model.embedding.word_embeddings(text)
x += self.model.language_model.embedding.position_embeddings
x = x + self.model.language_model.embedding.position_embeddings
x = x.permute(1, 0, 2) # NLD -> LND
x = self.text_transformer_forward(x, attn_mask=self.model.attn_mask)
x = self.model.language_model.encoder.final_layernorm(x)
Expand Down
Loading