updated docs

pytorch_lightning/loggers/__init__.py

@@ -3,21 +3,27 @@ Lightning supports the most popular logging frameworks (TensorBoard, Comet, Weig
 To use a logger, simply pass it into the :class:`~pytorch_lightning.trainer.trainer.Trainer`.
 Lightning uses TensorBoard by default.
 
->>> from pytorch_lightning import Trainer
+.. code-block:: python
->>> from pytorch_lightning import loggers
+
->>> tb_logger = loggers.TensorBoardLogger('logs/')
+    from pytorch_lightning import Trainer
->>> trainer = Trainer(logger=tb_logger)
+    from pytorch_lightning import loggers
+    tb_logger = loggers.TensorBoardLogger('logs/')
+    trainer = Trainer(logger=tb_logger)
 
 Choose from any of the others such as MLflow, Comet, Neptune, WandB, ...
 
->>> comet_logger = loggers.CometLogger(save_dir='logs/')
+.. code-block:: python
->>> trainer = Trainer(logger=comet_logger)
+
+    comet_logger = loggers.CometLogger(save_dir='logs/')
+    trainer = Trainer(logger=comet_logger)
 
 To use multiple loggers, simply pass in a ``list`` or ``tuple`` of loggers ...
 
->>> tb_logger = loggers.TensorBoardLogger('logs/')
+.. code-block:: python
->>> comet_logger = loggers.CometLogger(save_dir='logs/')
+
->>> trainer = Trainer(logger=[tb_logger, comet_logger])
+    tb_logger = loggers.TensorBoardLogger('logs/')
+    comet_logger = loggers.CometLogger(save_dir='logs/')
+    trainer = Trainer(logger=[tb_logger, comet_logger])
 
 Note:
     All loggers log by default to ``os.getcwd()``. To change the path without creating a logger set
@@ -30,31 +36,33 @@ You can implement your own logger by writing a class that inherits from
 :class:`LightningLoggerBase`. Use the :func:`~pytorch_lightning.loggers.base.rank_zero_only`
 decorator to make sure that only the first process in DDP training logs data.
 
->>> from pytorch_lightning.utilities import rank_zero_only
+.. code-block:: python
->>> from pytorch_lightning.loggers import LightningLoggerBase
+
->>> class MyLogger(LightningLoggerBase):
+    from pytorch_lightning.utilities import rank_zero_only
-...
+    from pytorch_lightning.loggers import LightningLoggerBase
-...     @rank_zero_only
+    class MyLogger(LightningLoggerBase):
-...     def log_hyperparams(self, params):
+
-...         # params is an argparse.Namespace
+        @rank_zero_only
-...         # your code to record hyperparameters goes here
+        def log_hyperparams(self, params):
-...         pass
+            # params is an argparse.Namespace
-...
+            # your code to record hyperparameters goes here
-...     @rank_zero_only
+            pass
-...     def log_metrics(self, metrics, step):
+
-...         # metrics is a dictionary of metric names and values
+        @rank_zero_only
-...         # your code to record metrics goes here
+        def log_metrics(self, metrics, step):
-...         pass
+            # metrics is a dictionary of metric names and values
-...
+            # your code to record metrics goes here
-...     def save(self):
+            pass
-...         # Optional. Any code necessary to save logger data goes here
+
-...         pass
+        def save(self):
-...
+            # Optional. Any code necessary to save logger data goes here
-...     @rank_zero_only
+            pass
-...     def finalize(self, status):
+
-...         # Optional. Any code that needs to be run after training
+        @rank_zero_only
-...         # finishes goes here
+        def finalize(self, status):
-...         pass
+            # Optional. Any code that needs to be run after training
+            # finishes goes here
+            pass
 
 If you write a logger that may be useful to others, please send
 a pull request to add it to Lighting!
@@ -65,18 +73,20 @@ Using loggers
 Call the logger anywhere except ``__init__`` in your
 :class:`~pytorch_lightning.core.lightning.LightningModule` by doing:
 
->>> from pytorch_lightning import LightningModule
+.. code-block:: python
->>> class LitModel(LightningModule):
+
-...     def training_step(self, batch, batch_idx):
+    from pytorch_lightning import LightningModule
-...         # example
+    class LitModel(LightningModule):
-...         self.logger.experiment.whatever_method_summary_writer_supports(...)
+        def training_step(self, batch, batch_idx):
+            # example
+            self.logger.experiment.whatever_method_summary_writer_supports(...)
 
             # example if logger is a tensorboard logger
             self.logger.experiment.add_image('images', grid, 0)
             self.logger.experiment.add_graph(model, images)
 
-...     def any_lightning_module_function_or_hook(self):
+        def any_lightning_module_function_or_hook(self):
-...         self.logger.experiment.add_histogram(...)
+            self.logger.experiment.add_histogram(...)
 
 Read more in the `Experiment Logging use case <./experiment_logging.html>`_.