Comment wrong in adversarial_crypto model

According to the code below:
```python
if key is not None:
      combined_message = tf.concat(axis=1, values=[message, key])
else:
      combined_message = message
```python
If the key=None, combined_message is just message, not the key.

adversarial_crypto/train_eval.py

@@ -118,7 +118,7 @@ class AdversarialCrypto(object):
 
   def model(self, collection, message, key=None):
     """The model for Alice, Bob, and Eve.  If key=None, the first FC layer
-    takes only the Key as inputs.  Otherwise, it uses both the key
+    takes only the message as inputs.  Otherwise, it uses both the key
     and the message.
 
     Args: