Removed unnecessary imports

OleksandrKlanovets · Nov 2, 2020 · 7e2bcf9 · 7e2bcf9
1 parent d2db0b2
commit 7e2bcf9
Show file tree

Hide file tree

Showing 3 changed files with 25 additions and 29 deletions.
diff --git a/models/single_objective/bat_search.py b/models/single_objective/bat_search.py
@@ -1,5 +1,4 @@
 import numpy as np
-import random
 import math
 from collections import namedtuple
 from models.swarm_algorithm import SwarmAlgorithm
@@ -26,15 +25,15 @@ class BatSearch(SwarmAlgorithm):
     params : BatSearchParams
         Model behavioral parameters.
     bounds : ndarray
-        A 2 by D matrix containing lower and upper bounds of the search space 
+        A 2 by D matrix containing lower and upper bounds of the search space
         for each dimension.
     seed : int, optional, default=None
         Random generator seed.
     max_iter : int, optional, default=100
         Maximum number of iterations (generations).
     stag_iter : int, optional, default=100
         Specifies the allowed number of iterations without solution improvement
-        by equal or more than a given tolerance. If the number is exceeded, 
+        by equal or more than a given tolerance. If the number is exceeded,
         the optimization process stagnations occurs and the algorithm stops.
     e : float, optional, default=1e-5
         Tolerance.
@@ -61,7 +60,7 @@ class BatSearch(SwarmAlgorithm):
         An array of size N representing the value of the fitness function
         for each particle.
     gbest : ndarray
-        A D-dimensional vector representing the position of the current 
+        A D-dimensional vector representing the position of the current
         global best particle.
     gbest_score : float
         The value of the fitness function for the current global best particle.
@@ -148,7 +147,7 @@ def __local_search(self, new_solutions):
         No value.
         '''
         apply_indices = np.random.rand(self.N) > self.rates
-        apply_size = np.sum(apply_indices) 
+        apply_size = np.sum(apply_indices)
 
         rand_coefs = np.random.normal(size=(apply_size, self.D))
         new_solutions[apply_indices] = np.copy(self.gbest)
@@ -168,7 +167,7 @@ def optimize(self):
         -------
         ndarray
             The coordinates of the global best particle at the end of
-            the optimization process. 
+            the optimization process.
         '''
         i = 0
         stag_count = 0
@@ -188,7 +187,7 @@ def optimize(self):
             # Increase rate and reduce loudness.
             self.loudness[apply_indices] *= self.alpha
             self.rates[apply_indices] = self.r_0 * (1 - math.exp(-self.gamma * (i + 1)))
-                
+
             self.update_best()
             self.eval_num += self.N
             i += 1

diff --git a/models/single_objective/cuckoo_search.py b/models/single_objective/cuckoo_search.py
@@ -25,15 +25,15 @@ class CuckooSearch(SwarmAlgorithm):
     params : CuckooSearchParams
         Model behavioral parameters.
     bounds : ndarray
-        A 2 by D matrix containing lower and upper bounds of the search space 
+        A 2 by D matrix containing lower and upper bounds of the search space
         for each dimension.
     seed : int, optional, default=None
         Random generator seed.
     max_iter : int, optional, default=100
         Maximum number of iterations (generations).
     stag_iter : int, optional, default=100
         Specifies the allowed number of iterations without solution improvement
-        by equal or more than a given tolerance e. If the number is exceeded, 
+        by equal or more than a given tolerance e. If the number is exceeded,
         the optimization process stagnation occurs and the algorithm stops.
     e : float, optional, default=1e-5
         Tolerance.
@@ -50,7 +50,7 @@ class CuckooSearch(SwarmAlgorithm):
         An array of size N representing the value of the fitness function
         for each particle.
     gbest : ndarray
-        A D-dimensional vector representing the position of the current 
+        A D-dimensional vector representing the position of the current
         global best particle.
     gbest_score : float
         The value of the fitness function for the current global best particle.
@@ -59,12 +59,12 @@ class CuckooSearch(SwarmAlgorithm):
     '''
     def __init__(self, D, N, fit_func, params, bounds, seed=None, max_iter=100,
                  stag_iter=100, e=0.00001):
-        super().__init__(D, N, fit_func, params, bounds, seed, max_iter, 
+        super().__init__(D, N, fit_func, params, bounds, seed, max_iter,
                          stag_iter, e)
 
         # Levy flight's coefficient calculation.
         self.beta = 1.5
-        self.sigma = math.gamma(1 + self.beta) 
+        self.sigma = math.gamma(1 + self.beta)
         self.sigma *= math.sin(math.pi * self.beta / 2)
         # math.gamma((1 + self.beta) / 2) * self.beta * 2 ** ((self.beta - 1) / 2)
         denominator = math.gamma((1 + self.beta) / 2)
@@ -83,7 +83,6 @@ def set_params(self, new_params):
         Returns
         -------
         No value.
-        
         '''
         self.xi = new_params.xi
         self.alpha = new_params.alpha
@@ -100,7 +99,7 @@ def optimize(self):
         -------
         ndarray
             The coordinates of the global best particle at the end of
-            the optimization process. 
+            the optimization process.
         '''
         i = 0
         # Initialize stagnating iterations counter.
@@ -172,7 +171,7 @@ def __get_new_solutions(self):
 
     def __empty_nests(self):
         '''
-        Replaces a fraction of nests (determined by xi) by generating 
+        Replaces a fraction of nests (determined by xi) by generating
         new solutions.
 
         Parameters

diff --git a/models/single_objective/firefly_algorithm.py b/models/single_objective/firefly_algorithm.py
@@ -1,5 +1,4 @@
 import numpy as np
-import math
 from collections import namedtuple
 from models.swarm_algorithm import SwarmAlgorithm
 
@@ -25,15 +24,15 @@ class FireflyAlgorithm(SwarmAlgorithm):
     params : FireflyAlgorithmParams
         Model behavioral parameters.
     bounds : ndarray
-        A 2 by D matrix containing lower and upper bounds of the search space 
+        A 2 by D matrix containing lower and upper bounds of the search space
         for each dimension.
     seed : int, optional, default=None
         Random generator seed.
     max_iter : int, optional, default=100
         Maximum number of iterations (generations).
     stag_iter : int, optional, default=100
         Specifies the allowed number of iterations without solution improvement
-        by equal or more than a given tolerance. If the number is exceeded, 
+        by equal or more than a given tolerance. If the number is exceeded,
         the optimization process stagnations occurs and the algorithm stops.
     e : float, optional, default=1e-5
         Tolerance.
@@ -52,15 +51,15 @@ class FireflyAlgorithm(SwarmAlgorithm):
         Light absorption coefficient.
     lambd : float
         Randomization coefficient determining the weight of the third component
-        of the update rule, which coordinates the movement towards 
+        of the update rule, which coordinates the movement towards
         the global best solution.
     particles : ndarray
         An N by D array representing the swarm of N particles.
     scores : ndarray
         An array of size N representing the value of the fitness function
         for each particle.
     gbest : ndarray
-        A D-dimensional vector representing the position of the current 
+        A D-dimensional vector representing the position of the current
         global best particle.
     gbest_score : float
         The value of the fitness function for the current global best particle.
@@ -69,7 +68,7 @@ class FireflyAlgorithm(SwarmAlgorithm):
     '''
     def __init__(self, D, N, fit_func, params, bounds, seed=None, max_iter=100,
                  stag_iter=100, e=0.0001):
-        super().__init__(D, N, fit_func, params, bounds, seed, max_iter, 
+        super().__init__(D, N, fit_func, params, bounds, seed, max_iter,
                          stag_iter, e)
 
     def reset(self):
@@ -98,7 +97,6 @@ def set_params(self, new_params):
         Returns
         -------
         No value.
-        
         '''
         self.beta_0 = new_params.beta_0
         self.alpha_0 = new_params.alpha_0
@@ -118,7 +116,7 @@ def optimize(self):
         -------
         ndarray
             The coordinates of the global best particle at the end of
-            the optimization process. 
+            the optimization process.
         '''
         i = 0
         # Initialize stagnating iterations counter.
@@ -148,7 +146,7 @@ def optimize(self):
                 stag_count += 1
             elif stag_count > 0:
                 stag_count = 0
-            
+
             prev_best_score = self.gbest_score
         return self.gbest
 
@@ -163,7 +161,7 @@ def __move_first_to_second(self, f1, f2):
             A vector representing firefly to move.
         f2 : ndarray
             A vector representing firefly to move to.
-        
+
         Returns
         -------
         No value.
@@ -174,17 +172,17 @@ def __move_first_to_second(self, f1, f2):
         # Get the attractiveness of the firefly we want to fly to.
         # beta = self.beta_0 * math.exp(-self.gamma * r ** 2)
         beta = self.beta_0 / (1 + self.gamma * r ** 2)
-        
+
         diff = f2 - f1
         urand_1 = np.random.randn(self.D)
 
         # Update coordinates according to the flight formula.
         f1 += beta * diff + self.alpha * urand_1
 
         # Modifications:
-        # Add an extra term to the flight formula, which uses global best 
+        # Add an extra term to the flight formula, which uses global best
         # solution to improve the efficiency.
-        urand_2 = np.random.randn(self.D)        
+        urand_2 = np.random.randn(self.D)
         f1 += self.lambd * urand_2 * (self.gbest - f1)
 
     def __move_all(self):
@@ -211,7 +209,7 @@ def __move_all(self):
 
     def __reduce_alpha(self, iteration):
         '''
-        Reduces alfa as the number of iterations increases. 
+        Reduces alfa as the number of iterations increases.
         Optional modification.
 
         Parameters