Configuration not syncing for Wandb Offline Mode #20571
Description
Bug description
When I train using wandb offline, pytorch-lightning doesn't upload the configuration of my job. It loads the metrics and the jobs summary. A workaround is to call wandb.init(config=model.hparams) before the pytorch-lightning wandbLogger initialization.
What version are you seeing the problem on?
v2.5
How to reproduce the bug
import pytorch_lightning as pl
from pytorch_lightning.loggers import WandbLogger
import torch
from torch import nn
from torch.utils.data import DataLoader, random_split, TensorDataset
import wandb
# Dummy dataset
data = torch.randn(1000, 10)
targets = torch.randint(0, 2, (1000,))
dataset = TensorDataset(data, targets)
# Split dataset
train_size = int(0.8 * len(dataset))
val_size = len(dataset) - train_size
train_dataset, val_dataset = random_split(dataset, [train_size, val_size])
# DataLoader
train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True)
val_loader = DataLoader(val_dataset, batch_size=32)
# Define a simple model
class SimpleModel(pl.LightningModule):
def __init__(self, hidden_size):
super().__init__()
self.save_hyperparameters()
self.layer_1 = nn.Linear(10, hidden_size,)
self.layer_2 = nn.Linear(hidden_size, hidden_size)
self.layer_3 = nn.Linear(64, 1)
self.loss = nn.BCEWithLogitsLoss()
def forward(self, x):
x = torch.relu(self.layer_1(x))
x = torch.relu(self.layer_2(x))
x = self.layer_3(x)
return x
def training_step(self, batch, batch_idx):
x, y = batch
y_hat = self(x).squeeze()
loss = self.loss(y_hat, y.float())
self.log('train_loss', loss)
return loss
def validation_step(self, batch, batch_idx):
x, y = batch
y_hat = self(x).squeeze()
loss = self.loss(y_hat, y.float())
self.log('val_loss', loss)
return loss
def configure_optimizers(self):
optimizer = torch.optim.Adam(self.parameters(), lr=1e-3)
return optimizer
# Initialize model
model = SimpleModel(64)
# Initialize trainer
wandblogger = WandbLogger(project='minimal-example', offline=True, save_dir='.')
trainer = pl.Trainer(max_epochs=10, logger=wandblogger)
# Train the model
trainer.fit(model, train_loader, val_loader)Error messages and logs
# No errors
Environment
Current environment
- CUDA:
- GPU: None
- available: False
- version: None - Lightning:
- lightning-utilities: 0.12.0
- pytorch-lightning: 2.5.0.post0
- torch: 2.5.1
- torchmetrics: 1.6.1 - Packages:
- annotated-types: 0.7.0
- appdirs: 1.4.4
- autocommand: 2.2.2
- backports.tarfile: 1.2.0
- brotli: 1.1.0
- certifi: 2024.12.14
- cffi: 1.17.1
- charset-normalizer: 3.4.1
- click: 8.1.8
- colorama: 0.4.6
- docker-pycreds: 0.4.0
- eval-type-backport: 0.2.2
- filelock: 3.17.0
- fsspec: 2025.2.0
- gitdb: 4.0.12
- gitpython: 3.1.44
- h2: 4.2.0
- hpack: 4.1.0
- hyperframe: 6.1.0
- idna: 3.10
- importlib-metadata: 8.0.0
- inflect: 7.3.1
- jaraco.collections: 5.1.0
- jaraco.context: 5.3.0
- jaraco.functools: 4.0.1
- jaraco.text: 3.12.1
- jinja2: 3.1.5
- lightning-utilities: 0.12.0
- markupsafe: 3.0.2
- more-itertools: 10.3.0
- mpmath: 1.3.0
- networkx: 3.4.2
- numpy: 2.2.2
- packaging: 24.2
- pip: 25.0
- platformdirs: 4.3.6
- protobuf: 5.28.3
- psutil: 6.1.1
- pybind11: 2.13.6
- pybind11-global: 2.13.6
- pycparser: 2.22
- pydantic: 2.10.6
- pydantic-core: 2.27.2
- pysocks: 1.7.1
- pytorch-lightning: 2.5.0.post0
- pyyaml: 6.0.2
- requests: 2.32.3
- sentry-sdk: 2.20.0
- setproctitle: 1.3.4
- setuptools: 75.8.0
- six: 1.17.0
- smmap: 5.0.0
- sympy: 1.13.3
- tomli: 2.0.1
- torch: 2.5.1
- torchmetrics: 1.6.1
- tqdm: 4.67.1
- typeguard: 4.3.0
- typing-extensions: 4.12.2
- urllib3: 2.3.0
- wandb: 0.19.5
- wheel: 0.45.1
- win-inet-pton: 1.1.0
- zipp: 3.19.2
- zstandard: 0.23.0 - System:
- OS: Windows
- architecture:
- 64bit
- WindowsPE
- processor: Intel64 Family 6 Model 154 Stepping 4, GenuineIntel
- python: 3.11.0
- release: 10
- version: 10.0.26100
More info
No response