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

cifar10: try unblock training #335

Merged
merged 7 commits into from
Jul 22, 2024
Merged

cifar10: try unblock training #335

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
dc32e45
cifar10: try unblock training
Borda Jul 22, 2024
18fa628
args
Borda Jul 22, 2024
f95ef2b
cuda
Borda Jul 22, 2024
355d6bb
on_train_end
Borda Jul 22, 2024
44ada63
lint
Borda Jul 22, 2024
105b3f9
Merge branch 'main' into fix/cifar
Borda Jul 22, 2024
a8cf7de
Apply suggestions from code review
Borda Jul 22, 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
18 changes: 9 additions & 9 deletions lightning_examples/cifar10-baseline/baseline.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@
from lightning.pytorch.callbacks import LearningRateMonitor
from lightning.pytorch.loggers import CSVLogger
from torch.optim.lr_scheduler import OneCycleLR
from torch.optim.swa_utils import AveragedModel, update_bn
from torch.optim.swa_utils import AveragedModel
from torch.utils.data import DataLoader, random_split
from torchmetrics.functional import accuracy
from torchvision.datasets import CIFAR10
Expand Down Expand Up @@ -173,15 +173,15 @@ def configure_optimizers(self):
model = LitResnet(lr=0.05)

trainer = L.Trainer(
max_epochs=30,
max_epochs=5,
accelerator="auto",
devices=1,
logger=CSVLogger(save_dir="logs/"),
callbacks=[LearningRateMonitor(logging_interval="step")],
)

trainer.fit(model, train_dataloader, val_dataloaders=val_dataloader)
trainer.test(model, test_dataloader)
trainer.fit(model, train_dataloaders=train_dataloader, val_dataloaders=val_dataloader)
trainer.test(model, dataloaders=test_dataloader)

# %%

Expand Down Expand Up @@ -229,22 +229,22 @@ def configure_optimizers(self):
optimizer = torch.optim.SGD(self.model.parameters(), lr=self.hparams.lr, momentum=0.9, weight_decay=5e-4)
return optimizer

def on_train_end(self):
update_bn(self.trainer.datamodule.train_dataloader(), self.swa_model, device=self.device)
# def on_train_end(self): # todo: failing as trainer has only dataloaders, not datamodules
# update_bn(self.trainer.datamodule.train_dataloader(), self.swa_model, device=self.device)


# %%
swa_model = SWAResnet(model.model, lr=0.01)

swa_trainer = L.Trainer(
max_epochs=20,
max_epochs=5,
accelerator="auto",
devices=1,
logger=CSVLogger(save_dir="logs/"),
)

swa_trainer.fit(swa_model, train_dataloader, val_dataloaders=val_dataloader)
swa_trainer.test(swa_model, test_dataloader)
swa_trainer.fit(swa_model, train_dataloaders=train_dataloader, val_dataloaders=val_dataloader)
swa_trainer.test(swa_model, dataloaders=test_dataloader)

# %%

Expand Down
Loading