Hydra MLFlow Clash

[ssakhavi]

🐛 Bug

When using the MLFlow logger with Hydra, because the parameters passed to the LightningModule is a DictConfig, the condition in the logger/base.py is not met.

https://github.com/PyTorchLightning/pytorch-lightning/blob/8211256c46430e43e0c27e4f078c72085bb4ea34/pytorch_lightning/loggers/base.py#L177

To Reproduce

Use Hydra and MLFlow together.

Traceback (most recent call last):
  File "/home/siavash/KroniKare/kwae2/kwae_ma/models/pl_train_segmentation_model.py", line 115, in <module>
    main()
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/hydra/main.py", line 24, in decorated_main
    strict=strict,
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/hydra/_internal/utils.py", line 174, in run_hydra
    overrides=args.overrides,
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/hydra/_internal/hydra.py", line 86, in run
    job_subdir_key=None,
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/hydra/plugins/common/utils.py", line 109, in run_job
    ret.return_value = task_function(task_cfg)
  File "/home/siavash/KroniKare/kwae2/kwae_ma/models/pl_train_segmentation_model.py", line 111, in main
    trainer.fit(wound_seg_pl)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 765, in fit
    self.single_gpu_train(model)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/pytorch_lightning/trainer/distrib_parts.py", line 492, in single_gpu_train
    self.run_pretrain_routine(model)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 843, in run_pretrain_routine
    self.logger.log_hyperparams(ref_model.hparams)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/pytorch_lightning/loggers/base.py", line 275, in log_hyperparams
    [logger.log_hyperparams(params) for logger in self._logger_iterable]
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/pytorch_lightning/loggers/base.py", line 275, in <listcomp>
    [logger.log_hyperparams(params) for logger in self._logger_iterable]
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/pytorch_lightning/utilities/distributed.py", line 10, in wrapped_fn
    return fn(*args, **kwargs)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/pytorch_lightning/loggers/mlflow.py", line 105, in log_hyperparams
    self.experiment.log_param(self.run_id, k, v)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/mlflow/tracking/client.py", line 206, in log_param
    self._tracking_client.log_param(run_id, key, value)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/mlflow/tracking/_tracking_service/client.py", line 177, in log_param
    _validate_param_name(key)
  File "/home/siavash/anaconda3/envs/kwae-ma/lib/python3.7/site-packages/mlflow/utils/validation.py", line 120, in _validate_param_name
    INVALID_PARAMETER_VALUE)
mlflow.exceptions.MlflowException: Invalid parameter name: ''. Names may be treated as files in certain cases, and must not resolve to other names when treated as such. This name would resolve to '.'

Expected behavior

Check whether the instance if dict or DictConfig in the given line.

[github-actions]

Hi! thanks for your contribution!, great first issue!

[Borda]

Check whether the instance if dict or DictConfig in the given line.

@ssakhavi that sounds reasonable solution, mind sending a PR - fix and its test?