Add strategies TF1, TF2, and TF3 #1036
Conversation
|
OK, I'll bite. What does the 'TF' stand for? |
Trained FSM (finite state machine): they're the 3 strategies trained for the moran paper :) |
|
|
||
| Names: | ||
|
|
||
| - TF1: Original name by Marc Harper |
There was a problem hiding this comment.
We should remember to include the moran paper as a reference for this once we have a title etc...
|
Aha! I hadn't made the connection between 'F' and 'FSM.' Must be time for further coffee.... |
| } | ||
|
|
||
| def test_strategy(self): | ||
| self.first_play_test(C) |
There was a problem hiding this comment.
As a regression test could you include a 5 turn play against alternator? (Same for the other 2)
| name = "TF1" | ||
| player = axelrod.TF1 | ||
| expected_classifier = { | ||
| 'memory_depth': 16, |
| name = "TF2" | ||
| player = axelrod.TF2 | ||
| expected_classifier = { | ||
| 'memory_depth': 16, |
| name = "TF3" | ||
| player = axelrod.TF3 | ||
| expected_classifier = { | ||
| 'memory_depth': 8, |
|
|
||
| name = "TF3" | ||
| classifier = { | ||
| 'memory_depth': 8, # Estimate |
There was a problem hiding this comment.
float('inf')
(explanation: Axelrod-Python/axelrod-moran#42).
|
|
||
| name = "TF2" | ||
| classifier = { | ||
| 'memory_depth': 16, # Estimate |
There was a problem hiding this comment.
float('inf') (not an estimate)
|
|
||
| name = "TF1" | ||
| classifier = { | ||
| 'memory_depth': 16, # Estimate |
There was a problem hiding this comment.
float('inf') and no need for estimate: it has a handshake.
| (2, C, 14, D), (2, D, 9, D), | ||
| (3, C, 0, C), (3, D, 1, D), | ||
| (4, C, 1, D), (4, D, 2, D), | ||
| # (5, C, 12, C), (5, D, 6, C), |
There was a problem hiding this comment.
I confirm these aren't reachable. Do we want to remove them (as opposed to having them commented out)?
There was a problem hiding this comment.
I left them in for historical reference, I guess it doesn't really matter since they are impossible states.
There was a problem hiding this comment.
Perhaps add another line comment saying: # Unreachable states?
I'm not overly bothered just trying to think of whether or not this will confuse future us.
There was a problem hiding this comment.
Just took them out. They are in the Moran repository if we ever need them for any reason.
|
Added tests and updated memory_depths |
|
Closes #884 |
Trained strategies used in the Moran paper.