fitness level added

Author: yahocan

geneticAlgorithm.py

@@ -88,24 +88,34 @@ def run_simulation(p_cross, p_mut):
         total_Q_history.append(total_Q)
         variance_history.append(Variance(outputs_list))
 
-        # A. RECOMBINATION (Tournament Selection)
-        mating_pool = []
+        # A. RECOMBINATION
+        # Calculate fitness scores (profits) for all firms
+        fitness_scores = []
+        for firm in Population:
+            q_i = BinaryToOutput(firm)
+            profit = CalculateProfit(q_i, current_price)
+            fitness_scores.append(profit)
 
-        for _ in range(n):
-            # Pick two random firms
-            firm_a = random.choice(Population)
-            firm_b = random.choice(Population)
+        # Check if we can use fitness-based selection
+        min_fitness = min(fitness_scores)
+        total_fitness = sum(fitness_scores)
 
-            q_a = BinaryToOutput(firm_a)
-            q_b = BinaryToOutput(firm_b)
+        # If negative fitness or zero total fitness, restart population
+        if min_fitness < 0 or total_fitness <= 0:
+            Population = []
+            for i in range(n):
+                genes = [random.randint(0, 1) for _ in range(l)]
+                Population.append(genes)
+            continue
 
-            profit_a = CalculateProfit(q_a, current_price)
-            profit_b = CalculateProfit(q_b, current_price)
+        # Normalize to get probabilities
+        probabilities = [f / total_fitness for f in fitness_scores]
 
-            if profit_a > profit_b:
-                mating_pool.append(firm_a.copy())
-            else:
-                mating_pool.append(firm_b.copy())
+        # Select firms based on their fitness probabilities
+        mating_pool = []
+        for _ in range(n):
+            selected_firm_idx = np.random.choice(len(Population), p=probabilities)
+            mating_pool.append(Population[selected_firm_idx].copy())
 
         # B. CROSSOVER
         offspring_pool = []