Skip to content

Fix Wrong Text-encoder Grad Setting in Custom_Diffusion Training #7302

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 2 commits into from
Mar 13, 2024
Merged

Fix Wrong Text-encoder Grad Setting in Custom_Diffusion Training #7302

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8dc0230
fix index in set textencoder grad
Rbrq03 Mar 13, 2024
dbea2b1
Merge branch 'main' into custom_diffusion_train_fix
sayakpaul Mar 13, 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
2 changes: 1 addition & 1 deletion examples/custom_diffusion/train_custom_diffusion.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1178,7 +1178,7 @@ def main(args):
grads_text_encoder = text_encoder.get_input_embeddings().weight.grad
# Get the index for tokens that we want to zero the grads for
index_grads_to_zero = torch.arange(len(tokenizer)) != modifier_token_id[0]
for i in range(len(modifier_token_id[1:])):
for i in range(1, len(modifier_token_id)):
index_grads_to_zero = index_grads_to_zero & (
torch.arange(len(tokenizer)) != modifier_token_id[i]
)
Expand Down