genetical_algorithm.py

@@ -1,3 +1,8 @@
+import random
+from copy import copy
+import settings
+import tree
+
 
 def evaluate_values(fuel, water, feritizer, seeds, fields_with_plants, k):
     fields_to_sow = 0
@@ -61,3 +66,56 @@ def fitness(solution, fields_plants, k):
         ans = evaluate_values(fuel, water, feritizer, seeds, fields_plants, k)
     return ans
 
+
+def solution(fields_with_plants, l):
+    solutions = []
+    ranked_solutions = []
+    for i in range(10000):
+        solution = []
+        fuel = random.randint(0, 2000)
+        water = random.randint(0, 100)
+        feritizer = random.randint(0, 100)
+        seeds = random.randint(0, 100)
+        solution.append(fuel)
+        solution.append(water)
+        solution.append(feritizer)
+        solution.append(seeds)
+        solutions.append(solution)
+    for s in solutions:
+        ranked_solutions.append((fitness(s, fields_with_plants, l), s))
+    ranked_solutions.sort()
+    ranked_solutions.reverse()
+    best_solutions = ranked_solutions[:100]
+    k = 1
+    print("Gen 1 best solution: " + str(ranked_solutions[0]))
+    for q in range(4):
+        k += 1
+        random_cross = random.randint(0, 3)
+        random_mutation_place = random.randint(0, 3)
+        random_mutation = random.randint(1, 5)
+        ranked_solutions = []
+        solutions = []
+        solutions.append(best_solutions[0][1])
+        for i in range(100):
+            for j in range(100):
+                if i == j:
+                    continue
+                random_mutation_chance = random.randint(1, 100)
+                solution = copy(best_solutions[i][1])
+                solution[random_cross:4] = best_solutions[j][1][random_cross:4]
+                if random_mutation_chance <= 3:
+                    if random_mutation_place == 0:
+                        solution[random_mutation_place] = solution[random_mutation_place] + random_mutation * 30
+                    else:
+                        solution[random_mutation_place] = solution[random_mutation_place] + random_mutation
+                solutions.append(copy(solution))
+
+        for s in solutions:
+            ranked_solutions.append((fitness(s, fields_with_plants, l), s))
+        ranked_solutions.sort()
+        ranked_solutions.reverse()
+
+        best_solutions = ranked_solutions[:100]
+        print("Gen " + str(k) + " best solution: " + str(ranked_solutions[0]))
+
+    return ranked_solutions[0][1][0], ranked_solutions[0][1][1], ranked_solutions[0][1][2], ranked_solutions[0][1][3]