Added mutation factor

brain.py

@@ -31,10 +31,10 @@ class Brain:
             clone.directions[i] = copy.copy(self.directions[i])
         return clone
 
-    # mutates the brain by setting some of the directions to random movements
-    def mutate(self):
+
+    def mutate(self, mutation_factor):  # mutates the brain by setting some of the directions to random movements
         for i in range(len(self.directions)):
-            if random.random() < 0.1:  # changes roughly 10% of the movements
+            if random.random() <= mutation_factor:
                 self.directions[i] = np.random.uniform(-1, 1, 4)
 
     def save(self):

main.py

@@ -5,11 +5,13 @@ import matplotlib.pyplot as plt
 INCREASE_BY = 5
 BRAIN_SIZE = 50
 POP_SIZE = 50
+MUTATION_FACTOR = 0.1  # 0 <= x <= 1
 GAME_CANCELLED = False
 LOAD_BRAIN = True
+RENDER_BEST = False
 
 if __name__ == '__main__':
-    population = Population(POP_SIZE, BRAIN_SIZE, LOAD_BRAIN)
+    population = Population(POP_SIZE, BRAIN_SIZE,MUTATION_FACTOR, LOAD_BRAIN, RENDER_BEST)
 
     while GAME_CANCELLED is False:  # this is our game
         if population.all_players_finished():  # this is our genetic algorithm after one generation of players

population.py

@@ -9,9 +9,10 @@ MAX_STEPS = 1599  # after 1600 steps the Environment gives us a done anyway.
 
 class Population:
 
-    def __init__(self, size, brain_size, load_brain):
+    def __init__(self, size, brain_size, mutation_factor, load_brain, render_best):
         self.size = size
         self.brain_size = brain_size
+        self.mutation_factor = mutation_factor
         self.fitness_sum = 0.0
         self.gen = 1
         self.best_walker_index = 0  # index of the best player in self.players
@@ -21,7 +22,7 @@ class Population:
         self.envs = []
         self.fitnesses = None
         for i in range(self.size):
-            self.walkers.append(Walker(self.brain_size, load_brain))
+            self.walkers.append(Walker(self.brain_size, load_brain, render_best))
         self.reset_environments()
         if load_brain:
             self.mutate_babies()
@@ -89,7 +90,7 @@ class Population:
 
     def mutate_babies(self):  # mutates all the brains of the babies
         for i in range(1, len(self.walkers)):  # we don't want to mutate the champion's brain
-            self.walkers[i].brain.mutate()
+            self.walkers[i].brain.mutate(self.mutation_factor)
 
     def set_best_walker(self):  # finds the player with the highest fitness and sets it as the best one
         max_index = np.argmax(self.fitnesses)

walker.py

@@ -3,13 +3,14 @@ import gym
 
 class Walker:
 
-    def __init__(self, brain_size, load_brain):
+    def __init__(self, brain_size, load_brain, render_best):
         self.brain = Brain(brain_size, load_brain)  # new brain with X instructions
         self.dead = False
         self.reached_goal = False
         self.is_best = False  # true if this dot is the best dot from the previous generation
         self.fitness = 0.0
         self.env = gym.make('BipedalWalker-v3')
+        self.render_best = render_best
         # self.pos = copy.copy(self.map.startpoint)
 
     def update(self):  # moves the dot according to the brains directions
@@ -25,7 +26,7 @@ class Walker:
         elif done is True:
             self.reached_goal = True
             self.fitness += 10000000
-        if self.is_best:
+        if self.is_best and self.render_best:
             self.env.render()
 
     """ def get_fitness(self):
@@ -40,7 +41,7 @@ class Walker:
         self.env.reset()
 
     def get_baby(self):
-        baby = Walker(0, False)
+        baby = Walker(0, False, self.render_best)
         baby.brain = self.brain.clone()  # babies have the same brain as their parents
         self.env.close()
         return baby