Update ptcheat.rst (#1083)

scheduler.step() should not be called at the start of every epoch as of PyTorch 1.1.0. Instead it should be called after the optimizer has updated the weights (after optimizer.step() is called)

Co-authored-by: Brian Johnson <brianjo@fb.com>
Co-authored-by: holly1238 <77758406+holly1238@users.noreply.github.com>

Author: 3 people

beginner_source/ptcheat.rst

@@ -237,7 +237,7 @@ Learning rate scheduling
 .. code-block:: python
 
     scheduler = optim.X(optimizer,...)      # create lr scheduler
-    scheduler.step()                        # update lr at start of epoch
+    scheduler.step()                        # update lr after optimizer updates weights
     optim.lr_scheduler.X                    # where X is LambdaLR, MultiplicativeLR,
                                             # StepLR, MultiStepLR, ExponentialLR,
                                             # CosineAnnealingLR, ReduceLROnPlateau, CyclicLR,