can't pass resume flag to WandbLogger

[AsaphLightricks]

https://github.com/PyTorchLightning/pytorch-lightning/blob/e7abd8e89bcc26a97203119f77f9c59dcb0d8791/pytorch_lightning/loggers/wandb.py#L161

I can't override the resume=allow. If i try passing the WandbLogger resume=True for instance i get this error:

Traceback (most recent call last):
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/hydra/_internal/utils.py", line 361, in _run_hydra
    run_and_report(
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/hydra/_internal/utils.py", line 213, in run_and_report
    raise ex
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/hydra/_internal/utils.py", line 210, in run_and_report
    return func()
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/hydra/_internal/utils.py", line 362, in <lambda>
    lambda: hydra.run(
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 93, in run
    return run_job(
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/hydra/core/utils.py", line 129, in run_job
    ret.return_value = task_function(task_cfg)
  File "/Users/ashamir/PycharmProjects/pytorch-lightning-template/src/train.py", line 113, in main
    trainer.fit(model=lightning_module, datamodule=data_module)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 474, in fit
    self.setup_trainer(model)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 421, in setup_trainer
    self.logger.log_hyperparams(model.hparams_initial)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/loggers/base.py", line 380, in log_hyperparams
    logger.log_hyperparams(params)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/utilities/distributed.py", line 40, in wrapped_fn
    return fn(*args, **kwargs)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/loggers/wandb.py", line 178, in log_hyperparams
    self.experiment.config.update(params, allow_val_change=True)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/loggers/base.py", line 41, in experiment
    return get_experiment() or DummyExperiment()
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/utilities/distributed.py", line 40, in wrapped_fn
    return fn(*args, **kwargs)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/loggers/base.py", line 39, in get_experiment
    return fn(self)
  File "/Users/ashamir/opt/miniconda3/envs/pytorch-lightning-template/lib/python3.8/site-packages/pytorch_lightning/loggers/wandb.py", line 152, in experiment
    self._experiment = wandb.init(
TypeError: init() got multiple values for keyword argument 'resume'

This flag should be configureable just like the others.
@borisdayma

[borisdayma]

That's a good point!
In the meantime you could call wandb.init separately and WandbLogger will not create a new run.

[stale]

This issue has been automatically marked as stale because it hasn't had any recent activity. This issue will be closed in 7 days if no further activity occurs. Thank you for your contributions, Pytorch Lightning Team!

[borisdayma]

There's PR #6989 open for this issue.

[awaelchli]

somehow nobody labelled this.
labelling now as bugifx.
thanks @borisdayma