Merge branch 'master' into func_metric_docs

docs/source/datamodules.rst

@@ -40,7 +40,7 @@ This class can then be shared and used anywhere:
 Why do I need a DataModule?
 ---------------------------
 In normal PyTorch code, the data cleaning/preparation is usually scattered across many files. This makes
-sharing and reusing the exact splits, and transforms across projects.
+sharing and reusing the exact splits and transforms across projects impossible.
 
 Datamodules are for you if you ever asked the questions:
 

docs/source/hyperparameters.rst

@@ -9,7 +9,7 @@
 
 Hyperparameters
 ---------------
-Lightning has utilities to interact seamlessly with the command line ArgumentParser
+Lightning has utilities to interact seamlessly with the command line ``ArgumentParser``
 and plays well with the hyperparameter optimization framework of your choice.
 
 ----------
@@ -37,15 +37,15 @@ Argparser Best Practices
 ^^^^^^^^^^^^^^^^^^^^^^^^
 It is best practice to layer your arguments in three sections.
 
-1.  Trainer args (gpus, num_nodes, etc...)
-2.  Model specific arguments (layer_dim, num_layers, learning_rate, etc...)
-3.  Program arguments (data_path, cluster_email, etc...)
+1.  Trainer args (``gpus``, ``num_nodes``, etc...)
+2.  Model specific arguments (``layer_dim``, ``num_layers``, ``learning_rate``, etc...)
+3.  Program arguments (``data_path``, ``cluster_email``, etc...)
 
 |
 
-We can do this as follows. First, in your LightningModule, define the arguments
+We can do this as follows. First, in your ``LightningModule``, define the arguments
 specific to that module. Remember that data splits or data paths may also be specific to
-a module (ie: if your project has a model that trains on Imagenet and another on CIFAR-10).
+a module (i.e.: if your project has a model that trains on Imagenet and another on CIFAR-10).
 
 .. testcode::
 
@@ -58,7 +58,7 @@ a module (ie: if your project has a model that trains on Imagenet and another on
             parser.add_argument('--data_path', type=str, default='/some/path')
             return parser
 
-Now in your main trainer file, add the Trainer args, the program args, and add the model args
+Now in your main trainer file, add the ``Trainer`` args, the program args, and add the model args
 
 .. testcode::
 
@@ -81,7 +81,7 @@ Now in your main trainer file, add the Trainer args, the program args, and add t
 
     args = parser.parse_args()
 
-Now you can call run your program like so
+Now you can call run your program like so:
 
 .. code-block:: bash
 
@@ -109,12 +109,12 @@ Finally, make sure to start the training like so:
 LightningModule hyperparameters
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 Often times we train many versions of a model. You might share that model or come back to it a few months later
-at which point it is very useful to know how that model was trained (ie: what learning_rate, neural network, etc...).
+at which point it is very useful to know how that model was trained (i.e.: what learning rate, neural network, etc...).
 
 Lightning has a few ways of saving that information for you in checkpoints and yaml files. The goal here is to
 improve readability and reproducibility
 
-1. The first way is to ask lightning to save the values anything in the __init__ for you to the checkpoint. This also
+1. The first way is to ask lightning to save the values of anything in the __init__ for you to the checkpoint. This also
 makes those values available via `self.hparams`.
 
 .. code-block:: python
@@ -198,7 +198,7 @@ In that case, choose only a few
 
 Trainer args
 ^^^^^^^^^^^^
-To recap, add ALL possible trainer flags to the argparser and init the Trainer this way
+To recap, add ALL possible trainer flags to the argparser and init the ``Trainer`` this way
 
 .. code-block:: python
 
@@ -217,7 +217,7 @@ Multiple Lightning Modules
 ^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 We often have multiple Lightning Modules where each one has different arguments. Instead of
-polluting the main.py file, the LightningModule lets you define arguments for each one.
+polluting the ``main.py`` file, the ``LightningModule`` lets you define arguments for each one.
 
 .. testcode::
 

docs/source/lr_finder.rst

@@ -53,7 +53,7 @@ which can be accessed via ``self.learning_rate`` or ``self.lr``.
 
     trainer.tune(model)
 
-If your model is using an arbitrary value instead of ``self.lr`` or ``self.learning_rate``, set that value as auto_lr_find
+If your model is using an arbitrary value instead of ``self.lr`` or ``self.learning_rate``, set that value as ``auto_lr_find``:
 
 .. code-block:: python
 

docs/source/metrics.rst

@@ -84,7 +84,7 @@ To implement your custom metric, subclass the base ``Metric`` class and implemen
 - ``update()``: Any code needed to update the state given any inputs to the metric.
 - ``compute()``: Computes a final value from the state of the metric.
 
-All you need to do is call add_state correctly to implement a custom metric with DDP.
+All you need to do is call ``add_state`` correctly to implement a custom metric with DDP.
 ``reset()`` is called on metric state variables added using ``add_state()``.
 
 To see how metric states are synchronized across distributed processes, refer to ``add_state()`` docs

docs/source/multiple_loaders.rst

@@ -8,16 +8,16 @@ Multiple Datasets
 =================
 Lightning supports multiple dataloaders in a few ways.
 
-1. Create a dataloader that iterates both datasets under the hood.
+1. Create a dataloader that iterates multiple datasets under the hood.
 2. In the validation and test loop you also have the option to return multiple dataloaders
    which lightning will call sequentially.
 
 ----------
 
 Multiple training dataloaders
 -----------------------------
-For training, the best way to use multiple-dataloaders is to create a Dataloader class
-which wraps both your dataloaders. (This of course also works for testing and validation
+For training, the best way to use multiple dataloaders is to create a ``DataLoader`` class
+which wraps your multiple dataloaders (this of course also works for testing and validation
 dataloaders).
 
 (`reference <https://discuss.pytorch.org/t/train-simultaneously-on-two-datasets/649/2>`_)
@@ -63,8 +63,8 @@ dataloaders).
 
 Test/Val dataloaders
 --------------------
-For validation, test dataloaders lightning also gives you the additional
-option of passing in multiple dataloaders back from each call.
+For validation and test dataloaders, lightning also gives you the additional
+option of passing multiple dataloaders back from each call.
 
 See the following for more details:
 

pytorch_lightning/metrics/functional/__init__.py

@@ -29,3 +29,6 @@
     rmsle,
     ssim
 )
+from pytorch_lightning.metrics.functional.self_supervised import (
+    embedding_similarity
+)