Skip to content

fix dit training #752

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
Oct 17, 2024
Merged

fix dit training #752

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
71ef171
fix dit training
nemonameless Oct 17, 2024
25a93ec
fix dit training
nemonameless Oct 17, 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
8 changes: 7 additions & 1 deletion ppdiffusers/examples/class_conditional_image_generation/DiT/diffusion/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,13 @@
from .dit import DiT
from .dit_llama import DiT_Llama
from .respace import SpacedDiffusion, space_timesteps
from .trainer import LatentDiffusionTrainer, LatentDiffusionAutoTrainer
from .trainer import LatentDiffusionTrainer
try:
from paddlenlp.trainer.auto_trainer import AutoTrainer
from .trainer_auto import LatentDiffusionAutoTrainer
except:
print(f'please install paddlepaddle-gpu>=3.0.0b2 if using auto trainer')

from .trainer_args import (
DataArguments,
ModelArguments,
Expand Down
15 changes: 0 additions & 15 deletions ppdiffusers/examples/class_conditional_image_generation/DiT/diffusion/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,6 @@
from paddle.distributed import fleet
from paddle.io import get_worker_info
from paddlenlp.trainer import Trainer
from paddlenlp.trainer.auto_trainer import AutoTrainer
from paddlenlp.trainer.integrations import (
INTEGRATION_TO_CALLBACK,
TrainerCallback,
Expand Down Expand Up @@ -295,20 +294,6 @@ def __impl__():

return __impl__

class LatentDiffusionAutoTrainer(AutoTrainer):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)

def _get_meshes_for_loader(self):
def _get_mesh(pp_idx=0):
return fleet.auto.get_mesh().get_mesh_with_dim("pp")[pp_idx]

return _get_mesh(0) # label_id is not label

def _wrap_for_dist_loader(self, train_dataloader):
dist_loader = super()._wrap_for_dist_loader(train_dataloader)
dist_loader._input_keys = ["latents", "label_id"]
return dist_loader

class LatentDiffusionTrainer(Trainer):
def __init__(self, **kwargs):
Expand Down
Loading