Merge pull request #1399 from Axelrod-Python/marcharper-patch-3

Add a thesis to citations

Author: meatballs

axelrod/compute_finite_state_machine_memory.py

@@ -113,7 +113,10 @@ def get_accessible_transitions(
         if trans.state in accessible_states:
             accessible_transitions[
                 (trans.state, trans.last_opponent_action)
-            ] = (trans.next_state, trans.next_action)
+            ] = (
+                trans.next_state,
+                trans.next_action,
+            )
 
     return accessible_transitions
 

axelrod/fingerprint.py

@@ -42,7 +42,7 @@ def _create_points(step: float, progress_bar: bool = True) -> List[Point]:
     num = int((1 / step) // 1) + 1
 
     if progress_bar:
-        p_bar = tqdm.tqdm(total=num ** 2, desc="Generating points")
+        p_bar = tqdm.tqdm(total=num**2, desc="Generating points")
 
     points = []
     for x in np.linspace(0, 1, num):

axelrod/random_.py

@@ -29,7 +29,7 @@ def randint(self, *args, **kwargs):
         return self._random.randint(*args, **kwargs)
 
     def random_seed_int(self) -> int:
-        return self.randint(low=0, high=2 ** 32 - 1, dtype="uint64")
+        return self.randint(low=0, high=2**32 - 1, dtype="uint64")
 
     def choice(self, *args, **kwargs):
         return self._random.choice(*args, **kwargs)
@@ -132,7 +132,7 @@ def _fill_ints(self):
         # Generate more random values. Store as a list since generators
         # cannot be pickled.
         self._ints = self._random_generator.randint(
-            low=0, high=2 ** 32 - 1, size=self._batch_size, dtype="uint64"
+            low=0, high=2**32 - 1, size=self._batch_size, dtype="uint64"
         )
         self._index = 0
 

axelrod/strategies/hmm.py

@@ -327,7 +327,7 @@ def _normalize_parameters(
         emission_probabilities = list(map(float, emission_probabilities))
         num_states = len(emission_probabilities)
         if mutation_probability is None:
-            mutation_probability = 10 / (num_states ** 2)
+            mutation_probability = 10 / (num_states**2)
         else:
             mutation_probability = mutation_probability
         return (
@@ -432,7 +432,7 @@ class with self.num_states.
         entry is the initial_action.
         """
 
-        assert len(vector) == 2 * self.num_states ** 2 + self.num_states + 1
+        assert len(vector) == 2 * self.num_states**2 + self.num_states + 1
 
         def deserialize(vector):
             matrix = []
@@ -442,9 +442,9 @@ def deserialize(vector):
                 matrix.append(row)
             return matrix
 
-        break_tc = self.num_states ** 2
-        break_td = 2 * self.num_states ** 2
-        break_ep = 2 * self.num_states ** 2 + self.num_states
+        break_tc = self.num_states**2
+        break_td = 2 * self.num_states**2
+        break_ep = 2 * self.num_states**2 + self.num_states
         initial_state = 0
         self.hmm = SimpleHMM(
             deserialize(vector[0:break_tc]),
@@ -457,7 +457,7 @@ def deserialize(vector):
 
     def create_vector_bounds(self):
         """Creates the bounds for the decision variables."""
-        vec_len = 2 * self.num_states ** 2 + self.num_states + 1
+        vec_len = 2 * self.num_states**2 + self.num_states + 1
         lb = [0.0] * vec_len
         ub = [1.0] * vec_len
         return lb, ub

axelrod/tests/strategies/test_hmm.py

@@ -266,7 +266,7 @@ def test_vector_to_instance(self):
 
     def test_create_vector_bounds(self):
         num_states = 4
-        size = 2 * num_states ** 2 + num_states + 1
+        size = 2 * num_states**2 + num_states + 1
 
         player = self.player_class(num_states=num_states, seed=1)
         lb, ub = player.create_vector_bounds()

axelrod/tests/strategies/test_player.py

@@ -138,7 +138,7 @@ def test_clone(self):
         player2 = player1.clone()
         turns = 50
         for op in [axl.Cooperator(), axl.Defector(), axl.TitForTat()]:
-            seed = random.randint(0, 10 ** 6)
+            seed = random.randint(0, 10**6)
             for p in [player1, player2]:
                 m = axl.Match((p, op), turns=turns, reset=True, seed=seed)
                 m.play()

citations.md

@@ -1,7 +1,6 @@
 # Citations
 
 ### 2021
-
 * Glynatsi, Nikoleta E., and Vincent A. Knight. "A bibliometric study of research topics, collaboration, and centrality in the iterated prisoner’s dilemma." Humanities and Social Sciences Communications 8.1 (2021): 1-12. [Nature.com](https://www.nature.com/articles/s41599-021-00718-9) [ArXiv](https://arxiv.org/abs/1911.06128)
 * Glynatsi, Nikoleta E., and Vincent A. Knight. "Game Theory and Python: An educational tutorial to game theory and repeated games using Python." Journal of Open Source Education, 4(39), 78. [jose.theoj.org](https://doi.org/10.21105/jose.00078)
 
@@ -47,6 +46,7 @@
 
 ### Disserations
 
+* Dollbo, Anna. MIXED MEMORY Q-LEARNER: An adaptive reinforcement learning algorithm for the Iterated Prisoner’s Dilemma. [BS thesis](https://gupea.ub.gu.se/handle/2077/69664). 2021.
 * Glynatsi, Nikoleta. "Understanding responses to environments for the Prisoner's Dilemma: A meta analysis, multidimensional optimisation and machine learning approach". PhD Thesis, Cardiff University. (2020) [Available Online](http://orca.cf.ac.uk/135221/).
 * Santana, Jonas Cardoso Carvalho. "Aprendizado de máquina e o dilema dos prisioneiros". (2020)
 * Black, Ben. "An Empirical Evaluation of the Ohtsuki-Nowak Approximation." [Master's Thesis](https://www.lancaster.ac.uk/~blackb/documents/MScDiss.pdf). (2018).