WandbLogger throws error with mode="disabled" (pytorch lightning 1.7)

[hogru]

🐛 Bug

After upgrading to pytorch lightning 1.7 instantiating WandbLogger with mode="disabled" throws the following error:

Traceback (most recent call last):
  File "/Users/stephan/Library/Mobile Documents/com~apple~CloudDocs/Ablage/AI Master/Courses/2022S/Master Thesis/molgen/src/playground/bug_report_model.py", line 70, in <module>
    run()
  File "/Users/stephan/Library/Mobile Documents/com~apple~CloudDocs/Ablage/AI Master/Courses/2022S/Master Thesis/molgen/src/playground/bug_report_model.py", line 48, in run
    wandb_logger = WandbLogger(mode="disabled")  # <== Added
  File "/Users/stephan/Library/Caches/pypoetry/virtualenvs/molgen-6oMP0hTK-py3.9/lib/python3.9/site-packages/pytorch_lightning/loggers/wandb.py", line 315, in __init__
    _ = self.experiment
  File "/Users/stephan/Library/Caches/pypoetry/virtualenvs/molgen-6oMP0hTK-py3.9/lib/python3.9/site-packages/pytorch_lightning/loggers/logger.py", line 54, in experiment
    return get_experiment() or DummyExperiment()
  File "/Users/stephan/Library/Caches/pypoetry/virtualenvs/molgen-6oMP0hTK-py3.9/lib/python3.9/site-packages/pytorch_lightning/utilities/rank_zero.py", line 32, in wrapped_fn
    return fn(*args, **kwargs)
  File "/Users/stephan/Library/Caches/pypoetry/virtualenvs/molgen-6oMP0hTK-py3.9/lib/python3.9/site-packages/pytorch_lightning/loggers/logger.py", line 52, in get_experiment
    return fn(self)
  File "/Users/stephan/Library/Caches/pypoetry/virtualenvs/molgen-6oMP0hTK-py3.9/lib/python3.9/site-packages/pytorch_lightning/loggers/wandb.py", line 368, in experiment
    assert isinstance(self._experiment, Run)
AssertionError

To Reproduce

import os

import torch
from torch.utils.data import DataLoader, Dataset

from pytorch_lightning import LightningModule, Trainer
from pytorch_lightning.loggers import WandbLogger  # <== Added


class RandomDataset(Dataset):
    def __init__(self, size, length):
        self.len = length
        self.data = torch.randn(length, size)

    def __getitem__(self, index):
        return self.data[index]

    def __len__(self):
        return self.len


class BoringModel(LightningModule):
    def __init__(self):
        super().__init__()
        self.layer = torch.nn.Linear(32, 2)

    def forward(self, x):
        return self.layer(x)

    def training_step(self, batch, batch_idx):
        loss = self(batch).sum()
        self.log("train_loss", loss)
        return {"loss": loss}

    def validation_step(self, batch, batch_idx):
        loss = self(batch).sum()
        self.log("valid_loss", loss)

    def test_step(self, batch, batch_idx):
        loss = self(batch).sum()
        self.log("test_loss", loss)

    def configure_optimizers(self):
        return torch.optim.SGD(self.layer.parameters(), lr=0.1)


def run():
    wandb_logger = WandbLogger(mode="disabled")  # <== Added

    train_data = DataLoader(RandomDataset(32, 64), batch_size=2)
    val_data = DataLoader(RandomDataset(32, 64), batch_size=2)
    test_data = DataLoader(RandomDataset(32, 64), batch_size=2)


    model = BoringModel()
    trainer = Trainer(
        default_root_dir=os.getcwd(),
        limit_train_batches=1,
        limit_val_batches=1,
        limit_test_batches=1,
        num_sanity_val_steps=0,
        max_epochs=1,
        enable_model_summary=False,
    )
    trainer.fit(model, train_dataloaders=train_data, val_dataloaders=val_data)
    trainer.test(model, dataloaders=test_data)


if __name__ == "__main__":
    run()

Expected behavior

No error, pass parameter to wandb (as it did until 1.6.5)

Environment

* CUDA:
	- GPU:
	- available:         False
	- version:           None
* Packages:
	- lightning:         None
	- lightning_app:     None
	- numpy:             1.23.1
	- pyTorch_debug:     False
	- pyTorch_version:   1.12.0
	- pytorch-lightning: 1.7.0
	- tqdm:              4.64.0
* System:
	- OS:                Darwin
	- architecture:
		- 64bit
		- 
	- processor:         i386
	- python:            3.9.12
	- version:           Darwin Kernel Version 21.5.0: Tue Apr 26 21:08:37 PDT 2022; root:xnu-8020.121.3~4/RELEASE_ARM64_T6000

Additional context

cc @awaelchli @morganmcg1 @borisdayma @scottire @manangoel99

[carmocca]

@gautierdag The PR #13483 seems to have caused this. Would you like to take a look?

[gautierdag]

@carmocca Indeed the typing PR introduced this bug. I've fixed it in #14112, the _experiment can be a RunDisabled object if mode="disabled" is passed. I can write an additional test for this behavior but since we mock the wandb library during testing it feels a little redundant.

Thanks @hogru for the mock example!