added pytorch lightning integration docs

docs/index.rst

@@ -120,6 +120,7 @@ Documentation contents
    TensorBoard <integrations/tensorboard.rst>
    MLflow <integrations/mlflow.rst>
    Fast.ai <integrations/fast_ai.rst>
+   PyTorchLightning <integrations/pytorch_lightning.rst>
    Neptune Contrib <integrations/neptune-contrib.rst>
 
 .. External links

docs/integrations/pytorch_lightning.rst

@@ -0,0 +1,207 @@
+Log PyTorchLightning metrics to neptune
+=======================================
+.. image:: ../_static/images/others/pytorchlightning_neptuneai.png
+   :target: ../_static/images/others/pytorchlightning_neptuneai.png
+   :alt: PyTorchLightning neptune.ai integration
+
+Prerequisites
+-------------
+Integration with |PyTorchLightning| framework is introduced as a part of logging module so just need to have |neptune-client| installed.
+
+.. code-block:: bash
+
+    pip install neptune-client
+
+
+Create the **LightningModule**
+------------------------------
+.. code-block:: python3
+
+    import os
+
+    import torch
+    from torch.nn import functional as F
+    from torch.utils.data import DataLoader
+    from torchvision.datasets import MNIST
+    from torchvision import transforms
+
+    import pytorch_lightning as pl
+
+
+    class CoolSystem(pl.LightningModule):
+
+        def __init__(self):
+            super(CoolSystem, self).__init__()
+            # not the best model...
+            self.l1 = torch.nn.Linear(28 * 28, 10)
+
+        def forward(self, x):
+            return torch.relu(self.l1(x.view(x.size(0), -1)))
+
+        def training_step(self, batch, batch_idx):
+            # REQUIRED
+            x, y = batch
+            y_hat = self.forward(x)
+            loss = F.cross_entropy(y_hat, y)
+            tensorboard_logs = {'train_loss': loss}
+            return {'loss': loss, 'log': tensorboard_logs}
+
+        def validation_step(self, batch, batch_idx):
+            # OPTIONAL
+            x, y = batch
+            y_hat = self.forward(x)
+            return {'val_loss': F.cross_entropy(y_hat, y)}
+
+        def validation_end(self, outputs):
+            # OPTIONAL
+            avg_loss = torch.stack([x['val_loss'] for x in outputs]).mean()
+            tensorboard_logs = {'val_loss': avg_loss}
+            return {'avg_val_loss': avg_loss, 'log': tensorboard_logs}
+
+        def configure_optimizers(self):
+            # REQUIRED
+            # can return multiple optimizers and learning_rate schedulers
+            # (LBFGS it is automatically supported, no need for closure function)
+            return torch.optim.Adam(self.parameters(), lr=0.02)
+
+        @pl.data_loader
+        def train_dataloader(self):
+            # REQUIRED
+            return DataLoader(MNIST(os.getcwd(), train=True, download=True, transform=transforms.ToTensor()), batch_size=32)
+
+        @pl.data_loader
+        def val_dataloader(self):
+            # OPTIONAL
+            return DataLoader(MNIST(os.getcwd(), train=True, download=True, transform=transforms.ToTensor()), batch_size=32)
+
+Create the **NeptuneLogger** with all the information you want to track
+------------------------------------------------------------------------
+.. code-block:: python3
+
+    from pytorch_lightning.logging.neptune import NeptuneLogger
+
+    neptune_logger = NeptuneLogger(
+        api_key=os.environ["NEPTUNE_API_TOKEN"],
+        project_name="jakub-czakon/examples",
+        experiment_name="default",  # Optional,
+        params={"max_epochs": 10,
+                "batch_size": 32},  # Optional,
+        tags=["pytorch-lightning", "mlp"]  # Optional,
+    )
+
+Create the **Trainer** and pass **neptune_logger** to logger
+------------------------------------------------------------
+.. code-block:: python3
+
+    from pytorch_lightning import Trainer
+
+    trainer = Trainer(max_epochs=10, logger=neptune_logger)
+    trainer.fit(CoolSystem())
+
+Monitor your PyTorchLightning training in Neptune
+--------------------------------------------------
+Now you can watch your pytorch-lightning model training in neptune!
+
+.. image:: ../_static/images/pytorch_lightning/pytorch_lightning_monitoring.gif
+   :target: ../_static/images/pytorch_lightning/pytorch_lightning_monitoring.gif
+   :alt: PyTorchLightning logging in neptune
+
+Full PyTorchLightning monitor script
+------------------------------------
+Simply copy and paste it to ``pytorch_lightning_example.py`` and run.
+Remember to change your credentials in the **NeptuneLogger**:
+
+.. code-block:: python3
+
+    neptune_logger = NeptuneLogger(
+        api_key=os.environ["NEPTUNE_API_TOKEN"],
+        project_name="USERNAME/PROJECT_NAME",
+        ...
+        )
+
+.. code-block:: python3
+
+    import os
+
+    import torch
+    from torch.nn import functional as F
+    from torch.utils.data import DataLoader
+    from torchvision.datasets import MNIST
+    from torchvision import transforms
+
+    import pytorch_lightning as pl
+
+
+    class CoolSystem(pl.LightningModule):
+
+        def __init__(self):
+            super(CoolSystem, self).__init__()
+            # not the best model...
+            self.l1 = torch.nn.Linear(28 * 28, 10)
+
+        def forward(self, x):
+            return torch.relu(self.l1(x.view(x.size(0), -1)))
+
+        def training_step(self, batch, batch_idx):
+            # REQUIRED
+            x, y = batch
+            y_hat = self.forward(x)
+            loss = F.cross_entropy(y_hat, y)
+            tensorboard_logs = {'train_loss': loss}
+            return {'loss': loss, 'log': tensorboard_logs}
+
+        def validation_step(self, batch, batch_idx):
+            # OPTIONAL
+            x, y = batch
+            y_hat = self.forward(x)
+            return {'val_loss': F.cross_entropy(y_hat, y)}
+
+        def validation_end(self, outputs):
+            # OPTIONAL
+            avg_loss = torch.stack([x['val_loss'] for x in outputs]).mean()
+            tensorboard_logs = {'val_loss': avg_loss}
+            return {'avg_val_loss': avg_loss, 'log': tensorboard_logs}
+
+        def configure_optimizers(self):
+            # REQUIRED
+            # can return multiple optimizers and learning_rate schedulers
+            # (LBFGS it is automatically supported, no need for closure function)
+            return torch.optim.Adam(self.parameters(), lr=0.02)
+
+        @pl.data_loader
+        def train_dataloader(self):
+            # REQUIRED
+            return DataLoader(MNIST(os.getcwd(), train=True, download=True, transform=transforms.ToTensor()), batch_size=32)
+
+        @pl.data_loader
+        def val_dataloader(self):
+            # OPTIONAL
+            return DataLoader(MNIST(os.getcwd(), train=True, download=True, transform=transforms.ToTensor()), batch_size=32)
+
+
+    from pytorch_lightning.logging.neptune import NeptuneLogger
+
+    neptune_logger = NeptuneLogger(
+        api_key=os.environ["NEPTUNE_API_TOKEN"],
+        project_name="USERNAME/PROJECT_NAME",
+        experiment_name="default",  # Optional,
+        params={"max_epochs": 4,
+                "batch_size": 32},  # Optional,
+        tags=["pytorch-lightning", "mlp"]  # Optional,
+    )
+
+    from pytorch_lightning import Trainer
+
+    trainer = Trainer(max_epochs=4, logger=neptune_logger)
+    trainer.fit(CoolSystem())
+
+
+.. External links
+
+.. |PyTorchLightning| raw:: html
+
+    <a href="https://github.com/PyTorchLightning/pytorch-lightning" target="_blank">PyTorchLightning</a>
+
+.. |neptune-client| raw:: html
+
+    <a href="https://github.com/neptune-ml/neptune-client" target="_blank">neptune-client</a>