two docstring changes

axelrod/strategy_transformers.py

@@ -200,7 +200,7 @@ def generic_strategy_wrapper(player, opponent, proposed_action, *args,
 
 
 def flip_wrapper(player, opponent, action):
-    """Applies Action.flip() at the class level."""
+    """Flips the player's original actions."""
     return action.flip()
 
 
@@ -240,7 +240,7 @@ def dual_wrapper(player, opponent, proposed_action):
 
 
 def noisy_wrapper(player, opponent, action, noise=0.05):
-    """Applies Action.flip() at the class level."""
+    """Flips the player's actions with probability: `noise`."""
     r = random.random()
     if r < noise:
         return action.flip()

docs/tutorials/further_topics/noisy_tournaments.rst

@@ -3,7 +3,7 @@ Noisy tournaments
 
 A common variation on iterated prisoner’s dilemma tournaments is to add
 stochasticity in the choice of actions, simply called noise. This noise is
-introduced by flipping plays between ‘C’ and ‘D’ with some probability that is
+introduced by flipping plays between C and D with some probability that is
 applied to all plays after they are delivered by the player [Bendor1993]_.
 
 The presence of this persistent background noise causes some strategies to