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

[Wuerstchen] fix fp16 training and correct lora args #6245

Merged
merged 3 commits into from
Dec 26, 2023
Merged

[Wuerstchen] fix fp16 training and correct lora args #6245

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c0ad660
fix fp16 training
kashif Dec 20, 2023
81d00e1
Merge branch 'main' into lora-float16
kashif Dec 20, 2023
55c5c27
Merge branch 'main' into lora-float16
sayakpaul Dec 25, 2023
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
10 changes: 9 additions & 1 deletion examples/wuerstchen/text_to_image/train_text_to_image_lora_prior.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -527,9 +527,17 @@ def deepspeed_zero_init_disabled_context_manager():

# lora attn processor
prior_lora_config = LoraConfig(
r=args.rank, target_modules=["to_k", "to_q", "to_v", "to_out.0", "add_k_proj", "add_v_proj"]
r=args.rank,
lora_alpha=args.rank,
target_modules=["to_k", "to_q", "to_v", "to_out.0", "add_k_proj", "add_v_proj"],
)
# Add adapter and make sure the trainable params are in float32.
prior.add_adapter(prior_lora_config)
if args.mixed_precision == "fp16":
for param in prior.parameters():
# only upcast trainable parameters (LoRA) into fp32
if param.requires_grad:
param.data = param.to(torch.float32)

# create custom saving & loading hooks so that `accelerator.save_state(...)` serializes in a nice format
def save_model_hook(models, weights, output_dir):
Expand Down