Erreur avec le le logger mlflow

[brunorigal]

Bug description

With the mlflow logger, I have the following error:

Traceback (most recent call last):
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py", line 48, in _call_and_handle_interrupt
    return trainer_fn(*args, **kwargs)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 599, in _fit_impl
    self._run(model, ckpt_path=ckpt_path)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 1025, in _run
    call._call_teardown_hook(self)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py", line 148, in _call_teardown_hook
    logger.finalize("success")
  File "/opt/venvs/lib/python3.10/site-packages/lightning_utilities/core/rank_zero.py", line 41, in wrapped_fn
    return fn(*args, **kwargs)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/loggers/mlflow.py", line 289, in finalize
    self._scan_and_log_checkpoints(self._checkpoint_callback)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/loggers/mlflow.py", line 369, in _scan_and_log_checkpoints
    self.experiment.log_artifact(self._run_id, p, artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/tracking/client.py", line 2379, in log_artifact
    self._tracking_client.log_artifact(run_id, local_path, artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/tracking/_tracking_service/client.py", line 931, in log_artifact
    artifact_repo.log_artifact(local_path, artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/store/artifact/local_artifact_repo.py", line 33, in log_artifact
    verify_artifact_path(artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/store/artifact/artifact_repo.py", line 464, in verify_artifact_path
    raise MlflowException(
mlflow.exceptions.MlflowException: Invalid artifact path: 'ordering-model-epoch=09-val_loss=0.00'. 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 'ordering-model-epoch=09-val_loss=0.00'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3.10/pdb.py", line 1726, in main
    pdb._runscript(mainpyfile)
  File "/usr/lib/python3.10/pdb.py", line 1586, in _runscript
    self.run(statement)
  File "/usr/lib/python3.10/bdb.py", line 597, in run
    exec(cmd, globals, locals)
  File "<string>", line 1, in <module>
  File "/workspace/perso/train.py", line 66, in <module>
    train_model(config)
  File "/workspace/vrsgen/training/train.py", line 145, in train_model
    trainer.fit(model, train_loader, val_loader)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py", line 561, in fit
    call._call_and_handle_interrupt(
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py", line 68, in _call_and_handle_interrupt
    _interrupt(trainer, exception)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py", line 82, in _interrupt
    logger.finalize("failed")
  File "/opt/venvs/lib/python3.10/site-packages/lightning_utilities/core/rank_zero.py", line 41, in wrapped_fn
    return fn(*args, **kwargs)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/loggers/mlflow.py", line 289, in finalize
    self._scan_and_log_checkpoints(self._checkpoint_callback)
  File "/opt/venvs/lib/python3.10/site-packages/pytorch_lightning/loggers/mlflow.py", line 369, in _scan_and_log_checkpoints
    self.experiment.log_artifact(self._run_id, p, artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/tracking/client.py", line 2379, in log_artifact
    self._tracking_client.log_artifact(run_id, local_path, artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/tracking/_tracking_service/client.py", line 931, in log_artifact
    artifact_repo.log_artifact(local_path, artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/store/artifact/local_artifact_repo.py", line 33, in log_artifact
    verify_artifact_path(artifact_path)
  File "/opt/venvs/lib/python3.10/site-packages/mlflow/store/artifact/artifact_repo.py", line 464, in verify_artifact_path
    raise MlflowException(
mlflow.exceptions.MlflowException: Invalid artifact path: 'ordering-model-epoch=09-val_loss=0.00'. 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 'ordering-model-epoch=09-val_loss=0.00'
Uncaught exception. Entering post mortem debugging
Running 'cont' or 'step' will restart the program

The problem comes from the path_not_unique function (line 168 in mlflow.utils.validation) which works only with str and here receives a pathlib.Path object ->posixpath.normpath(name) will return a str even for a Path as input and then norm != name is always false

to correct this, in the mlflow logger in pytorch_ligntning.loggers.mlflow, line 369 should be:
self.experiment.log_artifact(self._run_id, p, str(artifact_path))

What version are you seeing the problem on?

v2.5

Reproduced in studio

No response

How to reproduce the bug

Error messages and logs

# Error messages and logs here please

Environment

Current environment

#- PyTorch Lightning Version (e.g., 2.5.0):
#- PyTorch Version (e.g., 2.5):
#- Python version (e.g., 3.12):
#- OS (e.g., Linux):
#- CUDA/cuDNN version:
#- GPU models and configuration:
#- How you installed Lightning(`conda`, `pip`, source):

More info

No response

[Clara-liu]

A fixed was recently merged, then reverted for some reason. See: #20669