Added kwargs

model/pytorch/dcrnn_supervisor.py

@@ -121,7 +121,7 @@ class DCRNNSupervisor:
 
     def _train(self, base_lr,
                steps, patience=50, epochs=100, lr_decay_ratio=0.1, log_every=1, save_model=1,
-               test_every_n_epochs=10):
+               test_every_n_epochs=10, **kwargs):
         # steps is used in learning rate - will see if need to use it?
         min_val_loss = float('inf')
         wait = 0