add more detail to tbptt example (#755)

* add more detail to tbptt example

* warn user about new arg in training_step

pytorch_lightning/core/lightning.py

@@ -168,6 +168,14 @@ class LightningModule(ABC, GradInformation, ModelIO, ModelHooks):
             # Truncated back-propagation through time
             def training_step(self, batch, batch_idx, hiddens):
                 # hiddens are the hiddens from the previous truncated backprop step
+                ...
+                out, hiddens = self.lstm(data, hiddens)
+                ...
+
+                return {
+                    "loss": ...,
+                    "hiddens": hiddens  # remember to detach() this
+                }
 
         You can also return a -1 instead of a dict to stop the current loop. This is useful
          if you want to break out of the current training epoch early.

pytorch_lightning/trainer/trainer.py

@@ -448,6 +448,10 @@ class Trainer(TrainerIOMixin,
                     # backprop every 5 steps in a batch
                     trainer = Trainer(truncated_bptt_steps=5)
 
+                Using this feature requires updating your LightningModule's `training_step()` to include
+                a `hiddens` arg.
+
+
             resume_from_checkpoint (str): To resume training from a specific checkpoint pass in the path here.k
                 Example::