source: releases/mystic-0.2a1-patch2.diff @ 248

mystic-0.2a1/mystic/tools.py

@@ -132 +132 @@
         self._y = []   
 
     def __call__(self, x, y):
+        from numpy import ndarray
+        if isinstance(x,ndarray): x = list(x)
         self._x.append(x)
         self._y.append(y)
      
@@ -153 +155 @@
     """
     import numpy
     def __init__(self, interval = 10, xinterval = numpy.inf):
-        Sow.__init__(self)
+       #Sow.__init__(self)
+        super(VerboseSow,self).__init__()
         self._step = 0
         self._yinterval = interval
         self._xinterval = xinterval
         return
     def __call__(self, x, y):
         from numpy import ndarray
-        Sow.__call__(self, x, y)
+       #Sow.__call__(self, x, y)
+        super(VerboseSow,self).__call__(x, y)
         if isinstance(y,(list,ndarray)):
             y = y[0] #XXX: get the "best" fit... which should be in y[0]
         if isinstance(x[0],(list,ndarray)): #XXX: x should always be iterable
@@ -169 +173 @@
            #print "Generation %d has best Chi-Squared: %s" % (self._step, y)
             print "Generation %d has best Chi-Squared: %f" % (self._step, y)
         if int(self._step % self._xinterval) == 0:
+            if isinstance(x,ndarray): x = list(x)
             print "Generation %d has best fit parameters:\n %s" % (self._step, x)
         self._step += 1
         return
@@ -182 +187 @@
     import numpy
     def __init__(self, interval=1, filename='log.txt', new=False):
         import datetime
-        Sow.__init__(self)
+       #Sow.__init__(self)
+        super(LoggingSow,self).__init__()
         self._filename = filename
         self._step = 0
         self._yinterval = interval
@@ -197 +203 @@
     def __call__(self, x, y):
         self._file = open(self._filename,'a')
         from numpy import ndarray
-        Sow.__call__(self, x, y)
+       #Sow.__call__(self, x, y)
+        super(LoggingSow,self).__call__(x, y)
         if isinstance(y,(list,ndarray)):
             y = y[0] #XXX: get the "best" fit... which should be in y[0]
         if isinstance(x[0],(list,ndarray)): #XXX: x should always be iterable
             x = x[0] #XXX: get the "best" fit... which should be in x[0]
         if int(self._step % self._yinterval) == 0:
+            if isinstance(x,ndarray): x = list(x)
             self._file.write(" %d   %f   %s\n" % (self._step, y, x))
         self._step += 1
         self._file.close()

mystic-0.2a1/mystic/differential_evolution.py

@@ -215 +215 @@
 Further Inputs:
 
     strategy -- the mutation strategy for generating new trial
-        solutions [default = Best1Exp]
+        solutions [default = Best1Bin]
     CrossProbability -- the probability of cross-parameter mutations
-        [default = 0.5]
+        [default = 0.9]
     ScalingFactor -- multiplier for the impact of mutations on the
-        trial solution [default = 0.7]
+        trial solution [default = 0.8]
     callback -- an optional user-supplied function to call after each
         iteration.  It is called as callback(xk), where xk is
         the current parameter vector.  [default = None]
@@ -227 +227 @@
 
         """
         #allow for inputs that don't conform to AbstractSolver interface
-        from mystic.strategy import Best1Exp
-        strategy=Best1Exp    #mutation strategy (see mystic.strategy)
-        CrossProbability=0.5 #potential for parameter cross-mutation
-        ScalingFactor=0.7    #multiplier for mutation impact
+        from mystic.strategy import Best1Bin
+        strategy=Best1Bin    #mutation strategy (see mystic.strategy)
+        CrossProbability=0.9 #potential for parameter cross-mutation
+        ScalingFactor=0.8    #multiplier for mutation impact
         callback=None        #user-supplied function, called after each step
         disp=0               #non-zero to print convergence messages
         if kwds.has_key('strategy'): strategy = kwds['strategy']
@@ -399 +399 @@
 Further Inputs:
 
     strategy -- the mutation strategy for generating new trial
-        solutions [default = Best1Exp]
+        solutions [default = Best1Bin]
     CrossProbability -- the probability of cross-parameter mutations
-        [default = 0.5]
+        [default = 0.9]
     ScalingFactor -- multiplier for the impact of mutations on the
-        trial solution [default = 0.7]
+        trial solution [default = 0.8]
     callback -- an optional user-supplied function to call after each
         iteration.  It is called as callback(xk), where xk is
         the current parameter vector.  [default = None]
@@ -415 +415 @@
     disp -- non-zero to print convergence messages.
         """
         #allow for inputs that don't conform to AbstractSolver interface
-        from mystic.strategy import Best1Exp
-        strategy=Best1Exp    #mutation strategy (see mystic.strategy)
-        CrossProbability=0.5 #potential for parameter cross-mutation
-        ScalingFactor=0.7    #multiplier for mutation impact
+        from mystic.strategy import Best1Bin
+        strategy=Best1Bin    #mutation strategy (see mystic.strategy)
+        CrossProbability=0.9 #potential for parameter cross-mutation
+        ScalingFactor=0.8    #multiplier for mutation impact
         callback=None        #user-supplied function, called after each step
         disp=0               #non-zero to print convergence messages
         if kwds.has_key('strategy'): strategy = kwds['strategy']

mystic-0.2a1/examples/test_mogi.py

@@ -114 +114 @@
     #
     desol, dstepmon = de_solve()
     print "desol: ", desol
-    print "dstepmon 10: ", dstepmon.x[50]
-    print "dstepmon 20: ", dstepmon.x[100]
+    print "dstepmon 50: ", dstepmon.x[50]
+    print "dstepmon 100: ", dstepmon.x[100]
     #
     # this will try to use nelder_mean from a relatively "near by" point (very sensitive)
     point = [1234., -500., 10., 0.001] # both cg and nm does fine