AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-29 07:22:12 +03:00
Code Activity

use backend-agnostic cache and pass devide.

Browse Source
This commit is contained in:
sayakpaul
2025-04-09 11:48:26 +05:30
parent f50de75b69
commit b56112db6e
1 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
examples/controlnet/train_controlnet_sd3.py
View File

@@ -53,6 +53,7 @@ from diffusers.optimization import get_scheduler
from diffusers.training_utils import compute_density_for_timestep_sampling, compute_loss_weighting_for_sd3, free_memory
from diffusers.utils import check_min_version, is_wandb_available, make_image_grid
from diffusers.utils.hub_utils import load_or_create_model_card, populate_model_card
from diffusers.utils.testing_utils import backend_empty_cache
from diffusers.utils.torch_utils import is_compiled_module
@@ -118,7 +119,7 @@ def log_validation(controlnet, args, accelerator, weight_dtype, step, is_final_v
del pipeline
gc.collect()
torch.cuda.empty_cache()
backend_empty_cache(accelerator.device.type)
pipeline = StableDiffusion3ControlNetPipeline.from_pretrained(
args.pretrained_model_name_or_path,
@@ -131,7 +132,7 @@ def log_validation(controlnet, args, accelerator, weight_dtype, step, is_final_v
variant=args.variant,
torch_dtype=weight_dtype,
)
pipeline.enable_model_cpu_offload()
pipeline.enable_model_cpu_offload(device=accelerator.device.type)
pipeline.set_progress_bar_config(disable=True)
image_logs = []