Document how to use TensorBoardLogger with fsspec

docs/source-pytorch/api_references.rst

@@ -82,6 +82,8 @@ core
     ~optimizer.LightningOptimizer
 
 
+.. _loggers-api-references:
+
 loggers
 -------
 

docs/source-pytorch/common/remote_fs.rst

@@ -17,7 +17,8 @@ Working with different filesystems can be accomplished by appending a protocol l
     trainer = Trainer(default_root_dir="s3://my_bucket/data/")
     trainer.fit(model)
 
-You could pass custom paths to loggers for logging data.
+
+For logging, remote filesystem support depends on the particular logger integration being used. Consult :ref:`the documentation of the individual logger <loggers-api-references>` for more details.
 
 .. code-block:: python
 

src/lightning/pytorch/loggers/csv_logs.py

@@ -42,6 +42,8 @@ class ExperimentWriter(_FabricExperimentWriter):
     Currently, supports to log hyperparameters and metrics in YAML and CSV
     format, respectively.
 
+    This logger supports logging to remote filesystems via ``fsspec``. Make sure you have it installed.
+
     Args:
         log_dir: Directory for the experiment logs
     """

src/lightning/pytorch/loggers/tensorboard.py

@@ -46,12 +46,15 @@
 
 class TensorBoardLogger(Logger, FabricTensorBoardLogger):
     r"""
-    Log to local file system in `TensorBoard <https://www.tensorflow.org/tensorboard>`_ format.
+    Log to local or remote file system in `TensorBoard <https://www.tensorflow.org/tensorboard>`_ format.
 
     Implemented using :class:`~tensorboardX.SummaryWriter`. Logs are saved to
     ``os.path.join(save_dir, name, version)``. This is the default logger in Lightning, it comes
     preinstalled.
 
+    This logger supports logging to remote filesystems via ``fsspec``. Make sure you have it installed
+    and you don't have tensorflow (otherwise it will use tf.io.gfile instead of fsspec).
+
     Example:
 
     .. testcode::