GWO PSO

GWO.py

@@ -56,10 +56,11 @@ def GWO(objf,lb,ub,dim,SearchAgents_no,Max_iter, Positions = None, fopt=float("-
 
 
             # Calculate objective function for each search agent
-            scores = objf(Positions[i,:])
+            #score = objf(Positions[i,:])
             fitness = objf(Positions[i, :])
-            #print Positions
-            #print scores
+            #score =   abs(fitness - fopt)
+            #print fitness, Positions[i, :]
+            #print score, fitness
 
             # Update Alpha, Beta, and Delta
             if fitness<Alpha_score :

PSO.py

@@ -8,7 +8,7 @@
 import random
 import numpy
 import math
-from colorama import Fore, Back, Style
+#from colorama import Fore, Back, Style
 from solution import solution
 import time
 
@@ -17,7 +17,7 @@
 
 
 
-def PSO(objf,lb,ub,dim,PopSize,iters):
+def PSO(objf,lb,ub,dim,PopSize,iters, pos = None, fopt=float("-inf") , **kwargs ):
 
     # PSO parameters
 
@@ -48,7 +48,7 @@ def PSO(objf,lb,ub,dim,PopSize,iters):
 
     gBestScore=float("inf")
 
-    pos=numpy.random.uniform(0,1,(PopSize,dim)) *(ub-lb)+lb
+    #pos=numpy.random.uniform(0,1,(PopSize,dim)) *(ub-lb)+lb
 
     convergence_curve=numpy.zeros(iters)
 
@@ -92,14 +92,16 @@ def PSO(objf,lb,ub,dim,PopSize,iters):
 
         convergence_curve[l]=gBestScore
 
-        if (l%1==0):
+        if (l%1==0) and 'verbose' in kwargs:
                print(['At iteration '+ str(l+1)+ ' the best fitness is '+ str(gBestScore)]);
-    timerEnd=time.time()  
+    timerEnd=time.time()
+    s.best=gBestScore
     s.endTime=time.strftime("%Y-%m-%d-%H-%M-%S")
     s.executionTime=timerEnd-timerStart
-    s.convergence=convergence_curve
+    s.convergence=list(convergence_curve)
     s.optimizer="PSO"
     s.objfname=objf.__name__
+    s.pop = list(pos)
 
     return s