Working refactoring.

Author: skytreader

python/stochastic/common.py

@@ -1,6 +1,7 @@
 #! /usr/bin/env python3
 
 import math
+import random
 
 """
 Not in the original code listing. This factors out the common functions
@@ -11,3 +12,33 @@ def euc_2d(p1, p2):
 	dx = p1[0] - p2[0]
 	dy = p1[1] - p2[1]
 	return math.sqrt((dx ** 2) + (dy ** 2))
+
+def random_permutation(cities):
+    """
+    Given a list of cities, return a random permutation _of the indices_.
+    """
+    indices = [idx for idx, val in enumerate(cities)]
+    for i in indices:
+        # Guaranteed not to choose i as well
+        r = random.randint(0, len(cities) - 1 - i) + i
+        indices[r], indices[i] = indices[i], indices[r]
+
+    return indices
+
+def path_cost(permutation, cities):
+    """
+    Given a permutation of indices and the actual location of the cities,
+    the total cost (assessed as Euclidean distance) of traversing the path
+    described by the permutation.
+    """
+    distance = 0
+    limit = len(permutation)
+
+    for i in range(limit):
+        if i == (limit - 1):
+            c2 = 0
+        else:
+            c2 = i + 1
+        distance += euc_2d(cities[permutation[i]], cities[c2])
+
+    return distance

python/stochastic/iterated_local_search.py

@@ -1,6 +1,6 @@
 #! usr/bin/env python3
 
-from .common import euc_2d
+from .common import path_cost, random_permutation
 
 import math
 import random
@@ -23,29 +23,6 @@
 7542 units.
 """
 
-def cost(permutation, cities):
-    distance = 0
-    limit = len(permutation)
-    
-    for i in range(limit):
-        if i == (limit - 1):
-            c2 = permutation[0]
-        else:
-            c2 = permutation[i + 1]
-        
-        distance += euc_2d(cities[permutation[i]], cities[c2])
-    
-    return distance
-
-def random_permutation(cities):
-    perm = [i for i in range(len(cities))]
-    
-    for i in perm:
-        r = random.randint(0, len(perm) - 1 - i) + i
-        perm[r], perm[i] = perm[i], perm[r]
-    
-    return perm
-
 def stochastic_two_opt(permutation):
     perm = [permutation[i] for i in range(len(permutation))]
     upper_bound = len(perm) - 1
@@ -81,7 +58,7 @@ def local_search(best, cities, max_no_improv):
     while count < max_no_improv:
         candidate = {}
         candidate["vector"] = stochastic_two_opt(best["vector"])
-        candidate["cost"] = cost(candidate["vector"], cities)
+        candidate["cost"] = path_cost(candidate["vector"], cities)
 
         if candidate["cost"] < best["cost"]:
             count = 0
@@ -104,13 +81,13 @@ def double_bridge_move(perm):
 def perturbation(cities, best):
     candidate = {}
     candidate["vector"] = double_bridge_move(best["vector"])
-    candidate["cost"] = cost(candidate["vector"], cities)
+    candidate["cost"] = path_cost(candidate["vector"], cities)
     return candidate
 
 def search(cities, max_iterations, max_no_improv):
     best = {}
     best["vector"] = random_permutation(cities)
-    best["cost"] = cost(best["vector"], cities)
+    best["cost"] = path_cost(best["vector"], cities)
     best = local_search(best, cities, max_no_improv)
 
     for i in range(max_iterations):    

python/stochastic/stochastic_hill_climbing.py

@@ -21,8 +21,6 @@
 Implementation notes:
 The reference implementation uses a list of (one-character) strings.
 I opted to use a String object directly.
-
-@author Chad Estioco
 """
 
 def onemax(vector):

python/stochastic/variable_local_search.py

@@ -13,10 +13,3 @@
 global minimum is local minimum for all possible neighborhood structures, and
 3) local minima are relatively close to global minima for many problem classes.
 """
-
-def cost(permutation, cities):
-    distance = 0
-    limit = len(permutation)
-    
-    for idx, perm:
-        c2 = permutation[idx % limit]